Akira Toh [1], Kristy B. Arbogast, Ph.D [2], Matthew R. Maltese [3]
[1] Department of Bioengineering, Unviersity of Pennsylvania, Philadelphia, Pennsylvania
[2] Center for Injury Research and Prevention, Childrens Hospital of Philadelphia, Philadelphia, Pennsylvania
[3] Department of Anethesiology and Critical Care Medicine, Childrens Hospital of Philadelphia, Philadelphia, Pennsylvania
A biofidelic anthropomorphic test device (ATD) is essential for developing crash safety systems for occupants. The ubiquitous Kroell blunt hub impacts to the thoraces of post-mortem human subject (PMHS) have formed the biofidelity requirements for the adult-sized ATD thorax. Recently collected thoracic force-deflection data from cardio-pulmonary resuscitation patients offer a large dataset of biomechanical data across a broad age range. However, the applicability of this CPR data to inform ATD biofidelity requirements is unknown. Thus, the objective of this study was to evaluate the performance of CPR-derived thoracic stiffness and damping properties in a mathematical model of blunt thoracic impact validated to the Kroell experimental data.
A previously validated (Neathery 1971) spring-mass-damper (SMD) model of the adult chest during thoracic impact was recreated in MatLab. The model was modified (SMD-CPR), with mass and flesh stiffness from SMD, and thorax stiffness and damping from CPR. Comparison of the two models revealed that the damping constant from CPR was too low to produce a model that could be validated to the Kroell experiments. Thus, we created a third model (SMD-CPR-K), by changing the damping of SMD-CPR model back to the value from SMD model. During blunt hub impact simulations, the SMD-CPR-K model yielded a 47.0% increase in chest deflection from the SMD model, over impact velocities of 7.1 m/s.
The SMD-CPR-K model was then parameterized to quantify the mass, stiffness and damping effects on the force-deflection response. Increases to sternum mass or flesh stiffness led to increased force only in the first 40 mm of deflection. Increased thorax stiffness led to increased maximum thorax deflection only. Finally, increased thorax damping led to increased force in the first 40 mm of deflection and decreased maximum deflection.
These results indicate that stiffness but not damping characteristics from CPR are compatible with the impact environment.
Antony Francis, M.A.Sc [1], Ron El-Hawary, MD FRCS [1], J. Michael Lee, PhD FBSE [1]
[1] School of Biomedical Engineering, Dalhousie University, Halifax, Nova Scotia
[2] Department of Surgery, Dalhousie University, Halifax, Nova Scotia
[3] IWK Health Center, Halifax, Nova Scotia
Introduction: The goal of this project is to develop a standardized radiostereometric analysis (RSA) bead placement protocol for studying thoracic spinal fusion in adolescent idiopathic scoliosis (AIS) patients.
Methods: The bead placement protocol is performed via a computer simulation, which aims to electronically mimic an RSA exam following thoracic spinal fusion. The simulation is being done in three main steps: 1) Solid Edge is used to develop a CAD model to simulate the spine with the AIS instrumentation and to “insert” beads into the vertebrae of interest as per the proposed placement protocol; 2) POV-Ray software is used to simulate the left and right x-ray images obtained in a real RSA exam; 3) Model-Based RSA software is used to evaluate the developed bead placement protocol and to detect motion between the marked vertebral segments. To obtain an objective view of the performance of the developed RSA system, the accuracy and precision of the system are validated. Accuracy, defined as the closeness of agreement between a test result and a reference “true” value, is determined via comparing the migration generated by the Model-Based RSA software with the corresponding “true” manipulation performed in Solid Edge. Precision, defined as the closeness of agreement between two independent test results, is determined via repeated measures of zero displacement.
Results: The average translational accuracy of the developed system is 0.21 mm, 0.1 mm and 0.55 mm along the x, y and z-axes, respectively. The average translational precision of the system is 0.012 mm, 0.0021 mm and 0.067 mm along the x, y and z-axes, respectively.
Conclusion: Based on complete computer simulation, the developed RSA system allows measurement of the success of thoracic spinal fusion with high accuracy and precision. The computer simulation is a powerful tool than can be used to facilitate the development of an RSA system, when (1) complex anatomy is involved, (2) RSA markers can be obscured via the shadowing of adjacent instrumentation, and (3) the surface area in which the RSA markers are inserted is small.
Nan Zhang, B.S. [1], Kim L. Wagoner, B.S. [1], Rebecca A. Bader, Ph.D [1]
[1] Department of biomedical and chemical engineering,syracuse univeristy, syracuse, New York
Introduction: Despite over expression of the hyaluronic acid/CD44 receptor by inflamed synovial tissue, the use of hyaluronic acid (HA) as a targeted drug delivery vehicle in the treatment of rheumatoid arthritis has been relatively unexplored. The nature of the disease necessitates systemic delivery; however, native hyaluronic acid is too large to safely be administered systemically, and the half life of HA is short due to rapid clearance by the HARE receptor of liver endothelial cells. As a first step in generating a systemically administrable HA based drug delivery vehicle, we degraded HA to a molecular weight that will permit passive targeting through the highly permeable vasculature of the inflamed synovial tissue. The resultant low molecular weight HA was then conjugated to methotrexate (MTX) through an adipic dihydrazide linker using known methodology. The goal of the current study was to demonstrate that MTX conjugated to HA (HA-MTX) is more effective at inducing an anti-inflammatory response in rheumatoid arthritis synovial fibroblasts (RASFs) than MTX alone or MTX conjugated to a synthetic polymer that cannot bind to the CD44 receptor.
Method: Adipic dihydrazide modified HA was obtained following known methods. In an analogous manner, adipic dihydrazide was coupled to the EDCI activated glutamic acid group of MTX to yield HA-MTX. The bioconjugates were fully characterized by 1H NMR and GPC coupled with a static and dynamic light scattering detector. GPC-light scattering yielded a molecular weight (58 kDa) and hydrodynamic radius (48 nm) that were both above the threshold necessary for passive targeting. MTX and HA-MTX were tested in vitro by administration of doses of 0.1 mg/ml to a TNF-a stimulated RASF cell line. For comparison, poly(ethylene glycol) conjugated MTX (PEG-MTX), which will not bind to the CD44 receptor was also evaluated. 24 hours after administration, the supernatant was removed (N=3 wells for each group), and the concentrations of the IL-6, IL8 and VEGF that have been implicated in rheumatoid arthritis pathogenesis were determined using multiplex immunoassay.
Results: MTX alone elicited no response, while PEG-MTX was pro-inflammatory in regards to the secretion of IL-6 and IL-8. In contrast, HA-MTX significantly reduced the secretion of VEGF and IL-8. IL-6 concentrations were also decreased for cells treated with HA-MTX relative to the other treatment groups, but the difference was not significant. The levels of VEGF were so low as to be undetectable after administration of HA-MTX.
Conclusion: The results obtained thus far suggest that HA can be used as an effective tool in improving the effectiveness of existing therapeutics in the treatment of rheumatoid arthritis by targeting the CD44 receptor.
Theodorus E. de Groot, B.S. [1], Scott S. Verbridge, Ph.D [1], Claudia Fischbach, Ph.D [1]
[1] Department Biomedical Engineering, Cornell University, Ithaca, NY
Three-dimensional cell culture models are a more accurate means to study biological phenomena than traditional 2-D culture. Cells in 3-D culture can be presented with metabolite and growth factor gradients, cell-cell interactions, and cell-matrix interactions similar to that experienced in vivo. As tumors exist and grow in vivo they are subject to differing concentrations of oxygen, with a normoxic region near vasculature and a hypoxic region towards the core. These differences in oxygen concentration elicit specific responses from the cells in the tumor. Applying tissue engineering techniques to create 3-D cultures allows precise manipulation of the cells’ microenvironment and creates a means for decoupling specific cues in terms of their influence on tumor angiogenesis.
We designed and worked with a 3D tumor model to study the response of multiple tumor cell lines (MDA-MB231, OSCC-3) to changes in oxygen concentration in a 3D culture context. Cells were seeded in an alginate scaffold, an inert material that provides no integrin or other engagement sites for cellular interaction, and grown at varying oxygen levels between hypoxic, normoxic, and ambient oxygen conditions. This effectively allowed for the study of the direct effect of oxygen levels on the cells while removing interference from other microenvironmental cues. To determine the effects of hypoxia we quantified cell secretion of two pro-angiogenic factors, vascular endothelial growth factor (VEGF) and interleukin 8 (IL-8) using ELISA. VEGF is a growth factor which promotes vascularization of tumor masses and its expression has been shown to be inversely proportional to oxygen concentration for a number of cell types. IL-8 is a chemokine which attracts endothelial and inflammatory cells which are both involved in tumor progression. Using western blot, we measured deposition of fibronectin, an extracellular matrix molecule, in the scaffold.
Overall, our results indicated that oxygen concentration had a significant effect on cell behavior. We found VEGF secretion in the scaffold to be inversely proportional to oxygen concentration and secretion at a substantial level at normoxic conditions while IL-8 secretion increased with oxygen. In contrast, 2-D experiments have shown low levels of IL-8 and VEGF secretion at high and low oxygen concentration. Fibronectin was detected in the scaffold, indicating the possibility of cell-matrix interactions which could impact cell behavior. We will culture cells in a bioactive RGD-modified alginate scaffold to test significance of the interaction. The use of 3-D scaffolds and physiological conditions in studying cells has been shown to be powerful tool for making biologically relevant predictions in tumor models.
Chelsey J. Schweikert [1], Kali R. Cole, B.S. [1], Julie M. Hasenwinkel, Ph.D. [1], Rebecca A. Bader, Ph.D. [1]
[1] Department of Biomedical & Chemical Engineering, Syracuse University, Syracuse, Ny
Introduction: A relatively unexplored, but potentially ideal, material for the prevention of premature clearance in the realm of drug delivery is the polysaccharide poly(sialic acid), for which the body possesses no known receptors. Use of poly(sialic acid) as a carrier system should, therefore, increase the circulatory stability of a therapeutic agent by reducing undesired uptake by the reticuloendothelial system. The goal of the current study was to employ poly(sialic acid) in the development of a novel nanoparticle carrier system for drug delivery.
Materials & Methods: Ionic gelation of chitosan with sodium tripolyphosphate (TPP) in the presence of poly(sialic acid) (PSA) was used to prepare nanoparticles. Following a procedure originally used for the preparation of hyaluronic acid-chitosan nanoparticles, chitosan was dissolved at a weight percentage of 0.25% in 0.4% aqueous acetic acid. A solution of PSA and TPP together in DI water at concentrations of 0.25% and 0.1% respectively was also prepared. 1.5 ml of the PSA-TPP solution was added slowly to 3.0 ml of the chitosan solution, and the mixture was magnetically stirred for 30 minutes. For TEM sample preparation, a drop of nanoparticle solution was placed on a mesh copper TEM grid with a carbon film and air dried. For AFM sample preparation, a drop of nanoparticles solution will be placed on a glass slide and air dried. Particle size was evaluated using dynamic light scattering (photon correlation spectroscopy) with a Malvern Zetasizer Nano. Methotrexate and bovine serum albumin were chosen as model drugs for incorporation into the nanoparticle system.
Results & Discussion: Dynamic light scattering yielded an average particle size of 708 nm +/- 49 nm. Both TEM and AFM confirmed the formation of small, spherical nanoparticles; however, TEM indicated that the nanoparticles aggregated into chains and irregular shapes.
Conclusion: Polysialic acid was successfully incorporated into a nanoparticle carrier system by ionic complexation with chitosan. The particle size was consistent with those obtained for other polysaccharide-based nanoparticles.
Ming-Jhi Tzeng [1], YI-Hui Wu [2], Y.Y. Huang [1], Jer-Junn Luh [3]
[1] Department of Biomedical Engineering, National Taiwan University, Taiwan
[2] Department of Electrical Engineering, National Taiwan University
[3] School and Graduate Institute of Physical Therapy, National Taiwan University College of Medicine
Ultrasound (US) has successfully been used for enhancing transdermal transport of a variety of molecules during in vitro and in vivo animal studies. However, owing to the difficulty in measuring and quantifying the amount of penetrated drug and response, not many in vivo data of sonophoresis on human subjects were available. In this study, we used a local anesthetic, Eutectic Mixture of Local Anesthetics (EMLA), as a model drug and assessed its nerve conduction velocity (NCV) in human subjects to evaluate the efficiency and response of ultrasound-enhanced transdermal delivery. Generally speaking, effects of an anesthetics were measured by pain response; nevertheless, pain measurement always varies with subjective experience. In our experiments, a short brief electrical shock was first applied to the peripheral nerve and the nerve conduction velocity and latency change after transdermal delivery of EMLA were then measured. This method could offer an objective measurement or quantification of pain. Experimental results showed that the nerve conduction velocity was dramatically decreased after ultrasound-enhanced delivery of EMLA transdermally. Ultrasound-facilitated delivery reduced significantly the onset time of anesthetic drugs through skin. This study also demonstrates that assaying changes in nerve conduction velocity is a possible tool for measuring pain response without depending on subject self-assessment.
Ravi Sinha [1], Russell Gould [1], Daniel Judge [2], Jonathan Butcher [1]
[1] Department of Biomedical Engineering, Cornell University, Ithaca, NY
[2] Division of Cardiology, Johns Hopkins University School of Medicine, Baltimore, MD
Biomechanics are an important functional readout for soft tissues. Despite advances in designing tools to measure biomechanics, it is still very hard to measure the biomechanics of small tissues like heart valves from the most widely used animal model that is mouse. This is due to the inability of conventional force transducers to be attached to such a small tissue. This limits the vast functional understanding that can be obtained from studying the biomechanics of genetically altered mice. We have addressed this issue with a new device which uses cylindrical polydimethylsiloxane (PDMS) cantilever posts as force transducers (force proportional to post deformation). The posts were calibrated by deforming them against a rigid mount on a weighing balance and measuring the generated force as increase in weight.
We validated the approach by studying the biomechanical remodeling of murine heart valves in a genetic model of Marfan syndrome. Marfan syndrome in humans is characterized by long limbs, dislocated lenses and most significantly more compliant mitral valve and aortic root which are not able to bear too much load. It is associated with mutations in the FBN1 gene that encodes for fibrillin-1. We tested rectangular samples from the mitral valve and the aortic root of fibrillin-1 knockout mice and control wild type mice. The mitral valves were tested under a confocal microscope imaging in real time. We found that the mitral valves as well as aortic roots from the knockout mice were significantly weaker (lower modulus) than those from wild type control mice of the same age. The moduli (in kPa) +/- standard error (n=6) found were (aortic root, mitral valve) = (179 +/- 4, 449 +/- 106) for marfan and (293 +/- 35, 1125 +/- 359) for wild type.
The confocal images of valves showed collagen fibers stretch and align along the stretch direction with increasing strain (alignment was quantified from a Fourier transform of images). It was also observed that valve cells essentially retain their shape up to a certain stretch but start stretching after that (cell deformation was quantified using a dimensionless circularity index). The cells started deforming around stretch ratio 1.2 for wild type and at a higher stretch ratio of about 1.5 for marfan valves. These results agree with histological evidence that marfan valves have a higher proteoglycan content making them more compliant and allowing cells to rearrange in the ECM till a higher stretch ratio. Thus our system was validated. The ability to make these observations is important in many ways, for e.g. the knowledge of cell stretching behavior is important for tissue engineering while the ability to read the sensitivity of cells to stress/strain can provide important insights into disease mechanisms.
Divya Raman [1], Anitha Priya Krishnan, MS [1], Walter O'Dell, PhD [2]
[1] Dept. of Biomedical Engineering, Univ. of Rochester
[2] Dept. of Radiation Oncology, Univ. of Rochester
[3] Dept. of Biophysics and Biochemistry, Univ. of Rochester
[4] Dept. of Imaging Sciences, Univ. of Rochester
[5] Dept. of Neurobiology and Anatomy, Univ. of Rochester
Paths of elevated water diffusion provides a preferred route of migration for cancer cells along the white matter tracts from primary tumor. Using MR-DTI, we have recently developed a computational model of cell migration to predict the location of human brain tumor recurrence following radiation treatment. In an attempt to validate this model, we proposed to extend this to a rat model and verify the results with histology images of rat brain injected with tumor cells. In vivo and ex vivo T2 MR rat brain images were obtained at comparable resolution. Fiber tractography was performed on ex vivo MR-DTI to identify major fiber bundles. Brain masks and Tumor masks were obtained. The constrained random walk model was run with the tumor seed point located near a major fiber bundle. The cells were allowed to walk for 500 steps with a step size of ¼th voxel size. The distribution obtained from the model showed that most cells remained closer to the major fiber bundle compared to the surrounding gray matter. The fixed rat brains were then sliced and stained with H&E stain and Silver Stain to highlight axonal fibers to ease registration with DTI images. Future studies includes injecting GFP labeled CNS1 native rat glioma cells at same seed points and comparing the histology images with results obtained from our migration model.
Luke J. Mortensen, M.S. [1], Supriya Ravichandran, B.S. [1], Renea Faulknor [1], Lisa A. DeLouise, Ph.D [1]
[1] Department of Biomedical Engineering, University of Rochester, Rochester, NY
[2] Department of Dermatology, University of Rochester, Rochester, NY
The growing presence of quantum dots (QD) in a variety of biological, medical, and electronics applications means an increased risk of human exposure in manufacturing, research, and consumer use. Preliminary studies have shown the ability of QD to penetrate the skin in an increased fashion under barrier compromised conditions (such as UVR or diseased states). The cytotoxic effect of QD on keratinocytes (the primary skin cell type) and systemic retention of the most commonly used QD once in the bloodstream have also been demonstrated. Despite work in the area, no published studies have correlated the barrier effect of UVR damage in vivo with an increase in nanoparticle skin penetration or reported the effect of UVR cellular damage on the cytotoxicity and uptake of QD in primary keratinocytes. To address these issues, we use transepidermal water loss (TEWL) to quantify in vivo barrier damage at a range of UVR doses and relate this to QD skin penetration. We then examine cellular impact of UVR and QD in vitro with fluorescence-activated cell sorting (FACS) to determine viability, uptake levels, and reactive oxygen species generation. Preliminary results suggest that quantifiably induced barrier defect can be associated with increased in vivo QD skin penetration and that stress and differentiation states of primary skin cells are important drivers in their risk for QD uptake. Our research suggests future work in specific targeting of nanoparticles, to prevent or enhance uptake. This knowledge will be used to develop powerful therapeutic agents, decreased toxicity cosmetic nanoparticles, and powerful skin imaging modalities.
Nate T. Greene [1], Oleg Lomakin [1], Kevin A. Davis [1]
[1] Department of Biomedical Engineering, University of Rochester, Rochester, NY
Most studies of single-unit responses in the lateral superior olive (LSO) have been conducted in anesthetized preparations; two have sampled responses in unanesthetized, decerebrate animals. In anesthetized cats, LSO units usually respond to ipsilateral best-frequency tone bursts with a chopper-type discharge pattern, characterized by regularly spaced peaks of activity initially time-locked to the stimulus onset. In contrast, LSO units in decerebrate cats often display less regular, primary-like discharge patterns. Barbiturates are known to alter the balance of excitation and inhibition, and thus may explain these differences.
The goal of the present modeling study was to investigate the role of inhibition in shaping the temporal discharge patterns of LSO units. Consistent with the known anatomy, model LSO units receive excitatory inputs on their distal dendrites from model spherical bushy cells in the ipsilateral cochlear nucleus, which in turn receive inputs from model auditory nerve fibers. LSO cells also receive inhibitory inputs on their soma from globular bushy cells in the contralateral cochlear nucleus via a synapse in the (not explicitly modeled) ipsilateral medial nucleus of the trapezoid body. In the current set of simulations, the input was strictly monaural, thus the contralateral inhibitory inputs simply provided tonic (spontaneous) inhibitory drive. Simulation results show that model LSO units display chopper-type discharge patterns when the inhibitory inputs are weak, which become primary-like when the inhibitory synaptic strength is increased. These results thus suggest that inhibition plays a substantially larger role in shaping response properties in decerebrate than anesthetized preparations. Supported by NIDCD grant R01 DC 05161.
Andrew Law, M.S. [1], Adam G. Davidson, Ph.D. [1], Gil Rivlis, Ph.D. [1], Marc H. Schieber, M.D./Ph.D. [1]
[1] Department of Biomedical Engineering, University of Rochester, Rochester, New York
[2] Department of Neurobiology & Anatomy, University of Rochester Medical Center, Rochester, New York
[3] Department of Neurology, University of Rochester Medical Center, Rochester, New York
Conventional neural prosthesis control requires a sophisticated algorithm to decode recorded neuronal activity into control signals that drive the prosthetic device. An alternative approach, which does not require neural decoding, is to assign specific prosthesis-encoding responsibilities to recorded neurons. With this approach, the utility of the prosthetic device depends on the user's ability to voluntarily modulate and/or control the firing rates of recorded neurons effectively. Previous studies have shown that, when given direct visual feedback of neuronal activity, human and non-human primates can learn to voluntarily modulate the firing rates of single primary motor cortex (M1) neurons to control brain-computer interfaces (BCIs). We investigate whether non-human primates can voluntarily modulate two-neuron M1 ensemble activity for BCI control.
We recorded neuronal activity from M1 in a male rhesus macaque using four, sixteen-channel microelectrode arrays. In separate experimental sessions, two well-isolated single units were selected to form two-neuron ensembles jointly encoding the vertical position of a cursor on a computer screen. The cursor's position was defined by the average normalized firing rates of the two selected neurons. To earn rewards, the monkey was required to move the cursor into a center home-target for 50-ms and move the cursor to a subsequently presented high- or low-rate target and maintain the cursor in the target for a requisite hold duration. The difficulty of the cursor-control task depended on the size of the targets and the duration of the high/low-rate target hold.
Within-session and across-session improvement in cursor-control task performance indicates that non-human primates can indeed voluntarily modulate two-neuron ensemble activity for BCI control. Moreover, the monkey's level of performance in the cursor-control task is higher when two neurons jointly encode cursor position than when only one neuron encodes cursor position, especially as task difficulty increases. By comparing the mutual information between ensemble neurons' joint firing rates and cursor position with the mutual information between ensemble neurons' individual firing rates and cursor position, we found that for most ensembles, ensemble neurons synergistically encode cursor position. Future work aims to investigate whether non-human primates can voluntarily modulate larger M1 ensembles for BCI control and to evaluate how ensemble size affects BCI task performance.
Kevin Heeyong Kang, M.Eng. [1], Laura Ann Hockaday, Ph.D [1], Nicholas W. Colangelo, B.S. [1], C.C. Chu [1], Jonathan T. Butcher [1]
[1] Deparment of Biomedical Engineering, Cornell University, Ithaca, New York
[2] Department of Human Ecology, Cornell University, Ithaca, New York
The aortic valve is a complex heterogeneous structure designed to ensure unidirectional blood flow and to provide blood to the heart through the coronary ostia. Ongoing extracellular matrix (ECM) remodeling and valve repair are mediated by dynamic populations of aortic valvular interstitial cells (VICs) in leaflets and smooth muscle cells in the root. Heterogeneous cell and ECM distribution in the valve are thus important for appropriate mechanical function of the relatively stiff root and flexible leaflets.
Current tissue engineering strategies do not adequately replicate internal tissue heterogeneities, where the challenge in controlling tissue geometry still remains to be addressed. Solid freeform fabrication (3D printing) is a powerful technique that allows production of geometrically complex constructs with various materials. For this study, we have developed an approach that combines MicroCT image processing, custom algorithms, and 3D printing to generate cell-seeded valve constructs incorporating anatomic heterogeneities and complex geometry.
The open source Fab@Home 3D printing system allows layer-by-layer deposition of extrudable materials. We have adapted several hydrogel blends consisting of poly(ethylene glycol)-diacrylate (PEG-DA), cell resorbable poly-(ester amide) (PEA), fibronectin, alginate, and photoinitiator (Irgacure). These gels were tuned to exhibit a broad range of mechanical properties in strain-to-failure tests (stress_UTS=[5,250]kPa, max_strain=[0.24,2.1]). These hydrogels are designed to photo-polymerize under ultraviolet radiation(365nm). Our lab has thus additionally developed UV LED cross-linking system that integrates with the printer for simultaneous printing and photo-crosslinking hydrogels.
To fabricate valves using hydrogels, MicroCT scans of porcine aortic valve in DICOM format were rendered via thresholding and region growing to highlight density differences between the root and the leaflets. Rendered images were then converted into printable interlocking stereolithography files (STL) of the leaflets and root. Stented aortic valve and full valve constructs were printed in less than 45 minutes in different hydrogel mixtures. To evaluate printing accuracy, MicroCT scan slices (70 keV) of the printed valve were compared to the print image files. To demonstrate the feasibility of printing tissue that incorporates anatomic heterogeneity, custom algorithms were developed specifically for heterogeneous printing: one specifies extrusion paths converted from pixilated image files and the other defines the path by a gradient or probability function.
In summary, this study utilizes photo-crosslinking hydrogels to fabricate cell-seeded valves of nonself-supporting complex heterogeneous geometries on a physiologically relevant scale. Further studies of cell behavior in fabricated valves are required, but we believe our approach show potential for cardiac valve regeneration.
Lowell M. Smoger, M.S. [1], Elizabeth A. DeBartolo, Ph.D [1]
[1] Department of Mechanical Engineering, Rochester Institute of Technology, Rochester, New York
The study of how aerosols flow through the airways and within the alveoli is of interest to many research areas such as particle deposition during smoking or inhaled medication dispensing. Current models of the lung do not include accurate mechanical properties. Even though individual alveoli are too small to reasonably examine on a true-scale level and tissue-level mechanical properties are not easily analyzed, bulk pressure-volume (PV) behavior and elasticity of the lung are well defined. However, these properties have not been correlated to typical engineering material properties. Current models use linear elastic behavior to represent the lung walls. The biaxial response of non-linear hyperelastic materials is required to accurately represent the lung mechanical response.
Research into current biaxial specimen designs and test methods was conducted to improve the current design and method. Developing an improved biaxial test fixture system that accurately predicts boiling flask PV data is necessary in order to easily test candidate lung model materials. This will be achieved by a machine design and manufacturing analysis. This analysis will quantify any misalignment in the biaxial test fixture so that identical numerical simulations can be conducted. The goal is to better understand the sensitivity of the biaxial tension test to deviations from equibiaxial, perpendicular displacement loading conditions.
Additionally, a stress decay factor (SDF) method is proposed to accurately predict equibiaxial internal stresses from measurable data. This SDF method will be applied to the popular cruciform specimen geometry that has been numerically optimized with respect to a newly proposed set of requirements.
To date, a minimum constrain design analysis of the test fixture has suggested that the number of constraints on the system may not be optimal. Additionally, an optimal planar specimen geometry for the selected biaxial test fixture has been developed through numerical simulation. Analysis of test fixture misalignment and resulting specimen responses are in progress.
As a means of verification, a bulb geometry will be employed as a final step to test the planar biaxial properties from an improved biaxial test fixture.
David J. Monahan, B.S [1], James K. Stern, B.S [1], Jahanavi S. Gauthaman, B.S [2], Cory B. Laudenslager, B.S [2], Brian D. Glod, B.S; Assis E. Ngolo, B.S [3]
[1] Department of Mechanical Engineering, Rochester Institute of Technology, Rochester, NY
[2] Department of Electrical Engineering, Rochester Institute of Technology, Rochester, NY
[3] Department of Computer Engineering, Rochester Institute of Technology, Rochester, NY
The Motion Tracking System (RIT Multidisciplinary Senior Design P10010) is a currently active project focusing on research and testing of various sensors capable of measuring patients' range of motion in their natural environments. The primary ranges of motion of interest are the motion of a human limb, where a limb is defined as a 3-bar linkage; and the motion of a human's lower back, where lower back is defined as the lumbar region, with 3 points of contact: sacrum, L1-L2, L3-L5. The various aspects of a motion tracking system: sensors, a portable micro-controller, interface circuitry, software, and human interfaces are explored. Customer and functional needs were identified, target specifications were established, and risks assessed.
Sensors with 1-, 3-, and 6- degrees of freedom, accelerometers, gyroscopes, inertial measurement units, and flex sensors were explored. The sensors are currently being tested for individual functionality and usability in a system as a whole. Test fixtures were designed and are being built for testing the sensors' accuracy. The test fixtures from this project leave prospects for further future testing. Another RIT research team, (P10011- Motion Tracking Human Interface), is working closely with this Motion Tracking System project to design sensor enclosures and attachment methods that can be easily sanitized, and are comfortable to wear throughout an individuals' daily activities.
The key project deliverable is to present future research teams with sufficient tools to create a portable motion tracking device for the Nazareth College Physical Therapy Clinic. The project aims to enhance the knowledge base of the RIT Biomedical Systems and Technologies Track regarding sensor usage in human motion tracking. Future market opportunities lie in research and development in university research, physical therapy clinics, athletics departments, military, entertainment, bio-robotics, and many other potential fields where motion tracking data can be implemented.
Vivek J. Khandwala, M.S. [1], Michelle A. Burack, M.D./Ph.D. [2], Jonathan W. Mink, M.D./Ph.D [2], Greg T. Gdowski, Ph.D [1], Martha J. Gdowski, Ph.D. [3]
[1] Department of Biomedical Engineering, University of Rochester, Rochester , NY
[2] Department of Neurology, University of Rochester, Rochester , NY
[3] Department of Neurobiology and Anatomy, University of Rochester, Rochester , NY
The therapeutic benefits of subthalamic nucleus (STN) deep brain stimulation (DBS) for motor symptoms of Parkinson’s disease (PD) are well-documented. However, the mechanisms and quality of motor improvements remain poorly characterized. We quantified the effects of PD and the treatment efficacy of DBS based on the performance of an unconstrained upper limb button press task requiring 3-dimensional (3-D) and planar motion. Movements of each limb were quantified using a 3-D motion analysis system. Data were recorded from two subjects with idiopathic PD who were chronically implanted with bilateral STN-DBS and 9 age-matched controls. PD subjects were studied off-medication in both DBS-off and DBS-on conditions. We tested the hypothesis that STN-DBS improves voluntary upper limb movement by altering the velocity and temporal sequencing of proximal joints (shoulder, elbow). We predicted that these changes would result in increased movement efficiency by minimizing total distance traveled. With stimulation, reaction and movement times decreased, while peak and average velocities (linear and angular) increased. Furthermore, temporal sequencing of the onset of changes in joint angles differed significantly during DBS-on versus DBS-off conditions. During planar movements, the frequency with which the PD subjects selected a preferred sequence of shoulder and elbow joint angle changes increased in the DBS-on condition. During stimulation, a preferred sequence was selected twice as frequently as in the DBS-off condition, and with increased frequency relative to the 9 age-matched controls. Neither kinematic changes nor increased frequency of sequence selection resulted in improved path ratios. These results suggest that STN-DBS improves movement execution at the cost of flexibility in movement strategy.
Ryan M. Burke, M.S. [1], Martha L. Zettel, M.S. [2], Khawarl Liverpool [1], Kelley S. Madden, Ph.D. [1], Edward B. Brown III, Ph.D. [1]
[1] Dept of Biomedical Engineering, University of Rochester, Rochester, NY
[2] Department of Neurobiology and Anatomy, University of Rochester Medical Center, Rochester, NY
The onset of metastasis represents an onerous milestone in the progression of breast cancer, as metastasis to distant organs is frequently the cause of mortality in breast cancer patients. One indicator of breast cancer patient prognosis observed in the clinic is the infiltration of tumor-associated macrophages (TAMs) the presence of these leukocytes in breast tumors is correlated with poor outcome due to increased metastatic burden.
Metastasis requires tumor cells to enter the blood or lymphatic circulation, or exit the tumor-host interface, and such movement has been witnessed by multiple laboratories along collagen type I fibers which emit strong second harmonic generation (SHG) signal. SHG is a nonlinear optical phenomenon characteristic of materials that are highly ordered on a molecular level. Tumor cells move along SHG-emitting fibers faster than when they are moving independently of them, and they follow those fibers through the tumor-host interface and towards blood vessels. Due to the known capacity of TAMs to induce local fibroblasts to remodel collagen, and the ability of macrophages to alter collagen ordering as quantified by SHG during breast gland development, it is here hypothesized that TAMs are the cells ultimately responsible for creating and maintaining the network of metastasis-promoting, SHG-producing fibers in the breast tumor. The process of establishing and maintaining these fibers is of therapeutic interest, as disruption of this ordering regulation capacity exhibited by TAMs may lead to decreased metastatic burden in cases of breast cancer.
The tumor draining lymph node (TDLN) is one of the earliest sites for establishment and subsequent propagation of metastases. Significant changes in SHG signatures between TDLNs in nonmetastatic and early metastatic cases have been seen in our laboratory, implicating collagen ordering in the TDLN as measured by SHG as another possible mechanism for influencing metastatic outcome. Hence we hypothesize that macrophages are the cells ultimately responsible for creating and maintaining the network of metastasis-promoting, SHG-producing fibers in the TDLN as well.
The correlation between TAM presence and poor prognosis, interestingly, is not evidenced in all tumor types for instance, colorectal tumor prognoses are improved by TAM infiltration. We believe that the ability of TAMs to maintain collagen ordering may be hindered in the colon, and that the natural immune function of this cell type then dominates. As fibroblasts from different tissues have been shown to retain a positional memory, acting in a manner specific to their tissue of origin even after transplant into new tissue, this clinical observation is possibly explained by fibroblasts in the colon not responding in the same manner to received TAM signals as fibroblasts in the breast. We therefore hypothesize that it is the tumor bed, not the tumor type, that dictates the role of TAMs in maintaining the network of metastasis-promoting, SHG-producing fibers in a given tumor.
Brendan J. Morse [1], Amy L. Lerner [1]
[1] Department of Biomedical Engineering, University of Rochester, Rochester, NY
Introduction: The main role of the meniscus is to provide stability to the knee and distribute load across the tibial plateau during weight bearing activities. Experimental and computational studies have shown that the load transmitted through the meniscus can vary between subjects due to differences in knee kinematics, anatomy or loading conditions. Because of the importance of knee alignment in the risk for knee osteoarthritis or the potential for implant loosening, finite element (FE) models have studied the effect of alignment on bone strains within the proximal tibia, however the role of the meniscus in the intact knee has not been investigated. Using the ISO 14243 standard, the aim of this study was to determine the medial-lateral ratio effects on contact force distribution, contact area and the location of the centroid of pressure in a finite element model of the natural knee joint.
Methods: A three-dimensional tibiofemoral joint FE model was built from magnetic resonance (MR) images of a healthy subject. Material properties for all soft-tissues were based on an integrated MR-imaging and computational validation study. A load of half body weight, 264.5N, was applied parallel to the tibial axis according to the ISO 14243 standard. To test for the effects of varus and valgus alignments, the load was then shifted 12mm medially and laterally. Shifting the load 12mm in either direction resulted in a 1.9 degree change in alignment. FE model outputs included medial-lateral reaction forces, contact forces transmitted through the meniscus and cartilage, contact area, and locations of the contact centroids on the tibial plateau.
Results: The 12 mm shift in applied loads resulted in a change in overall reaction force distribution from the ISO position, which was about a 60/40 medial/lateral ratio to 86/14 and 40/60, at 12mm medial and lateral respectively. The centroid of pressure does not move greatly as the load varies. Contact forces increased where the load was applied as expected. In all cases the menisci transmit the majority of the contact force, while the medial meniscus takes a large portion of the force regardless of the location.
Discussion: Our model suggests that substantial variations in static alignment do not necessarily result in large differences in pressure distributions on the tibial plateau. Although the ratio between medial and lateral reaction forces varied from 70/30 to 30/70, the portion of load supported by the meniscus remained relatively constant in this particular subject.
Arnold D. Gomez, M.S. [1]
[1] Mechanical Engineering Department, Rochester Insitute of Technology, Rochester, New York
Background: We present the methods for real-time measurement of fluidic parameters using the intrinsic pump signals of the LEVitated impeller Ventricular Assist Device (LEV-VAD). Each of these fluidic parameters; heart rate, blood flow rate of the pump, and differential pressure across a ventricular assist device can be used to assess patient health, pump status, and physiological blood flow requirements. Furthermore, the pump eliminates the need for dedicated sensors to measure the parameters because they are readily calculated directly from pump magnetic bearing control signals.
Methods: The pressure differential between the pump inlet and outlet, e.g. pressure differential between left ventricle and the aorta, cause an axial force on the rotor within the pump housing. The axial displacement can be perceived by the position sensors used for radial rotor stability. We correlate the real-time position sensor output to the differential pressure, and use it along with speed measurements to calculate flow based on a non-dimensional hydrodynamic performance curve.
Results: Bench top experiments were conducted to fully characterize the sensor output, differential pressure, and flow rate using blood analog at steady and pulsatile conditions. The pump was tested in vivo in acute trials to generate a comparative analysis between direct measurement of parameters (pressure, flow, and heart rate) and intrinsic measurements using the pump signals. The analysis shows a highly significant correlation within the target operating conditions.
Hung L. Chung, James L. McGrath/Ph.D [1], Jessica L. Snyder [2], David Z. Fang, Chris C. Striemer/Ph.D, Philippe M. Fauchet/Ph.D [3], Chris C. Striemer/Ph.D [4]
[1] Department of Biomedical Engineering, University of Rochester, Rochester, New York
[2] Department of Biophysics, University of Rochester, Rochester, New York
[3] Department of Electrical Engineering, University of Rochester, Rochester, New York
[4] SiMpore Inc., Rochester, New York
The use of membranes to fine-tune mass transport in microfludics is essential to many applications, including sample filtration, concentration, and micro-reaction. However, traditional polymeric membranes cannot be easily integrated into a microfluidic system and suffer from the lack of transport efficiency. In this work, we have presented a novel membrane, termed porous nanocrystalline silicon (pnc-Si), that is nanometer-thin and can be readily integrated into microfluidic systems. Because of its thinness, well-defined pores, and tight distribution, pnc-Si offers greater transport of materials as well as better resolution for filtration. Since pnc-Si is fabricated in a thin, planer silicon chip format by standard photolithography, it readily fits into microfluidic devices as a modular component. We have assembled a counterflow microfluidics with patterned PDMS, pnc-Si, and fused-silica capillary tubings and have used microspheres to visualize counterflow. We have also demonstrated that pnc-Si itself can be used as an inline electroosmotic pump to drive fluid flow. In fact, pnc-Si generated comparable flow rate and used less voltage than those typically required of electroosmotic pumps in the literature. These desirable properties of pnc-Si are likely attributable to the higher effective electric fields across the pores. Finally, we proposed a complex extension of pnc-Si-integrated microfluidics--- a shear-free chemotaxis chamber that can generate steady and complex 1D gradient (i.e. linear, delta, step, parabolic, and saw-shaped) within seconds to guide cell migration.
Brendan J. Morse [1], Amy L. Lerner [1], Paul D. Funkenbusch [2]
[1] Department of Biomedical Engineering, University of Rochester, Rochester, NY
[2] Department of Mechanical Engineering, University of Rochester, Rochester NY
Introduction: Osteoarthritis is related to altered mechanical stresses on the joint surfaces or underlying tissues, however it is unknown which tissue is most affected by these changes. Using a finite element (FE) model to non-invasively study mechanical responses in all tissues of the joint could be critical in understanding etiology and planning treatments or prevention. The aim of this study was to verify and validate subject specific models using several experimental studies to characterize contact variables, as well as stresses and strains in both bone and cartilage.
Methods: Three-dimensional proximal tibia FE models were built from magnetic resonance (MR) images of two healthy subjects to identify cartilage and bone boundaries. Loads were applied 50/50 onto the cartilage-cartilage and cartilage-meniscus contact areas using a previously presented pressure distribution method. The load was applied at about 20% of the gait cycle, equivalent to approximately 15-20 degrees of flexion. Loading magnitudes were varied to reflect the relevant experimental study, as well as the expected variation within the experiment. Eight different experimental studies were considered to study tibio-femoral contact variables such as peak and mean pressure, and contact area. To compare to the only reported in vivo measurement of cartilage strains during a static lunge activity and walking, a load of 3 times body weight (BW) was instantaneously applied to the model. Loads were applied to the model for investigating cortical and trabecular bone strains according to individual experimental studies. Regions analyzed in the experimental study were identified and predicted model values were compared.
Results: Experimental studies reflect wide variations in measurements of contact pressures and areas. Our FE model contact variables were within the ranges indentified. The model predictions are lower than Bingham (2008) as expected due to applying an instantaneous load. However our results are in close agreement with Liu (2010) which investigated normal gait. The mean cortical bone results for the anterior-medial and posterior regions are within 10% of the experimental results. In the trabecular bone, predicted FE results are a close approximation of the experimental principal strains.
Discussion: This study investigated multiple tissues in a FE model and found reasonable agreement within current literature for all areas. In collecting the appropriate studies for comparison it became there is a need for more experimental testing to address the variability seen within the current literature and to address the areas of limited understanding, such as cartilage and bone strains.
Junwen Mao, M.S. [1], Laurel H. Carney, Ph.D [2]
[1] Department of Electrical and Computer Engineering, University of Rochester, Rochester, New York
[2] Department of Biomedical Engineering, Department of Neurobiology and Anatomy, University of Rochester, Rochester, New York
Difficulty understanding speech in noise is a significant clinical problem. Despite decades of study, it is still not clear how listeners detect even pure tones in noise. This study focused on mechanisms for the diotic and the dichotic detection of a 500-Hz tone in wideband or narrowband reproducible noises. Previous analyses have focused on energy or temporal fine-structure cues over whole waveforms, yet listeners suggest that decisions are often based on short epochs. In this study, waveforms were separated into epochs to obtain temporally “local” cues, and the decision variables (DV) were computed by combining cues across epochs. For the dichotic case, the DVs were the means of the standard deviations of interaural time difference and interaural level difference across epochs. For the diotic case, the DVs were weighted sums across epochs, with weights based on reliability, evaluated from the distributions of cues for large sets of random noise-alone and tone-plus-noise waveforms using a likelihood ratio test. For most cues, these DVs yielded better correlations with the subjects’ results than the DVs computed for whole waveforms. Different epoch lengths were optimal for different cues. Epoch-based analysis also enabled an effective strategy for combining cues. Previous studies that fitted combined standard deviations of interaural time difference and interaural level difference cues to the data essentially selected the better of the two. In this study, cue weights for the dichotic case were derived based on the variability of the standard deviations of interaural time difference and interaural level difference information across time. With this method, correlation results improved compared to those for a single cue. For the diotic case, information provided by a single cue was combined across epochs, and then each cue was weighted based on its reliability. For most subjects, the resulting predictions were superior to those based on any single cue, and predictions depended on multiple cues, similar to actual performance, as opposed to being dominated by a single best cue. [Supported by NIDCD-DC001641]
Aaron F. Burger, B.S./Meng [1]
[1] Department of Mechanical Engineering, Rochester Institute of Technology, Rochester, New York
[2] Utah Artificial Heart Institute, Salt Lake City, Utah
In order to improve the performance of a magnetically levitated (maglev) axial flow blood pump, a three-dimensional (3-D) finite element method (FEM) is used to optimize the design of a hybrid magnetic bearing (HMB). Radial, axial, and current stiffness of multiple design variations of the HMB are calculated using a 3-D FEM package and verified by experimental results. As compared with the original design, the optimized HMB has twice the axial stiffness with the resulting increase of negative radial stiffness partially compensated for by increased current stiffness. Accordingly, the performance of the maglev axial flow blood pump with the optimized HMBs is greatly improved. The radial, axial and current stiffness of the HMB are found to be linear at nominal operational position from both 3-D FEA. Stiffness values determined by FEA and empirical measurements agree well with one another. The magnetic flux density distribution and flux loop of the HMB is also visualized via 3-D FEA which confirms the designers’ initial assumption on the HMB construction.
Tony Chen [1], Hani A. Awad [1]
[1] Department of Biomedical Engineering, University of Rochester, Rochester, NY
Surgical repair of articular cartilage typically yields fibrocartilage that lacks the stratified ECM architecture of native cartilage and does not integrate with the surrounding hyaline cartilage. Because of these limitations, there has been considerable interest in cartilage tissue engineering and that led to the development of various bioreactor systems to grow functional engineered cartilage constructs for in vivo implantation. Some of these bioreactors have used mechanical stimulation to enhance the synthetic activity of the cells while others have used perfusion fluid flow to enhance nutrient transport and produce homogenous constructs. To date, however, recapitulation of the zonal organization of native cartilage and the associated anisotropy in structure and mechanical properties in engineered constructs remains elusive. We hypothesize that in addition to improving nutrient transport via convection, interstitial fluid flow induces frictional drag and shear stress that can affect the alignment of fibrillar collagen in the constructs.
To test our hypothesis, we designed two systems, a parallel plate and rotating bioreactor systems, which create shear induced flow on the surface of chondrocyte-seeded agarose hydrogels, thereby generating depth-dependent interstitial flow fields within the hydrogel. The hydrogels were stimulated with fluid flow-induced shear stress up to 14 days. At that point, biopsies were collected from each hydrogel for biochemistry, histology, and in situ hybridization assays. Our results showed that compared to static controls, hydrodynamic bioreactor cultivation increases the synthesis of sulfated glycosaminoglycans and results in an aggrecan- and collagen II-rich superficial layer. Polarized light and 2-photon microscopy demonstrated alignment of a fibrillar collagen matrix parallel to the fluid flow streamlines on the rotating hydrogel surface, which was not observed in static controls. In situ hybridization data confirmed an increase in collagen II, aggrecan and lubricin gene expression in the flow stimulated layer compared to the static controls. These results demonstrate that fluid flow can induce cartilage ECM formation and fibrillar alignment in tissue engineered hydrogels and provides a promising approach for engineering stratified cartilage constructs that mimic the native tissue microstructure.
Etana C. Elegbe, M.S. [1], Stephen A. McAleavey, Ph.D [1]
[1] Department of Biomedical Engineering, University of Rochester, Rochester, NY
[2] Rochester Center for Biomedical Ultrasound, Rochester, NY
The Spatially Modulated Ultrasound Radiation Force (SMURF) method is a technique for estimating the shear modulus of an elastic material. SMURF determines shear modulus through measurement of the frequency of shear waves of known wavelength propagating in a medium of unknown modulus. Shear modulus is then computed as the product of the medium's density and the square of the product of the shear wave temporal frequency and wavelength. Since acoustic radiation force is proportional to intensity, generation of this spatially varying force is a matter of determining how to generate an equivalently varying ultrasound intensity field. Presented in a previous study are the Focal Fraunhofer and Intersecting Plane wave methods, which are two of the techniques that can be used to create the desired beam intensity pattern within a region of interest. To create a push that results in discernible displacements while maximizing the signal to noise ratio, we investigate the use of a multi-row 1.5D array over 1D linear array in implementing the aforementioned methods. A 1.5D array allows for more aggressive focusing in the elevation direction and thus a greater intensity in the region of interest. The efficiency of both configurations is analyzed based on the ability to generate beams of the desired well-defined spatial wavelength at various focal depths, the ability to localize the pushing beam to the region of interest and the ability to maximize the depth-of-field.
Xiaoxing Han [1], Edward B. Brown [2]
[1] Institute of Optics, University of Rochester, Rochester, NY
[2] Department of Biomedical Engineering,University of Rochester, Rochester, NY
Second Harmonic Generation (SHG) has proven to be a useful window into the amount and organization of fibrillar collagen in biological tissues due to its relative specificity and the fact that it is an intrinsic signal. SHG is a coherent phenomenon, which implies that SHG is sensitive not only to the amplitude of the illumination field but also to its phase. In addition to a spatial resolution that is equal to other imaging techniques (such as two photon excited fluorescence), SHG microscopy can provide information about the sample¡¯s molecular structure. For example, the ratio of the forward-propagating to backward propagating SHG signal (the ¡°F/B ratio¡±) can help us to understand the axial extent of ordering in collagen fibers.
Previously, in vitro measurements of SHG F/B ratios have been used to study collagen fiber ordering in various tissue samples such as rat tail, mouse models of breast cancer, and dermis from mouse models of Osteogenesis Imperfecta (OIM). In these measurements a second objective lens was needed to collect forward propagating SHG signal. Hence, the tissue sample had to be dissected from the animal and sectioned to 100um slices to allow signal to reach the second detection lens. For clinical application, such as in endoscopy, it is impossible to put an objective and a PMT detector underneath the tissue sample to collect the forward propagating SHG. The excision and sectioning required to use a second detector for forward propagating SHG also prevents dynamic measurements of collagen ordering over time.
Here, we present a method to determine, for the first time, the ratio of forward-propagating SHG signal to back-propagating SHG signal (F/B) in vivo on the surface of intact tissue samples without any biopsy or tissue sectioning, using only epidetection (i.e. via a single objective lens). This method has the additional benefit of using the confocal detection apparatus already contained within common commercially available two-photon laser-scanning microscopes, and hence can allow the measurement of the SHG F/B ratio in vivo with minimal purchase of new equipment.
Kelley A. Garvin, M.S. [1], Denise C. Hocking, Ph.D. [1], Diane Dalecki, Ph.D. [1]
[1] Department of Biomedical Engineering and The Rochester Center for Biomedical Ultrasound, University of Rochester, Rochester, New York
Controlling the spatial organization of cells and proteins within 3D engineered tissue represents a promising strategy to regulate cell behaviors that are essential for generating replacement tissues. Acoustic radiation forces in ultrasound standing wave fields (USWF) can align cells in suspension into planar bands. We demonstrate that USWF can control the spatial distribution of cells within 3D collagen-based engineered tissue. In this study, fibronectin-null myofibroblasts (FN-/- MF) suspended in a type-I collagen solution were exposed to a 1 MHz USWF for 15 min. Collagen solutions were allowed to polymerize during the exposure to maintain the USWF-induced cell distribution after removal of the sound field. Resultant collagen gels were characterized by a distinct pattern of banded cells. USWF exposure did not decrease cell viability. Aligning FN-/- MF into this arrangement resulted in a 1.5-fold increase in cell-mediated collagen gel contraction, suggesting that USWF-mediated cellular organization enhanced cell contractility. We also demonstrate co-localization of the extracellular matrix protein FN to USWF-induced cell bands by binding soluble FN to FN-/- MF prior to USWF exposure. In summary, USWF can control the spatial organization of cells and cell-bound proteins within 3D engineered tissue and the USWF-induced cell organization enhances cell contractility. This technique has applications to the fabrication of tissue constructs with improved mechanical properties, and thus has the potential to significantly advance the development of functional artificial tissues.
Benjamin Freer [1], Jayanti Venkataraman [1]
[1] Department of Electrical and Microelectronic Engineering, Rochester Institute of Technology, Rochester, NY
Diabetes is a condition in which insulin resistance results in an inability to control glucose levels in the blood stream. Monitoring of blood glucose levels is fundamental to diabetes care. Most at-home monitoring is performed with a blood glucose meter. Continuous monitoring systems also exist, but they require a subcutaneous sensor to be replaced every 3 to 7 days. While these monitoring systems are effective in reducing blood glucose to recommended levels, it is a painful long term method.
The present work investigates the feasibility of a non invasive monitoring of glucose levels using an antenna placed in a cuff and wrapped around the wrist without extraction of blood samples or subcutaneous sensors. Tissues exhibit frequency dependent permittivity and conductivity that vary vastly between tissue and for the same tissue in its healthy and diseased states. It has been shown in the literature that the concentration of glucose within red blood cells (erythrocytes) affects the lipid density of the cell walls, significantly changing the blood permittivity. The method is based on determining the complex dielectric permittivity of blood from a measurement of the input impedance of the antenna and the shift in its resonant frequency from which the dielectric permittivity of blood is determined and related to the glucose levels.
A wideband monopole antenna is used in this work that operates in the MHz range for better penetration into the tissue. Using a realistic tissue model of a human hand, simulations have been performed with Ansys's High Frequency System Simulator (HFSS) to obtain the antenna input impedance. The resonant frequency is seen to shift with changes in blood glucose levels. Based on our previous work of tissue characterization, an analytical technique is developed to relate this frequency shift to the permittivity and conductivity of blood, from which the glucose levels are determined.
Liwei An, M.S. [1], Bradley Mills [1]
[1] Department of Electrical and Computer Engineering, University of Rochester, Rochester, NY
[2] Department of Biomedical Engineering, University of Rochester, Rochester, NY
OBJECTIVE : To quantify the elastic contrast for heterogeneous elastic phantoms and to evaluate frequency dependence, bias, and spatial resolution of lesion detection based on Crawling Wave (CrW) sonoelastography. We focus on parameters of particular interest in prostate cancer detection.
METHODS: Heterogeneous elastic phantoms were prepared by embedding stiff spherical inclusions (11% gelatin) with diameter of 2 cm, 1 cm and 6 mm in otherwise soft homogeneous backgrounds (6% gelatin) separately. CrWs at 70 Hz, 100 Hz, 120 Hz, 140 Hz, 200 Hz and 300 Hz were acquired by offsetting a small frequency difference between two sources positioned at each side of the phantom. The CrW movies were first normalized to compensate for the gain difference at different depth levels of the image. A sinusoidal curve fitting over one cycle of crawling waves was then applied to improve the SNR. A 2-D local shear velocity estimator was employed with the CrW phase pre-conditioned to a favorable range of the estimator. A mean estimation over frequencies was obtained by globally selecting and averaging 90% qualified data points. Finally, the shear velocity information from both the hard region and the soft region was extracted from the estimation map and was compared against the reference values obtained from homogeneous phantoms.
RESULTS: The 2 cm inclusion phantom gives the highest contrast of hard and soft regions at 1.29 averaged over frequency, followed by the 1cm case at 1.27 and the 6mm case at 1.14. The estimations followed an upward tendency with frequency increase due to the higher resolution of CrW at higher frequencies. Better contrast was acquired for inclusions with larger size. The effect of under-estimation and loss of contrast, which is caused by noise and the spatial support of the estimator, was revealed.
CONCLUSIONS: Contrast details were demonstrated for phantoms with stiff inclusions of different sizes. The experiment showed the ability of the shear velocity estimator to distinguish the lesion from the background. Lesions similar in elastic contrast to prostate cancer that are 6 mm in diameter or larger are resolvable at frequencies above 100Hz, using our current system and methods.
Michael Pecoraro [1], Jayanti Venkataraman [1], Gill Tsouri [1], Sohail Dianat [1]
[1] Department of Electrical and Microelectronic Engineering, Rochester Institute of Technology, Rochester, NY
Cranially implanted sensors and electrodes have been used in practice for several years; their applications range from the recording of neural signals for use in Brain-Computer Interfaces to help the disabled, to the treatment of diseases and conditions ranging from Parkinson's disease, multiple sclerosis, depression, etc. Current communication methods with implants, however, are lacking; they run the gamut from physical, percutaneous connections that increase the risk of infection, to wireless links that are slow and uncomfortable for patients.
The present work focuses on the characterization of the effects of the human head on communication with cranially implanted antennas for its eventual use in improving current communication methods. A realistic human head model with frequency dependent tissue characteristics is used to obtain a transfer function that describes the magnitude and phase of an electromagnetic wave as it propagates through the human head over both frequency and depth into the skull; this data is obtained for multiple energy entry angles. The technique used to obtain transfer function measurements consists of taking the ratio of the electric fields at the receiver and transmitter and is developed through analysis of ultra-wideband transmit/receive antenna systems; verification for this technique is provided.
After the transfer function data described above is obtained, we posit a communication model to approximate the transfer function magnitude. This approximation takes the form of a modified log-distance, log-frequency path loss model and fits the data quite well. The final approximation describes the path loss of an electromagnetic wave over both frequency and distance for all simulated orientations.
Lastly, simulations are presented for communication from a cranially implanted dipole antenna. The received power of an external antenna - whose position is varied in both distance (from the head), as well as location (around the head) - is captured and plotted. We finally show that the transfer function that was obtained for all perpendicular communication through the head is able to, in most cases, correctly predict the results of these received power simulations.
Carlos Sevilla, M.S. [1], Diane Dalecki, Ph.D [1], Denise Hocking, Ph.D [1]
[1] Department of Biomedical Engineering, University of Rochester, Rochester, NY
[2] Department of Pharmacology and Physiology, University of Rochester, Rochester, NY
Fibronectin is an adhesive glycoprotein that is polymerized into extracellular matrices (ECM) via a tightly-regulated, cell-dependent process. Here, we demonstrate that fibronectin matrix polymerization induces the self-assembly of multicellular tissue-like structures in vitro. Fibronectin-null mouse embryonic fibroblasts (FN-null MEFs) adherent to compliant gels of polymerized type I collagen failed to spread or proliferate. In contrast, addition of fibronectin to collagen-adherent FN-null MEFs resulted in a dose-dependent increase in cell number, and induced the formation of three-dimensional (3D) multicellular structures that remained adherent and well-spread on the native collagen substrate. An extensive, fibrillar fibronectin matrix formed throughout the microtissues. Blocking fibronectin matrix polymerization inhibited both cell proliferation and microtissue formation, demonstrating the importance of fibronectin fibrillogenesis in triggering cellular self-organization. Cell proliferation, microtissue formation, and microtissue shape were dependent on both fibronectin and collagen concentrations, suggesting that the relative proportion of collagen and fibronectin fibrils polymerized into the ECM influences the extent of cell proliferation and the final shape of microtissues. These data demonstrate a novel role for cell-mediated fibronectin fibrillogenesis in the formation and vertical growth of microtissues, and provides a novel approach for engineering complex tissue architecture.
Dean G. Johnson, M.S. [1], Robert D. Frisina, Ph.D. [2], David A. Borkholder , Ph.D. [3]
[1] Department of Microsystems Engineering,Rochester Institute of Technology, Rochester, NY
[2] Departments of Otolaryngology , Biomedical Engineering, and Neurobiology & Anatomy, University of Rochester School of Medicine and Dentistry, Rochester, NY
[3] Department of Electrical and Microelectronic Engineering, Rochester Institute of Technology, Rochester, NY
Small mammals, particularly mice, are very useful animal models for biomedical research. Implantable Microsystems are used for site directed delivery of gene vectors and / or therapeutic compounds and require fluidic interconnection to other Microsystems or biological systems. In space constrained applications such as the human mastoid cavity, subcutaneous implantation, or small animal studies, the form factor and volume of the fluidic interconnects is of critical importance. For medical implant applications, additional coupling structures and out-of-plane approaches add unacceptable volume. Coupling approaches relying on adhesives (epoxies and elastomers) require micro-injection techniques, and can result in either blocked capillaries or gap formation and dead volumes depending on the material viscosity and gap widths. Existing microfluidic interconnect technologies fail to reliably meet the combined space and biocompatibility requirements of implantable Microsystems that are needed for many clinical applications and for use in small animal model systems such as mice.
A method for coupling external fluidic systems to microfluidic channels via in-plane interconnects is presented including modeling, fabrication, and testing with emphasis on added interconnect volume, dead volume, resistance to leakage from internal pressure, and robustness to applied force on the extending capillary. Capillary tubing is inserted into channels etched in the surface of a silicon wafer with a seal created by Parylene-C deposition. Characterization of Parylene-C deposition into channels was used in conjunction with equations describing molecular flow and diffusion to spatially model monomer concentration and polymer deposition within tapered channels. Device fabrication is described with testing on several designs bracketing the predicted optimum channel dimensions. Testing includes measurement of deposited Parylene and void characterization along the length of the interconnect, burst / leakage pressure, and pull-strength.
Low volume interconnects using biocompatible, chemical resistant materials have been demonstrated and shown to withstand pressure as high as 827 kPa (120 psi) with average pull test strength of 2.9 N. Each interconnect consumes less than 0.018 mm3 (18 nl) of volume. The low added volume makes this an ideal interconnect technology for medical applications where implant volume is critical. These results along with overall interconnect volume, dead volume and fabrication complexity are compared against commercial and other research devices.
Aswin Dhamodharan M.S [1]
[1] Department of Industrial and Systems Engineering
[2] Rochester Institute of Technology
[3] Rochester
[4] NewYork
Vaccines are perishable products stored in vials with a specified expiry period. The expiry period corresponds to the length of time for which vaccine doses stored in a vial can be safely used to immunize children since the time of opening (puncturing) of the vial. Vaccines are products with limited supply. The overall vaccine wastage rate in developing countries has been observed to be 50% by WHO and UNICEF. Hence there is a need to recommend inventory policies and an optimal vial size to be used to minimize vaccine wastage. During immunization campaigns, vaccine expiration occurs when not enough children in need of vaccination are found during the expiry period of an open vaccine vial, resulting in discarding of the remaining doses in the vial. Storing only one dose of vaccine in a vial could prevent the expiration from happening. However, doing so increases production and handling costs for the immunization campaign. This study aims to determine the optimal number of vaccine doses to be stored in a vial such that total cost associated with an immunization campaign is minimized. This is done by solving two sub-problems; the first problem considers the perishable content of an open vaccine vial as inventory to capture the wastage due to expiry and the wastage cost associated with it. The second problem deals with the vaccine vial reordering policy and positive lead-time. Poisson process is assumed to generate demand. An approximate expected cost function for the entire problem is obtained using policies from the literature and is evaluated using Mathematica. A Mixed Integer Program is used to simulate the problem and compute average cost for a given set of parameters and is compared with the expected cost obtained from the approximate cost function mentioned above.
References:
• G Hadley and T.M Whitin. Analysis of Inventory Systems. Prentice Hall, Inc., Englewood Cliffs, N.J., 1963
• Eylem Tekin, Ulku Gurler, and Emre Berk. Age-based vs. stock level control policies for a perishable inventory system. European Journal of Operational Research, 134(2):309 – 329, 2001.
Melissa A. Monahan [1]
[1] Department of Mechanical Engineering, Rochester Institute of Technology, Rochester, New York
The goal of this thesis project is to determine the lower limit of scale for the RIT robotic grasping hand. The problem of scalability aims to bridge the gap between the complex life-size robotic hand and less adept micro-manipulators currently available. Taking the functionality of the robotic hand and scaling down the design will result in a more sophisticated and precise tool for applications such as micro-surgeries.
This is accomplished by conducting a force analysis that will determine the size of air muscles required, to achieve appropriate contact forces at a smaller scale.
Input variables such as the actuation force and tendon return force are experimentally determined. A dynamic model of the hand system is created and used to predict the contact (grasping) force of the fingers. A correlation between the model and physical testing is achieved for both a life-size and half-scale finger assembly. Both half and quarter-scale robotic hand rapid prototype assemblies are built on 3D Printers. This thesis work identifies the point where further miniaturization would require a change in the manufacturing process to micro-fabrication.
Several techniques are compared as potential methods for making a production intent quarter-scale robotic hand. Investment casting, Swiss machining, and Selective Laser Sintering are the manufacturing techniques considered. A quarter-scale robotic hand would test the limits of each technology. Below this scale, micro-machining would be required. The break point for the current actuation method, air muscles, is also explored. Below the quarter-scale, an alternative actuation method would be required. Electroactive Polymers are explored as a potential option.
In summary, a dynamic model of the robotic hand is created and validated as scalable. The model is used to determine finger contact forces at the quarter-scale. The break point in terms of the current RIT robotic grasping hand is identified at the quarter-scale for both manufacturing and actuation. A novel quarter-scale robotic hand assembly is built by an additive manufacturing process, a high resolution 3D printer.
Next steps for the Robotic Hand Platform include applying controls to the half and quarter-scale robotic hand assemblies. Design recommendations based upon this thesis work are made to improve the robustness of the mini-hands. Future work may also include the incorporation of haptic feedback, which has become a standard for robotic surgical tools.
Xiaofan Luo [1], Patrick T. Mather [1]
[1] Department of Biomedical and Chemical Engineering, Syracuse University, Syracuse, NY
Shape memory polymers (SMPs) are responsive polymers that are capable of changing shapes on-demand. The broad application of SMPs requires material design strategies that allow not only the precise tailoring of shape memory properties but also facile incorporation of additional functionalities. In this poster we describe the fabrication and characterization of a series of shape memory polymer composites. The basic approach involves physically combining two or more components into an interpenetrating fiber/matrix structure, allowing them to function in a synergistic fashion yet remain physically separated. This latter aspect is critical since it enables the control of overall composite properties and functions by separately tuning each component. Utilizing the intrinsic versatility of this approach, novel properties and functions (in addition to “regular” shape memory) have been achieved in our composite systems including (1) being elastomeric in the application temperature range, (2) triple-shape memory (the ability to fix two independent temporary shapes and recover them sequentially upon heating), (3) high conductivity for fast electrical actuation, and (4) self-healing of mechanical damage.
Kevin A. Davis, B.S. [1], Kelly A. Burke, Ph. D [3], Patrick T. Mather, Ph. D [1], James H. Henderson, Ph. D [1]
[1] Department of Biomedical and Chemical Engineering, Syracuse University, Syracuse, New York
[2] Syracuse Biomaterials Institute, Syracuse, New York
[3] Department of Macromolecular Science, Case Western Reserve University, Cleveland, Ohio
Substrate topography can direct cell behavior. Shape-memory polymers (SMPs) are a class of active materials that have the ability to "memorize" a permanent shape, be manipulated and "fixed" to a temporary shape, and then later recover to the permanent shape by a triggering event. We hypothesized that programmed changes in surface topography of an SMP substrate could be used to direct cell behavior.
To test this hypothesis, we analyzed cell morphology on an SMP substrate programmed to change from a flat to a grooved topography and an SMP substrate programmed to change from a grooved to a flat topography. The shape memory response was activated via a hydro-thermal trigger. A photocured, glassy SMP was cured on an imprint of a vinyl record to produce films with a permanent surface topography of triangular grooves or between glass slides to produce films with a permanent flat topography. Samples were then flattened or embossed with the imprint of the vinyl record by compression above Tg and fixed by cooling to room temperature under force. This produced substrates with a temporary flat or a temporary grooved topography. C3H10T1/2 cells were allowed to attach for 5 h at 28C and then shape recovery was triggered by increasing to 37C (above the hydrated Tg). Two groups were tested to image cell morphology (i) before recovery and (ii) after recovery. Samples recovered by heat to their permanent topography and cultured at 28C and 37C were the control.
We found that a programmed change in surface topography of an SMP substrate can be used to direct cell morphology. Cells on recovered grooved samples showed greater alignment (13.1+/-0.7 degrees, aligned with groove = 0 degrees) along the grooves compared to unrecovered flat samples (23.9+/-1.2 degrees, p < 0.05). Qualitatively, cells on recovered flat samples showed a uniform angular distribution indicating random alignment while cells on unrecovered grooved samples showed a peak indicating alignment in a preferred direction.
Our results suggest that active substrates that incorporate surface shape memory could provide an unprecedented approach for controlling cell behavior during cell culture.
Alexander E. Ship [1], Matthew N. Giarra [1], Steven W. Day [1]
[1] Department of Mechanical Engineering, Rochester Institute of Technology, Rochester, NY
The lack of standard procedures for the use of computational fluid dynamics (CFD) to evaluate biomedical devices safety has led to a study of CFD analyses on a benchmark geometry. The benchmark geometry consists of, depending on flow direction, a conical nozzle (or diffuser) and a sudden expansion (or contraction). In a prior study, analysis was performed by 28 groups including self-identified beginner, intermediate, or expert CFD users.
Flow visualization data on the benchmark model was collected at three locations using particle image velocimetry (PIV) on physical models made out of acrylic and the resulting data was used to evaluate the accuracy of the submitted CFD data. Currently, similar plastic models are being used to evaluate red blood cell damage caused by flow through the geometry. The plastic models were manufactured at RIT to a 1% tolerance. The primary manufacturing issues were achieving adequate surface finish, optical quality and maintaining geometrical accuracy, specifically corner sharpness.
This work presented here includes an investigation on the manufacturing aspect of the experimental models as well as the use of a commercial CFD package to obtain a solution.
Adequate surface finish and optical quality required a series of polishing techniques. Initially, wet sanding was used to remove the tool markings, and then progressively finer polishing compounds were used to polish the surfaces to specification. However, the sharpness of the expansion corner has been difficult to maintain through the polishing process, as such the plastic model may not be an accurate representation of the theoretical model in some of these models. As this corner is a region of high shear stress, it is possible explanation for the variation in damage caused by the physical models.
In order to improve this corner sharpness, a polishing technique was developed to consistently achieve surface finish requirements. Using a cylindrical felt bob, progressively finer polish compounds, and a mill, the perpendicular face of the expansion (or contraction) region was polished while maintaining a corner radius of less than 0.001 inch.
The advances in manufacturing enable more accurate experimental data for both flow performance and blood damage studies. Without reliable experimental models it is impossible to develop standardized methods for evaluation of biomedical devices.
In parallel with the manufacturing efforts and FDA round-robin, the benchmark model will be analyzed by a student as a final project in the CFD Applications course at RIT in the Mechanical Engineering department. This course instructs students on the basic applications of mesh generation (Gambit) and CFD (Fluent).
Matt Kenyon [1], Dean Johnson [2], George Smith [3], Chris Trimby [3], David Borkholder [1]
[1] Dept of Electrical and Microelectronic Engineering, Rochester Institute of Technology, Rochester, NY
[2] Dept of Microsystems Engineering, Rochester Institute of Technology, Rochester, NY
[3] Dept of Physiology, Chandler Medical Center, University of Kentucky, Lexington, KY
Traditional neuronal interfaces rely on electrical stimulation for excitatory or inhibitory control; stimulation which is incapable of targeting specific neuronal subpopulations. The emerging field of targeted expression of light activated cationic or anionic channels (channelrhodopsins) with response times equivalent to ligand gated ion channels offers the opportunity for population based photonic stimulation. This work aims to demonstrate functional stimulation for locomotion in rats following a complete spinal cord transaction at T8-T9. Channelrhodopsin-2, a light activated non-specific Na+ ion channel, will be expressed in select neuronal populations within the injured spinal cord and spatially stimulated with blue light coupled to the spinal cord through fiber optics and a photonic microsystem. A scalable optical stimulation system has been developed which allows activation of multiple sites independently. The system consists of control electronics which independently drive four high intensity blue LEDs. These LEDs are directly coupled to 200μm diameter optical fibers which terminate at a custom fabricated, implantable, optical stimulation chip. Each of the LEDs intensity, pulse width, frequency and delay are individually controlled through a graphical user interface (GUI). The implantable chip was created by etching out V-groove channels and a passive mirror in single-crystal silicon. The system has proven capable of producing light outputs greater than 200 mW/mm2 and has elicited action potentials in DRG neurons expressing channelrhodopsin-2 in culture. The system architecture is scalable to larger numbers of channels and easily adaptable to an embedded, stand-alone solution.
Sylvan Hemingway [1]
[1] Mechanical Engineering, Rochester Institute of Technology
[2] Rochester, New York
Robotics devices have become commonplace in our world, and are developing rapidly, with many challenges to over come. An example of a robotic device in this realm is a device known as the pneumatic muscle actuator (PMA) From the time of their development by Mckibben et al. in the 1960’s as a potential prosthetic device these muscles have been investigated thoroughly. The muscles primarily consist of an inner latex tube, and an outer nylon braided shell. With an input of pressurized air to the inner tube, the muscle expands radically, while the braided structure of the nylon weave causes the muscle to contract axially. The motion generated is analogous to the movement of an organic muscle, exerting a contractile force.
The motions of these muscles have been proven controllable as means for robotic actuation. Two muscles in an antagonistic configuration provide a robotic joint. These joints can act as elbow, knees, and ankles. Full arm configurations have been produced and have an advance controls with high levels of dexterity. The limitation to these instruments is providing the muscles with a pressurized air source. Compressors and pressurized air canisters are restrictive due their size and weight. However, the muscles do have potential as a physical therapy device. The air muscles can be configured to be worn and provide a guided motion. The advantages of the air muscles are that they are lightweight, and compliant. They are ideal for human robotic interaction.
The current research projects with the muscles in the Bioengineering lab at RIT include investigation control of a robotic hand, and investigation of the potential uses as an underwater actuation device. While the drive for an underwater actuator is primarily for underwater robotic vehicles, the muscles could aid in water physical therapy. The water provides a low stress environment and a strengthening fluid resistance. An exoskeleton device designed to work in an aqueous environment has potential to act as a guiding mechanism for those in therapy.
Daniel C. Roy, M.S. [1], Denise C. Hocking, Ph.D [2]
[1] Department of Biomedical Engineering, University of Rochester, Rochester, New York
[2] Department of Pharmacology and Physiology, University of Rochester, Rochester, New York
Protein engineering approaches may be used to develop small, biologically-active proteins that can stimulate tissue regeneration by promoting cell behaviors critical for wound repair, including cell adhesion, growth, and migration. Fibronectin is a large adhesive glycoprotein that is normally rapidly up-regulated in response to tissue injury. Decreased fibronectin levels are associated with non-healing wounds. The insoluble, extracellular matrix (ECM) form of fibronectin stimulates cell growth, spreading, migration, and contractility in part via an exposed heparin-binding site in the III1 module of ECM fibronectin. We have engineered a recombinant, ECM fibronectin mimetic (GST/III-1H,8-10) that couples the heparin-binding fragment of FNIII1 (FNIII-1H) to the integrin-binding domain of fibronectin (FNIII8-10). GST/III-1H,8-10 supports cell adhesion and spreading and increases the rate of cell proliferation to a greater extent than full-length fibronectin. To produce smaller, equally bioactive ECM fibronectin mimetics, a series of deletion mutants of GST/III-1H,8-10 were constructed. Protein constructs containing the integrin-binding, RGD sequence, FNIII-1H and FNIII8 stimulated cell growth to a similar extent as GST/III-1H,8-10. When used as adhesive substrates, the fibronectin matrix mimetics supported fibronectin matrix assembly and increased the rate of cell migration compared to full-length fibronectin. These data indicate that small engineered ECM fibronectin mimetics containing the integrin-binding RGD sequence together with FNIII-1H and FNIII8 can serve as bioadhesive substrates to promote cellular activities critical to tissue repair.
Keerthi S. Valluru, M.S. [1], Bhargava K. Chinni, M.S. [1], Navalgund A. Rao, Ph.D [2], Shweta Bhatt, M.D. [1], Vikram S. Dogra, M.D. [1]
[1] Dept. of Imaging Sciences, University of Rochester, Rochester, NY
[2] Center for Imaging Sciences, Rochester Institute of Technology, Rochester, NY
From a fundamental perspective, image reconstruction tasks in transmission based ultrasound, pulse-echo ultrasound and photoacoustic imaging are similar. We propose a C-scan imaging scheme that is applicable to transmission based ultrasound and photoacoustic imaging modalities where the image reconstruction is achieved through focusing action of an acoustic lens. The proof of concept is demonstrated. Experimental methodology to determine the system point-spread-function is outlined and preliminary results are presented.
Sneha Narayanan [1], Sally P. Duarte [1], David J Pinto, PhD [3]
[1] Department of Biomedical Engineering, University of Rochester, Rochester, New York
[2] Department of Neurobiology and Anatomy, School of Medicine and Dentistry, University of Rochester, Rochester, New York
[3] Strong Epilepsy Center, School of Medicine and Dentistry, University of Rochester, Rochester, New York
The onset of an epileptic event, or ictogenesis, is deceptively simple to understand. A group of errant neurons become active and recruit activity in other neurons until a threshold is reached and the neuronal network explodes into an epileptic event. Describing the mathematics of this process rigorously, however, is more difficult. How can we use standard nonlinear analysis to frame the concept of network threshold? What experiments might we perform that can translate dynamic parameters into testable hypotheses? How can abstract concepts from dynamic systems lead to meaningful insights into clinical epilepsy? Our goal in this project is to answer these questions.
Understanding the dynamics of ictogenesis entails both mathematical and experimental methods. Mathematically, we use concepts from nonlinear analysis to derive a two-parameter model of a double-well potential that describes network threshold. We then use the model to design specific experimental methods to relate directly experimental data with model parameters. Finally, we carry out one of our proposed experiments to demonstrate how our model analysis can capture differences in the ictogenic properties between a region of the brain known for epileptic activity (piriform cortex) versus a region that is clinically stable (somatosensory cortex).
Taken together, our results suggest that our nonlinear model effectively describes the dynamics of experimental ictogenesis. These results represent a first step toward using nonlinear dynamics for devising targeted control strategies for preventing an epileptic event before it starts.
Amit Chainani, B.S. [1], Steven W Day, Ph.D [1], Risa J Robinson, Ph.D [1]
[1] Department of Mechanical Engineering, Rochester Institute of Technology, Rochester, New York
This study focuses on optical visualization of the flow field inside a Johnson & Johnson OCD MicroWell, a mixing device that is used with the Vitros ECI analyzers, manufactured by J&J. This product is used to mix blood serum samples and reagents during various assay cycles of the machine. A test rig emulating the injection and mixing process of the actual system has been built. This rig allows optical access for extensive PIV (Particle Image Velocimetry) and PLIF (Planar Laser Induced Fluorescence) analysis. The techniques have been implemented and optimized for accurate and real-time assessment of velocity and concentration during primary and secondary mixing stages of the process. Concentration profiles various stages of the process obtained using PLIF have been quantified to estimate “mixedness” of the fluid inside the MicroWell. Results from both PIV and PLIF experiments studied in conjunction and quantified for aid of future assay and immunodiagnostics development.
Xinzhu Gu [1], Patrick T. Mather, Ph. D [1]
[1] Department of Biomedical and Chemical Engineering and Syracuse Biomaterials Institute, Syracuse University, Syracuse, NY
Shape-memory polymers (SMPs) are stimuli-responsive materials which possess the ability to memorize permanent shapes that differ substantially from their temporary shapes. With an external stimulus, they change shapes to temporary ones which can be recovered by heating up to their transition temperatures. Nowadays, SMPs have drawn much attention because they possess a lot of advantages over shape memory alloys such as high-shape recovery, easy processing, and low manufacturing cost. Biocompatible and biodegradable shape-memory polymers with appropriate thermal and mechanical properties are expected as a unique approach to medical devices such as surgical sutures and catheters.
In this work, we hypothesized that entanglements could engender memory for multiblock polyurethanes. Thus, a family of hardblock-free multi-block thermoplastic polyurethanes (TPUs) consisting of PCL and poly(ethylene glycol) (PEG) was synthesized and characterized. As we expected, due to thermodynamic incompatibility between hydrophilic PEG blocks and hydrophobic PCL blocks, micro-phase separation happened, resulting in a unique “reversible plasticity” shape memory (RPSM) property. Remarkably, the recoverable deformations of all the samples are larger than 1200%. When heating upon the transition temperature, the highly deformed samples rapidly and completely recovered. Meanwhile, the entanglements that exist above the melting temperature further promote strain recovery and prevent flow. The large deformation, fast actuation and good recovery make these materials potential candidates for applications in such medical devices as self-tightening sutures.
Leif Melhus, B.S./M.Eng. [1], Michael Moy, B.S. [1], Risa Robinson, Ph.D [1], Edward Hensel, Ph.D [1]
[1] Department of Mechanical Engineering, Rochester Institute of Technology, Rochester, NY
Electronic cigarettes are the latest consumer product designed to provide cigarette smokers with a low-risk nicotine delivery system. Organic solutions containing nicotine are heated and delivered to the user as a nicotine enriched aerosol. These delivery systems contain a solution cartridge capable of storing the equivalence of two packs of standard cigarettes, an atomizer comprising of a heating element or ultra-sonic technology, flow channels, a battery, circuitry, and a LED all packaged in the size of a conventional cigarette. Electronic cigarettes are believed to contain no carcinogens, no particulate matter, and are less harmful than conventional cigarettes. At this time, there is no governmental regulation of these products. In addition, there are no standard test procedures. RIT has an established track record of modeling and performing experimental work in particle deposition in the human respiratory tract. Development of an electronic cigarette evaluation system will position RIT for future research contributions.
The primary goal of the Electronic Cigarette Evaluation System is to design, build, and utilize a machine capable of smoking and collecting the matter exhausted by an electronic cigarette. One electronic cigarette can be smoked per trial for a given time interval & flow rate. This machine must also be able to smoke the electronic cigarette simulating the puff profile of an actual smoker. To achieve all of these requirements, a vacuum box design was implemented. Not only does the vacuum chamber smoke the e-cigarette, but also simulates a human lung-breathing action during smoking. The system varies the smoking conditions by controlling the puff length and the flow rate through the pump via a Lab View based GUI. Both particulate & vapor contaminates can be collected using Cambridge filter pads and sorbent tubes, which are optional add-ons to the system.