Bioengineering Seminar Schedule
Fall 2003
Friday, September 5, 12:00 - 1:00 pm, Room 210 Hallowell at University Park and CG623 at Hershey
Wall Shear Rate Estimation Within the 50cc Penn State Artifical Heart Chamber
Using Particle Imagei Velocimetry (PIV)
Pramote Hochaeron
Penn State University
Abstract
A major problem after implantation is a clot formation within the chamber. Wall shear stress, is a prime suspect
in the undesirable clotting. Particle Image Velocimetry (PIV) was employed to estimate wall shear-rates within
the artificial heart. Three sources of errors in estimating wall shear rate from the PIV results are near wall
velocity bias, vector location errors, and the inaccuracy of a discrete gradient calculation.
Three techniques are introduced to resolve these problems, which we term: "zero masking", "fluid
centroid shift", and "wall coordinate decomposition." A first order (linear) interpolation with
a no-slip boundary condition was used to calculate wall shear-rate. Simulations were performed to investigate the
factors affecting the accuracy of the wall shear-rate estimation technique.
Interestingly, the results show the "first order" wall shear-rate estimation technique performs better
for the quadratic boundary layer profiles than for the linear profiles. A large interrogation window, 32×32
pixels, does not perform well for high wall gradients. The effect of the wall curvature is observed for a small
interrogation window, 8×8 pixels, and presents as a bias proportional to the wall gradients. Wall reflections
do not play a critical role for the simulations.
The best case is the quadratic boundary layer profile on the straight surface. A 16×16 pixels interrogation
window seems to be the most practical for our application.
Friday, September 12, 12:00 - 1:00 pm, Room 210 Hallowell at University Park and CG623 at Hershey
Aspects of the History of Bioelectricity and Electrocardiography
David Geselowitz
Penn State University
Abstract
The histories of bioelectricity, the physics of electricity, and electric technology, especially instrumentation,
are intertwined. Nerve and muscle cells are electrically active. Pioneering work in electrophysiology was conducted
by Galvani who studied the nerve and muscle of the frog leg. A challenge to Galvani's interpertation of his results
by Volta led to the invention of the Voltaic pile, or battery. This year is the 150th anniversary of a monumental
paper by Helmholtz in which he developed the laws governing the spread of currents in a volume conductor prompted
by his study of nerve propagation.
Bioelectric signals are small and brief, and their measurement has provided a major challenge. The 19th and 20th
centuries saw the development of more and more sophisticated insturmentation. This year marks the 101st anniversary
of the invention of the string galvanometer by Einthoven. This device made clinical electrocardiography feasible.
Einthoven presented a concept in which the heart was represented as a vector, and voltages measured across the
limbs were the dot product of the heart vector and a "lead vector." Following World War II several biophysicists
and electrical engineers put this concept on a firm physical basis. The introduction of the digital computer led
to the use of sophisticated models for the propagation of the electric impulse in the heart and the resulting potentials
on the skin. Interest in studies of firing patterns of neurons led to the invention of the personal computer.
Friday, September 19, 12:00 - 1:00 pm, Room 210 Hallowell at University Park and CG623 in Hershey
NOVEL TECHNIQUES FOR MINIMIZING ADVERSE OUTCOMES AFTER PEDIATRIC CARDIOPULMONARY BYPASS PROCEDURES: PULSATILE FLOW
AND COMPLEMENT INHIBITION
Akif Undar
Hershey Medical Center
Abstract
Pulsatile vs. Non-Pulsatile Perfusion
Congenital heart anomalies are the most common birth defects in the United States. Each year, approximately
10,000 babies (1 of every 115), born in the United States have a congenital heart defect that will require operative
repair with the aid of cardiopulmonary bypass (CPB) during early childhood. Approximately 22,000 children (aged
1 day to 18 years) underwent procedures that required CPB in 2002.
Injury to vital organs (brain, heart, and kidneys) following cardiopulmonary bypass with conventional non-pulsatile
perfusion is still a significant clinical problem in pediatric cardiac surgery patients. Several investigators
have shown that the use of pulsatile perfusion improves vital organ recovery in neonates and infants, as well as
in animal models, compared to the conventional non-pulsatile flow during and after open-heart surgery. To date,
however, only a few pediatric heart centers routinely use pulsatile flow during cardiac surgery. In this lecture,
the effects of different types of pulsatile pumps (pulsatile roller, hydraulically driven physiologic pulsatile
pumps) versus conventional non-pulsatile roller pumps on vital organ blood flow will be discussed in detail. Specifically,
I will talk about the benefits of pulsatile perfusion on cerebral, renal, and myocardial blood flow. The impact
of pediatric membrane oxygenators and aortic cannulas on perfusion mode (pulsatile vs. non-pulsatile) will be included.
The Energy Equivalent Pressure formula will also be described for use in the quantification of the pulsatile and
non-pulsatile pressure and flow waveforms for direct comparison.
Complement Inhibition
Patients undergoing cardiopulmonary bypass (CPB) frequently manifest generalized systemic inflammation and
occasionally manifest serious multi-organ failure. Inflammatory responses of CPB are triggered by contact of blood
with artificial surfaces of the bypass circuits, surgical trauma, and ischemia-reperfusion injury. We studied the
effects of specific inhibition of the alternative complement cascade by using an anti-factor D monoclonal antibody
(166-32) in extracorporeal circulation of human whole blood used as a simulated model of cardiopulmonary bypass
and in baboons.
Monoclonal antibody inhibited the alternative complement activation, the production of Bb, C3a, sC5b-9, C5a, upregulation
of CD11b on neutrophils, and CD62P on platelets in a simulated model of CPB. Anti-factor D Mab almost completely
inhibited plasma Bb, C3a, and sC5b-9 production during CPB, CD11b expression on neutrophils, and on monocytes was
lower in the treatment group in baboons.
The alternative complement pathway plays a major role in systemic inflammation during CPB. Inhibition of complement
activation via the alternative pathway by antifactor D Mab 166-32 significantly reduces leukocyte activation and
tissue injury in our baboon model.
Friday, September 26, 12:00 - 1:00 pm, Room 210 Hallowell at University Park and CG623 at Hershey
Ultrasound Energy-Based Devices for Surgical Applications
Inder Makin
Ethicon Endo-Surgery, Johnson & Johnson
Abstract
In its most common embodiment, ultrasound energy modality in medicine is used in the form of sonographic imaging
equipment. These systems today are considered standard imaging equipment in almost every medical center and are
used by professionals from multiple medical specialties. However, new areas of medical application of ultrasound
energy are emerging, whereby ultrasound energy deposited at powers much greater than that used for imaging result
in significant tissue effects (such as coagulative necrosis, hemostasis, and cutting), in situ. The use of intense
ultrasound for therapeutic (surgical) applications can further be segmented into systems operating at kHz frequencies
(17 - 100 kHz), and MHz frequency (1 - 10 MHz) systems.
This presentation will describe as an example, a 55 kHz clamp-based surgical device, used extensively by surgeons
to cut and coagulate tissue. An "engineer's approach" to understanding the basic biophysical model will
be presented. Comparison with experimental observations will be considered. Further, the role of therapeutic MHz
frequency ultrasound systems will be discussed. Considerations for designing MHz-frequency surgical systems through
numerical simulations and experiments will be presented.
Friday, October 3, 12:00 - 1:00 pm, Room 210 Hallowell at University Park and CG623 at Hershey
New Tools and Approaches to Understand, Manipulate, and Control Cytoskeletal Machinery
Alan Hunt
University of Michigan
Abstract
We are at the beginning of a revolution in our ability to control and study cellular and molecular processes using high precision quantitative techniques. This is a logical progression from the revolution in molecular biology that has dominated the life sciences for twenty years. Many biological molecules have been identified, and molecular genetics has provided evidence of their functions, but applying this knowledge to its full potential now requires quantitative perspective and methodologies to fully characterize mechanical, physical, and structural properties. With mitosis as our primary focus, our lab is developing methods to precisely manipulate subcellular structures using optical forces, lithographic protein patterning, and "structural-knockout" technology whereby sub-cellular structures can be targeted in space and quickly ablated using tightly focused ultrashort laser pulses. We are also investigating application of these techniques to develop bio-hybrid MEMS devices that are actuated by biological motor molecules.
Friday, October 10, 12:00 - 1:00 pm, Room 210 Hallowell at University Park and CG623 at Hershey
Quantifying Hydrodynamic Forces on Intercellular Bonds, Soluble Molecules and Cell Surface Receptors: Some Examples
Involving Human Leukocytes, Platelets and Plasma Protein von Willebrand Factor
Sriram Neelamegham
Chemical Engineering Department
State University of New York at Buffalo
Abstract
Cells and biomolecules exposed to blood circulation experience hydrodynamic forces that affect their function. Application of biophysical modeling strategies to analyze in vitro experiments provides an important method to distinguish between the contributions of the biological features of cells/biomolecules and the physical effects of mechanical shear flow in controlling cell function. This talk presents a description of some of these computational methods, which we have developed over the last several years to analyze cell adhesion and activation data. Some of the focus is on the application of low Reynolds number hydrodynamic theory to estimate the nature of fluid forces and force loading rates applied on cellular aggregates, cell-surface proteins and soluble molecules. Other aspects of the talk deal with methods to quantify adhesion efficiencies under fluid shear in two geometries, where cells collide and adhere in suspension and when cells attach to ligand-bearing substrates. For illustrative purposes, we will discuss the relative roles of selectins and ß2-integrins in mediating leukocyte adhesion in suspension and onto substrates. In another example, we will quantitatively examine the aspects of hydrodynamic shear that regulate shear-induced platelet activation and self-association of von Willebrand factor (vWF). In this last example, it is demonstrated that platelet activation under shear flow follows a two-step process that can be resolved based on the contributions of vWF, platelet surface receptor GpIb and fluid forces. In these experiments using light and neutron scattering, it is also demonstrated that human vWF when subjected to shear forces aggregate or undergoes self-association. This vWF self-association may be an additional feature involved in controlling cell adhesion rates in the arterial circulation.
Friday, October 17, 12:00 - 1:00 pm, Room 210 Hallowell at University Park and CG623 at Hershey
No Seminar
Friday, October 24, 12:00 - 1:00 pm, Room 210 Hallowell at University Park and CG623 at Hershey
How Heart Valves Accumulate Cholesterol
David Rumschitzki
Department of Chemical Engineering
City College, NY
Abstract
Atherosclerosis is the leading cause of death in the US and in all western countries. In addition to accumulating in walls of large arteries, low-density lipoprotein (LDL or "bad") cholesterol also infiltrates and accumulates in valve leaflets, mainly on the left side of the heart. Progression to calcification can impair valve function.
Tompkins and coworkers (1989) used quantitative autoradiography to measure concentration profiles of labeled
LDL as a function of depth in aortic valve leaflets of squirrel monkeys after 30 minutes of circulation. These
profiles show large variation in both magnitude and shape, even between different leaflets of the same aortic valve
and between different regions of the same valve leaflet, despite identical experimental conditions. Naturally,
the parameters that they determine by fitting each set of these data to a simple one-dimensional diffusion model
show the same variability. Using insight gained from our work on LD transport in the walls of large arteries, we
construct a theory that can rationally explain the variability in these data as well as other transport data on
valves so as to understand the early stages of valve disease.
We present experimental evidence that macromolecular transport into the valve leaflet is focal rather than uniform.
We show that after penetrating the valve's endothelium at a leaky site, tracer rapidly spreads to form a large
spot and we explain what this implies about the subendothelial structure and the role of convection. Based on these
findings, we propose a two-dimensional, convection-diffusion model and fit three parameters that cannot be determined
directly by comparison with one of Tompkins et al's curves. With all parameters fixed, we quantitatively explain
tracer spot growth rate. Numerical solutions of this same model account for the variability in the balance of Tompkins
et al's squirrel monkey data by allowing the number of leakage foci on each endothelium to vary and by varying
the distance from the nearest leak to the tissue section. Combination of the LDL transport model with our earlier
kinetics of lipid binding to extracellular matrix, with no new adjustable parameters, agrees well with data on
lipid accumulation in the valves of rabbits fed a high cholesterol diet.
Friday, October 31, 12:00 - 1:00 pm, Room 210 Hallowell at University Park and CG623 at Hershey
NO Seminar
Friday, November 7, 12:00 - 1:00 pm, Room 210 Hallowell at University Park and CG623 at Hershey
Friday, November 14, 12:00 - 1:00 pm, Room 210 Hallowell at University Park and CG623 at Hershey
MEMS-Based Fluidic Manipulation: Design, Modeling, and Experimentation
Qiao Lin
Department of Mechanical Engineering
Carnegie Mellon University
Abstract
Microelectromechanical systems (MEMS) are enabling remarkable advances in a broad spectrum of technical fields. Emphasizing the application of MEMS technology to fluidic manipulation, we will present two aspects of our work in microfluidic design, modeling, and experimentation. The first aspect involves the development of efficient and accurate models for microfluidic systems, with a particular focus on the modeling of analyte dispersion in microchip electrophoretic separations. These models are in closed form and parametrized, and are capable of accurately capturing the essential physics of dispersion to allow fast prediction and evaluation of the impact of microchip design parameter choices on separation performance. As such, these models are useful as building blocks for efficient, higher-level models to represent complex bioanalytical microsystems. We will also discuss models that follow a similar philosophical approach to address MEMS hotwire-type thermal sensors in microfluidic devices. The second aspect of this talk involves the experimentation and analysis of passive flow control in microfluidic systems. We will focus on polymeric micro devices that exploit large compliance of polydimethylsiloxane (PDMS) microstructures, and present our results in device fabrication, modeling and characterization. As valves, these devices can be used to gate liquid flows with low leak rates. As novel flow regulators, these devices, in analogy to electronic constant-current sources, provide virtually constant flow rates under large variations of driving pressures. The passiveness of device operation eliminates the need for power or active control components, and the planar device configuration is amenable to the incorporation of these devices in integrated microfluidic systems.
Friday, November 21, 12:00 - 1:00 pm, Room 210 Hallowell at University Park and CG623 at Hershey
Imaging Lungs with NMR
Dean Kuethe
New Mexico Resonance
Abstract
Lungs are considered very difficult to image with nuclear magnetic resonance imaging (NMRI). It is generally thought high resolution lung MRI is impossible. The canonical reasons are that, compared to other organs, there is very little water in lungs to image and that lungs mess up the nice homogeneous magnetic field we impose to perform MRI. It turns out that there are good ways to make high resolution images of lung tissue but the techniques are not yet available on clinical imagers. Additionally, a number of folk are making images of gases in lungs and making low resolution images of lung tissue that contain useful physiological information. I will show you recent results of high resolution images of rat lungs and images of sulfurhexafluoride gas (SF6) in rat lungs. The chemical name sounds scary, but it is harmless except that it makes one’s voice sound like a heavy metal rock singer.
Friday, November 28, 12:00 - 1:00 pm, Room 210 Hallowell at University Park and CG623 at Hershey
NO SEMINAR THANKSGIVING BREAK
Friday, December 5, 12:00 - 1:00 pm, Room 210 Hallowell at University Park and CG623 at Hershey
Student Presentations:
Osama Al-Bataineh
Optimized hyperthermia treatment of prostate cancer using an intracavitary ultrasound
array
Abstract
Localized uniformly distributed ultrasound-induced hyperthermia is a useful adjuvant to radiotherapy in the
treatment of prostate cancer. A two-dimensional, 20 x 4 element, transrectal phased-array probe was designed to
deliver a uniform and controllable amount of heat directly to the prostate without damaging the rectal wall or
surrounding tissue. A three-dimensional prostate model was created using anatomical markers from the Visible Human
Project to optimize the array. Sound speed, density, and absorption parameters were mapped to hue, saturation and
value of the photographic data to simulate sound speed propagation through inhomogeneous tissue using the k-space
method. To satisfy the requirements of this method from 1.2 to 1.8 MHz, the grid was adjusted to have 5 points
per millimeter in each Cartesian direction. A spherical wave was propagated through the model using tapered absorption
boundary conditions. The expected temperature rise due to sound was obtained using the bioheat transfer equation.
Optimal insonification parameters that uniformly heat the prostate to 43°C for 40-60 minutes will be determined
for use in the construction of a clinical hyperthermia array. [Research supported by the Department of Defense
Congressionally Directed Medical Prostate Cancer Research Program.] ===================================================================================
Min-Ho Kim
Fluid Filtration Rate Transient Following Capillary Occlusion
Abstract
Using a modified Landis technique, micropipette occlusion of microvessels allows a measure of fluid filtration
rate (Jv/S). In capillaries of the rat mesentery, we observe a transient decline in which Jv/S at 50 sec of occlusion
is only 55 +/- 8% of the value obtained at 5 sec. One hypothesis is that this decline in Jv/S is due to the step
reduction in shear rate (to zero) following
occlusion, i.e., hydraulic conductivity (Lp) could be dependent on shear rate, and therefore decreases during occlusion.
A second hypothesis is that the step increase in capillary pressure during the occlusion induces a pressure “sealing
effect” that decreases Lp. In 12 rats, baseline Jv/S was measured, then prior to a second capillary occlusion,
we increased baseline capillary shear rate by 99 +/- 9% by occluding the feeding arteriole just downstream of the
capillary
branch. This technique also increases arteriolar pressure by 17 +/-2%. With this increase in shear and pressure,
Jv/S increased by 92 +/- 6% in the intial 5 sec of measurement compared to baseline, but only by 21 +/- 4% by 50
sec. With shear rates equal to zero during capillary occlusion, the 21% difference at 50 sec appears to correspond
to the sustained pressure difference due to downstream arteriolar occlusion. We hypothesize that the larger change
in shear may be a more important factor than the smaller change in pressure in influencing the initial decline
in Jv/S following occlusion.
=====================================================
Huninan Liang
Highly Parallel Drug Screening in Integrated Bio-Optical Sensors
Abstract
A major research thrust in nanoscale science and technology is the development of autonomous platforms for cellular
analysis. An array of conventional systems already exist, and the pharmaceutical industry uses complicated automated
robotic systems and cell based assays to test potential therapeutic compounds.
Since these systems typically screen hundreds of thousands of compounds to develop a single commercial drug, there
are currently major research efforts to miniaturize these automated platforms. Batch fabricated disposable diagnostic
units hold great potential to enable both research and healthcare advances when applied to high throughput drug
screening. These biomedical applications of MicroElectroMechanicalSystems are an emerging field for sensing and
manipulating biological material with a precision not previously possible. The scope of this research is to develop
a multidisciplinary approach to investigate highly-parallel, high-throughput drug screening using fluorescence-based
cellular assays in a programmable, parallel, microfluidic cytometer (PPMC).
====================================================
Wanzhan Liu
Numerical Calculations of Radiofrequency Power Absorbed by Different Subjects in Magnetic Resonance
Imaging
Abstract
RF power absorbed by the subject's body during MRI are calculated for a male and a female subject with finite
difference time domain (FDTD)method. The male subject absorbs more RF power than the female subject does. The absorbed
RF
power in both subjects are below the limits of regulation.
Friday, December 12, 12:00 - 1:00 pm, Room 210 Hallowell at University Park and CG623 at Hershey
Student Presentations:
Shile Liang
Effects of Shear Rate Versus Shear Stress on PMN-Mediated Melanoma Adhsion and Migration
Abstract
The primary cause of death among cancer patients is cancer metastasis. The process of tumor metastasis consists
of a complex cascade of adhesive interactions between tumor cells and host tissues. Metastatic cancer cells must
penetrate the interendothelial junctions to invade the underlying tissue. The initial step in melanoma extravasation
is tumor cell adhesion to the endothelium of blood capillaries. Recent data have expanded the concept that inflammation
is a critical component of tumor progression and many cancers arise from sites of infection and inflammation .
Neutrophils (PMNs) are the first recruited effectors of the acute inflammatory response . Therefore, we hypothesize
that PMNs potentially influence melanoma adhesion by directly tethering melanoma cells to endothelium cells (EC)
and facilitate the binding of melanomas to activated ECs and contribute to conditions favorable for melanoma cell
extravasation. Our lab has recently reported that PMNs enhance tumor cell adhesion and subsequent migration toward
extracellular matrix proteins under shear conditions . Of interest is that how these two cell types aggregate under
shear force and what hydrodynamic parameters modulate the heterotypic cell-cell aggregation. We design experiments
to characterize the effects of shear rate and fluid viscosity on melanoma-PMN aggregation by separately varying
g (with a fixed µ) or varying µ (with a fixed g).
A parallel-plate chamber is used to quantify PMN-melanoma adhesion behavior and the heterotypic cell-cell adhesion
efficiency and collision frequency. Next, using a flow-migration chamber to characterize PMN-mediated melanoma
transmigration at the same set of flow parameters, we can characterize the effectiveness of PMNs in modulating
melanoma extravasation.
===================================================================================
Varun Reddy
Organic/aqueous Two Phase Microflow for Biological Sample Preparation
Abstract
Genomic or plasmid DNA extraction using aqueous-organic liquid extraction is one of many standard techniques
commonly performed in biology laboratories. Briefly, the procedure consists of lysing cells in a protease solution,
and adding a phenol:chloroform 1:1 by volume mixture to the aqueous solution. A vortexing step mixes the two phases
and allows the different cellular components to partition into either the aqueous or organic phases. With effective
mixing, the cell
components naturally distribute themselves into the two phases in order to minimize interaction energies of the
biological components with the surrounding solvents. The cell membrane components and protein partition into the
organic phase while the DNA stays in the aqueous phase. This partitioning occurs over molecular dimensions at the
aqueous-organic interface. Thus, effective mixing maximizes the surface area over which this partitioning occurs.
Therefore, the smaller the discrete phase domains, the more effective the DNA extraction procedure is. After the
mixing step, the two immiscible phases are allowed to separate and the aqueous phase is removed using a micropipette.
The DNA is then concentrated by precipitation in ethanol and resuspended in an aqueous buffer. This technique has
only been useful for a large number of cells and uses a large volume of liquid (~ 1 ml) because of DNA loss during
the aqueous phase removal and the size of micropipette tools.This work studies two phase fluidic interactions to
promote the mixing of two fluid phases consisting of an aqueous and organic phase for a miniaturized organic/aqueous
liquid extraction mTAS(micro total analysis system) device for DNA processing from whole cell samples. A microfluidic
device fabricated by a combination of standard lithographic techniques such as glass etching, sputtering and lift-off
processes is used to extract the DNA. Analogus to the mixing mentioned earlier the mixing of the two liquid phases
in the microfluidic device is brought about by a combination of hydrostatic pressure differences between the two
phases, unsteady flow and electrohydrodynamic instability. The device once fabricated will be packaged in a external
housing.
Hydrostatic flow control in combination with electrically promoted mixing should promote efficient liquid extraction.
======================================================================================
Ben Snyder
Design, Development, and Implementation of the 3x3 Cymbal Array for
Insulin Delivery
Abstract
Previous research, conducted both within this Bioengineering Department and other external laboratories, has
demonstrated the feasibility of using ultrasound to transdermally deliver drugs - scientifically known as sonophoresis
or phonophoresis. Previous studies have utilized either large, relatively immobile sonicators or contrastingly
undersized arrays too small for human use. The 3x3 Cymbal array, operating at 20 kHz frequency and measuring 57x57x7
mm3 (35 grams), is modeled after the smaller 2x2 Cymbal array prototype. The proposed advantage of the 3x3 Cymbal
array is an increased cross-sectional area of ultrasound power output, thereby making the array practical for human
use. This proposal is verified using ultrasound exposimetry, including an omnidirectional reference hydrophone
and a precise, computer controlled positioning system. Spatial peak-temporal peak power output intensities were
obtained from both the 2x2 and 3x3 Cymbal arrays, and then compared. From the exposimetry experiments, correct
output parameters were determined so that the spatial peak-temporal peak intensity was 100 mW/cm2 at a distance
of 1 mm from the array surface. This power intensity was then employed to deliver insulin to hyperglycemic rabbits,
with an ultrasound exposure time of 60 minutes. Blood glucose levels were normalized between a test sample of rabbits
and compared to the results obtained using the 2x2 Cymbal array.
========================================================================================
Ming Yang
Design and Fabrication of Multifunctional Ophthalmologic Instrument Prototype
Abstract
New surgical instruments are currently being developed that function as multiple instruments in one. Use of multifunctional instruments would provide a major benefit in vitreo-retinal surgery because it would minimize the time required to bring the instrument function to the surgical site. Disposable multifunctional instruments would provide an additional benefit because of frequent instrument damage and the need for meticulous cleaning. Furthermore, the industry is moving toward even smaller instruments. In this work, a new multifunctional ophthalmologic instrument prototype was designed and fabricated using UV-LIGA microfabrication techniques. Electroless plated nickel was used as material of the instrument. The mechanical property of the prototype was tested and the materials property was investigated using SEM. In ongoing work, a new prototype was designed for better mechanical function. Features for Self-alignment package and in situ mechanical test were also incorporated into this design.
For additional information, contact Ms. Doretta Garvey, Dept of Bioengineering, Tel: 814.865.1407 or E-Mail: bioe@psu.edu