BEE 4530 - 2015 Student Papers

Permanent URI for this collection

This is a collection of student research papers for Professor Ashim Datta's Biomed BEE 4530/Computer-aided Engineering course for 2015.


Recent Submissions

Now showing 1 - 10 of 10
  • Item
    Optimization of Combined Radiation and Gold Nanoparticle Hyperthermia Therapy for Treating Cutaneous Squamous Carcinoma
    Sankar, Sitara; Zhang, Michelle (2015-05-19)
    Radiation therapy offers the ability to kill and shrink tumors non-invasively and serves as adjuvant therapy post-surgery and as primary therapy for patients unable to undergo surgery. But the process is nonspecific and nearby healthy tissue exposed to the radiation can also undergo the same ablation process that kills tumor cells. A possible solution is coupling the radiation therapy with directed hyperthermia. Tumors exposed to elevated temperatures of 40-45C during hyperthermia exhibit increased sensitivity to radiation therapy and are more likely undergo ablation. The heating process of the tissue can be controlled by directly injecting the tumor with gold nanoparticles (AuNPs), which can absorb near infrared light and heat the tumor faster. This project models the treatment of subcutaneous squamous carcinoma using AuNP controlled hyperthermia and X-ray radiation therapy. Using COMSOL Multiphysics 4.3b using 2D axisymmetric coordinates, we simplified the geometry of the tumor and surrounding tissue. We modeled two overall processes vital to the treatment: diffusion of AuNPs injected into the tumor and the heating process of hyperthermiaradiation therapy. We optimized the heating and radiation dosage combination for maximized tumor death and minimized tissue damage. Survival fraction of tumor and tissue were evaluated using a linear quadratic model of radiation dosage modified for thermal treatments. Our model shows the results for AuNPs diffusion and hyperthermia-radiation tissue ablation. The optimized combination of hyperthermia and X-ray dosage was determined to be 60 seconds of heating using a 1.5 W/cm2 infrared lamp and 0.35 Gy. This hyperthermia and radiation dosage model takes advantage of AuNP’s ability to increase tumor sensitivity to radiation therapy. By optimizing the heating time and radiation dosage combination, we can reduce the total radiation exposure and the length of treatment, allowing for overall faster and less harmful treatments for patients.
  • Item
    Optimization of Transdermal Drug Delivery by Hydrogel-Enhanced Sonophoresis
    Chapin, Nate; Hung, Michelle; Larsen, Wesley; Pramil, Varsha (2015-05-19)
    External-field mediated transdermal drug delivery is a new alternative to oral delivery and hypodermic injections. This method allows patients to receive treatments via a drug-infused patch applied to the skin and offers continuous release of drug for up to a week. It is also relatively inexpensive compared to regular treatments. In our design, ultrasound and an applied electric field are used to increase the rate of drug diffusion and propagation from the patch into and across the skin. Previous research has shown that the application of ultrasound results in higher rates of transdermal transport by increasing the permeability of the stratum corneum. This permeabilization can be so great that the patch or drug reservoir itself becomes the most significant barrier to the overall transdermal drug transport. An applied voltage has been shown to be capable of creating a driving force for transport that counteracts this effect. In this project, we combined the effects of applied voltage and ultrasound in order to model the transdermal transport of insulin from a methyl-cellulose hydrogel through the skin. Because the typical drug-delivery patch is circular, we modeled the problem in a 2Daxisymmetric geometry. Next, we determined the effects of intensity and voltage on the average insulin concentration in the skin. To do this, we conducted a sensitivity analysis by varying the applied voltages and intensities on the hydrogel and calculating the corresponding average insulin concentration in the skin at a specific time point. Finally, by using an objective function, we maximized the flux of insulin through the skin while minimizing patient discomfort. The iontophoretic and sonophoretic aspects of our model were validated individually through comparison with experimental data. We found that voltage and intensity from ultrasound combined provide the greatest increase in insulin transport through the skin. An intensity of 1 kW/m2 in conjunction with an applied voltage of -2 V resulted in the optimum insulin flux through the skin while maintaining minimal patient discomfort. The optimal intensity was found to be at the lower end of the range of experimentally and clinically utilized values. This suggests that higher intensities may contribute unnecessary heating without significantly enhancing insulin transport through the skin. Based on our optimization results, it can be seen that transdermal delivery of insulin through the skin is efficient when coupled with ultrasound and applied voltage. Our results show that with the application of sonophoresis and iontophoresis, insulin is effectively able to diffuse through the skin into the bloodstream. Our optimization also shows that this type of insulin delivery would cause minimal discomfort or skin damage. This suggests that transdermal delivery of insulin through the skin is a promising treatment for patients with diabetes.
  • Item
    A Computational Model of Vortex Ring Formation in Hydrogels
    Das, Shoshana; Fiore, Alyssa; Hickman, Dillon; Vasthare, Varsha (2015-05-19)
    Duo An, a PhD student at Cornell University, is developing a method that uses electrospray technique to expel clay hydrogel droplets into an ion solution. Upon contact, the droplets crosslink and form various solid geometries, including spheres and toroids. Depending on a few factors, including composition of the hydrogel, volume, viscosity, and speed of falling, the hydrogel will form a variety shapes. Toroidal shapes have many applications because they have a greater surface area to volume ratio than spheres. Therefore, clay hydrogels in the shape of a torus have a greater diffusive capacity than spherically shaped hydrogel formations. This allows for greater mass transfer and thus more efficient drug treatment therapies. The current problem is that the specific conditions that cause micro-toroid formation are not well understood. The majority of An’s data has used a larger scale droplet size diameter. This project aimed to elucidate the formation of a toroid shape by creating a COMSOL® simulation that can model the fluid flow and physics on a smaller scale. A pseudo three-phase model was designed to represent the three parts: the droplet of hydrogel, the air, and the ion solution. However, the model had to be implemented as a two-phase problem, due to the three-phase restraints in COMSOL. This was accomplished by implementing the solution and hydrogel as a single phase by defining an initial ion concentration for both the hydrogel and water bath. Additionally, the particle-tracing feature was used in COMSOL to help track the hydrogel species during the simulation. A larger scale model with a diameter of 3.0 mm was first developed to demonstrate toroid formation on a larger scale. Performing a sensitivity analysis on particle number, density, diameter and initial velocity showed that initial velocity had the largest impact on formation of the final product. This reflects An’s findings that varying initial velocities of the hydrogel droplet leads to different spreading and shape formation. In order to further evaluate the model, simulations with different initial velocities were run and their shape formations were qualitatively compared to images and descriptions from An's data. Success in obtaining vortex formation at the large scale indicated that a smaller scale model could be useful in predicting and understanding micro-toroid formation of these hydrogels. Next, the model was scaled down to a 500 μm diameter droplet and the results were compared to the larger droplet. When compared, the small model had similar vorticity trends to the large model at a faster rate. In order to compare the model’s results to An’s, the Reynolds and Weber numbers were calculated and final shape formation images were compared with experimental images for models that matched these numbers. Two of the models showed a similar mushroom shape formation to experimental data. The use of COMSOL to create a scaled down model of An's work helps to provide a better understanding of the underlying physics behind toroid formation of these clay hydrogels. This understanding will lead to better control of the process, in order to produce hydrogels of desired geometries and dimensions, which could give promise for production in future biomedical applications.
  • Item
    Predicting Embryonic Dynamics and Development
    Francken, Miles; Patish, Levi; Streitman, David; Yoshitake, Sho (2015-05-19)
    Although the chick embryo has long been used as a model for mammalian embryogenesis, there is a need to understand the changing tissue properties that cause physiological changes. Most imaging techniques have proven inadequate for real-time in vivo imaging of soft tissues. However, Micro-CT technology with gold nanoparticles as contrast agents can be used as an alternative. These nanoparticles readily dissolve into soft tissues and thus mitigate the need to fix samples (Murphy, 2008). Using X-rays to excite these nanoparticles produces 2-D image slices, which can then be combined to create a complete 3-D image. In this study, this imaging technique was coupled with computational modeling to determine the dynamics of tissue properties in chick embryos. Gold nanoparticles were injected into chick embryos, and Micro-CT images were taken at various time points within 24 hours following injection. A novel image analysis technique was then developed and used to quantify the accumulation of the nanoparticles in the myocardial tissue directly from the images. The heart and surrounding pericardial cavity were isolated and approximated as a sphere with a semi-aqueous layer surrounding the sphere, respectively. Using this geometry, a 1-D radially axisymmetric diffusion model was created in COMSOL. A function for diffusivity of nanoparticle in myocardial tissue as well as partition coefficient in the heart lumen were found by minimizing the discrepancy between the model and experimental values. Although the experimental results followed expectations, the model did not match the results. As expected, there was accumulation of nanoparticle in the myocardial tissue, and the rate of this accumulation decreased with time. Optimization of the model yielded a diffusivity function and partition coefficient of xxxx=1.0∗10−14xx−10−7xx and 80, respectively. But, these values corresponded to a total discrepancy of 29.94%. This difference could be attributed to variability in the image processing methods or the computational model. However, analysis of the image processing methods showed robustness and reliability. When three group members independently analyzed ten images, the difference in total mass was less than 5%. Consequently, the contribution of each parameter on the discrepancy was resolved. This was done by performing sensitivity analyses on the parameters. From this, it was clear that, for the diffusivity function of the nanoparticle in the myocardial tissue, the initial diffusivity had a much larger impact on the total discrepancy than the rate of decay. Furthermore, the partition coefficient had a large impact on the total discrepancy. However, the total discrepancy did not differ greatly for values similar to the optimal values. From this, it was evident that changing these parameters would not lead to a lower difference; the study needed to be improved. Possible enhancements include including more experimental data and connecting the heart to other organs in the model, among others.
  • Item
    That bites!: the transport of Bothrops asper venom in leg
    Hatter, Nick; MacMullan, Melanie; Kwasnica, Marek; Hu, Victoria (2015-05-19)
    Snake venom has been known to be a deadly toxin for centuries. Although many studies about the dangerous effects of snake venom have been conducted, the spread of venom throughout human tissue has not yet been modeled. The goal of this study is to examine the contributions of relevant modes of mass transport on the spread of venom in human skeletal muscle tissue. In this study, computational models that mimic the localized propagation of venom were developed using parameter values for diffusivity, injection pressure, and injection volume determined from available research papers, empirical formulas, and clinical case studies. COMSOL Multiphysics 4.3b software was used to simulate the dissemination of venom from a fang into human flesh. A simplified 2D axisymmetric geometry was used initially to model the human tissue punctured by a single snake fang. This model allows us to examine the diffusion of BAP1 metalloprotease in human tissue. The results were compared to data from in vivo snake bite case studies and a study of the diffusion of similarly sized non-toxic protein. It was determined that diffusion alone could not be responsible for the expected extent of venom spread. Therefore, additional modes of mass transport, such as convection and Darcy’s flow, were evaluated and integrated in a 3D model of a human leg. Darcy flow at the point of injection simulated the effect of the injection pressure pushing venom out of the fang into the tissue. The model results were compared to findings in a series of studies on venom mechanics and metering in rattlesnakes and were found to be consistent. The convection of venom away from the injection site due to circulation of blood and interstitial fluid in the leg was modeled as a source term. Sensitivity analyses were performed in order to study how sensitive the solutions are to these input parameters. These models enable researchers to gain insight into how the different modes of mass transport influence the progression of edema. Knowing this can allow researchers to develop more effective treatment methods and antidotes for snakebites. Researchers can also modify these input parameters in order to model venom transport for different species of snakes, allowing them to tailor treatment methods for a variety of snakes.
  • Item
    Modeling Reverse Iontophoresis for Noninvasive Glycemic Monitoring
    Cheng, Stephanie; Kuehlert, Esra; Panchap, Latha; Young, Kathleen (2015-05-19)
    Current methods of glucose monitoring for diabetics are invasive and require blood to be drawn. Despite improved technologies for self-monitoring glucose, patient adherence for these methods is low due to the inconvenience and complexity of the regimen. Several iterations of noninvasive devices have been developed for monitoring glucose levels, but these devices still require direct blood glucose analysis for calibration purposes. Therefore, a current focus in diabetes research is developing standalone noninvasive methods that do not require direct measurements. Although still in the proof-of-concept phase, a promising alternative to current treatments is a temporary tattoo glucose sensor. This glucose monitoring system consists of a removable device in the form of a decal-style temporary tattoo with an electrochemical sensor. The sensor has two electrodes that apply a voltage to the skin that brings glucose to the surface, a concept known as reverse iontophoresis. In reverse iontophoresis, ions flow diffusively in the medium in the direction opposite of the current resulting from the applied voltage. In the glucose sensor, the movement of sodium ions towards the cathode creates an osmotic pressure gradient. This pressure gradient drives water to the cathode, bringing interstitial glucose along to the cathode. At the cathode, the glucose reacts with glucose oxidase, transducing the glucose concentration into a measurable visual signal. Research on non-invasive glucose monitoring is still very much in its nascent stage. This project seeks to enhance the current understanding of the coupled physics behind the glucose monitoring process, as this has yet to be established. There is little modeling on the physics of any glucose monitoring system, despite their existence for the past 20 years. Published research mainly focuses on direct glucose measurements or signal correlations rather than describing the physical processes involved. The model described in this report focuses on defining the coupling of the different physics in the glucose sensor, and does not simulate the visual signal for the glucose level reading. The main goal was to optimize the design of a proof-of-concept glucose sensor described by Bandodkar, et al. in 2015. We validated our model by comparing our extracted glucose values to experimental values reported by Ching, et al. in 2008. Our final extracted glucose values were consistently within the range of their trial results, suggesting that our model closely mimicked experimental results. From our sensitivity analysis, we found that our model was most sensitive to glucose diffusivity, initial glucose concentration, and the reactive flux. However, it was not very sensitive to the parameters that we wanted to use to optimize the design: voltage and inter-electrode distance. We determined that this was because diffusion through the skin occurred slower than the reactive flux boundary extracted glucose, creating a diffusive shell around the cathode. Further work on our model will involve understanding and optimizing the reactive flux boundary, such that we minimize the presence of the diffusive shell. Additionally, the model will incorporate the glucose oxidase-containing gel layer between the skin surface and electrodes and the conversion of the extracted glucose into a measurable signal for glucose monitoring.
  • Item
    Optimizing Throughput in a Fetal Cell Magnetophoretic Separator
    Flood, Zachary; Jasty, Naveen; Jiang, Chiyu; Park, Jino (2015-05-19)
    Magnetophoretic separation is a non-invasive method under development for the sampling of nucleated fetal red blood cells (NRBCs). By inducing a magnetic field gradient symmetrically about a microfluidic separatory channel, blood cells may be separated based on their inherent magnetic properties [1]. Diamagnetic red blood cells (RBCs) are magnetically attracted into lateral side channels, while others cells, e.g. white blood cells (WBCs), are paramagnetic and repelled into the center waste channel. We aimed to model the first phase of a complete lab-on-chip device for the separation of RBCs from maternal circulation in order to isolate NRBCs for fetal diagnostics. Simulation results from COMSOL implementation revealed that seven out of ten RBCs successfully traveled to the intended lateral channels while three out of ten traveled through the center channel. All ten WBCs ended up in the center channel as intended. The three RBCs were collected in the middle channel due to their starting position. At the center of the channel before the fork, the magnetophoretic force was very weak and close to zero because magnets from both sides of the channel apply a force on those cells. In order to improve the efficiency of the magnetophoretic separator, the geometry of the channel was changed. An initial bifurcation was implemented that caused unilateral application of the magnetophoretic force before the main channel re-converged. By doing so, the initial force balance was broken and separation effi ciency was improved from 70% to 90%. Additionally, in order to improve device throughput the inlet velocity was subject to optimization. Using an objective function penalizing loss of RBCs and WBC contamination, an optimum inlet velocity of 0:875 mm x s-1 was determined. These actions constituted steps in solving the speed and accuracy constraints that have prevented successful commercialization of such a device. In doing so, model validation provided the basis for future manufacturing and testing and satisfied our motivation in developing safer fetal diagnostics.
  • Item
    Melting fat: Modeling the Effects of RFAL using Invasix BodyTiteTM system
    Fung, Ashley; Hua, Kevin; Nadig, Malavika; Wan, Gary (2015-05-19)
    Radio-Frequency Assisted Liposuction (RFAL) is a recent technique that shows promise compared to traditional procedures in terms of safety, recovery times, and results. Currently, however, there lacks an accurate computational model that can be used by researchers to study the efficacy of this technology with optimal parameters. Here we focus on the use of a novel RFAL device (the Invasix BodyTite™) that consists of a RF-emitting, fat-aspirating probe and a grounding sensor. A 3D COMSOL model was implemented with the bio-heat equation coupled with a joule heating mechanism in order to simulate the temperature and fat aspiration profiles, allowing analysis of varying levels of output power and probe velocity. The model had mean dimensions of skin, adipose, and muscle thickness proportional to measurements of arms and thighs found in literature, while the material properties were collected as statistical mean from a compilation of online databases and literature sources. Verification was conducted throughout each step of the design process through temperature and heat source graphs, and a mesh convergence was reached at a quality of higher than 30,000 elements. The cumulative fat aspiration is calculated with a volume integration of nodes that reach above the fat melting temperature of 316 K. The total fat volumes extrapolated from our model, within the devices range of power, are compliant with clinically observed results from literature. Furthermore, a sensitivity analysis of key fat parameters showed density, heat capacity, and thermal conductivity to have the largest effect on cumulative fat aspiration. To reach the original objective of assisting researchers with a computational mode, an optimization of probe velocity to cumulative fat aspiration was conducted. The optimal probe velocity for maximum rate of fat aspiration was found to be 3 cm/s.
  • Item
    Modeling an Injection Profile of Nanoparticles to Optimize Tumor Treatment Time with Magnetic Hyperthermia
    Eagle, Sonja; Wadsworth, Samantha; Wnorowski, Alexa (2015-05-19)
    Hyperthermia treatment to destroy cancerous tissue is a highly effective treatment strategy for malignant tumors. The goal of hyperthermia treatment is to raise the tumor temperature high enough to kill cancerous cells while minimizing damage to normal surrounding tissue. This project focuses on optimizing the treatment time using iron oxide magnetic nanoparticles (MNPs) to induce hyperthermia in cancerous tumors. In this treatment, the MNPs are injected into the center of the tumor, and their movement through the tissue is modeled using pressure-driven Darcy flow and simple mass diffusion. The MNPs are activated by a magnetic coil surrounding the tissue that produces an AC magnetic field, and heat is produced due to friction between the nanoparticles as they change orientation with the alternating current. This friction is sufficient to produce hyperthermia. Because of the many parameters that can be changed in hyperthermia treatments, computational modeling of this process could provide a more efficient way of determining optimal treatments. However, most previous models do not model the injection and diffusion of nanoparticles, but rather have an exponential decay power equation as a heat source at the site of injection. To create a more accurate model, the injection process and mass diffusion of the nanoparticles can be modeled and coupled to the heating process through an electromagnetic heat source term. In this COMSOL model, a tumor was approximated as a sphere surrounded by a sphere of normal tissue. Nanoparticle heat production within the tumor during exposure to a magnetic field is proportional to the nanoparticle concentration, which can be determined from the diffusion model including Darcy fluid flow. The transient temperature profile of the tissue was then monitored to observe the extent of damage to both the tumor tissue and surrounding healthy tissue. Treatment time was then optimized for a specific initial nanoparticle fluid concentration and injection velocity. For a tumor with properties of a common liver tumor, nanoparticles with a concentration of 78600 g/m3 were injected at a flow rate of 20 μL/min for fifteen minutes and allowed to diffuse for 24 hours. Under these conditions, optimal heating time was determined to be 11.5 minutes. In the future, this model could be adjusted based on tumor size, geometry, and specific parameters such as density, as well as various types of nanoparticles, and used in a clinical setting to determine optimal treatment prior to beginning the hyperthermia treatment.
  • Item
    Optimization of reversible electroporation for the destruction of an irregular brain tumor
    Ahmad, Taha; Putrevu, Karann; Walk, Remy; Zhang, Xuexiang (2015-05-19)
    The ability to localize treatments is of great relevance to many diseases, from cancer to blood clots. The drugs used for treatment are often cytotoxic or otherwise deleterious to healthy as well as diseased tissue, and simply flooding the body is not an option. Conversely, the physiochemical properties of these drug molecules can often prevent their penetration of the cell membrane in even the targeted area. Reversible tissue electroporation serves as a method to bypass this barrier and introduce the drug into only targeted tissue. The instantaneous application of an electrical field (pulse) causes the transient formation of pores in the cell membrane that allow large molecules, including drugs, to pass into the cytosol. Such electroporation has found use clinically in electrochemotherapy, electrogenetherapy, and transdermal drug delivery. However, the introduction of reversible electroporation into body tissue can have unintended consequences that require consideration. Sensitive areas of the body, such as the brain, may be subject to overheating if the applied voltage is too high, and the electrodes must be oriented so as to minimize the penetrance of drug into healthy tissue. In vivo experimentation on animal models has several limitations, including imperfect correlation to human application, small sample sizes, and prohibitive costs. Thus predictive models are required prior to human clinical trials to ensure that minimal collateral damage is done to surrounding tissue. For this study, a two-dimensional model was developed in order to maximize death within a brain tumor while minimizing death in the surrounding healthy tissue. Given the complexity of the physics involved, the model was numerically implemented via the available software, COMSOL Multiphysics 4.3. A single pulse was applied to the domain and studied over five seconds. Pore distribution and pore size were linked to the strength of the electrical pulse implemented and the time post pulse. The reaction rate modeled the intake of drug (in this study, bleomycin) into the cell and was dependent on pore size, pore distribution, and local bleomycin concentration. Bleomycin was used for direct comparison with several other studies on electroporation that used the same agent. Using this model, it was found that for the chosen irregular tumor, thermal stress was of minimal concern as it failed to kill any cells not already killed by bleomycin. The optimization of electrode orientation, distancing and voltage application yielded a horizontal orientation with 4.8 mm between the electrodes and an applied voltage of 275 Volts, which killed 91.28% of the tumor. The application of larger voltages killed more of the tumor, but relatively more healthy cells. Sensitivity analysis on the voltage applied showed that the profile of electroporated cells followed the distribution of electric field lines, generally the 1x10^4 V/m line. It follows that there is a great deal of flexibility in the shape of the electroporated region that can be adjusted on a case-by-case basis. The goal of this model was to provide an accurate and clinically relevant simulation of reversible electroporation for tumor destruction. This is the first model that optimizes a two-electrode approach to reversible electroporation with an irregular tumor model and surrounding healthy tissue, and thus provides clinicians with finer control and understanding of this methodology for application. However further refined models must still be completed for increased predictive power; such refinements include a three dimensional model, multiple pulses, and modeling of intracranialpressure changes.