BEE 4530 - 2010 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 2010.


Recent Submissions

Now showing 1 - 10 of 11
  • Item
    Long Term Flux Profile of Implanon Birth Control Implant
    Aiwazian, Jonathan; Baker, Courtney; Nissanka, Nadee; Sarker, Upal (2010-05-21T18:50:55Z)
    Hormonal birth control methods have become increasingly popular since 2000, as the technology becomes more convenient for users, moving from daily pills, to weekly patches, to yearly implants. Implanon is an example of a long-term birth control implant. The goal of this project is to create an accurate computational model of Implanon’s hormone diffusion over its standard prescription length: three years. Because Implanon is intended for long-term use, any potential improvements in the drug or release mechanism take several years to clinically test. It is time and cost inefficient to develop several new designs and test them all over this long time period. Instead, we can eliminate those designs that fail to reach specifications in our computer model and be more confident in the clinical success of those that function properly in our computer model, thus reducing the number of clinical trials needed and the time and money spent. We created a two-dimensional cross section model of the Implanon implant and surrounding tissue under mass transfer conditions model using COMSOL Multiphysics software. We confirmed our model’s accuracy with comparisons to published Implanon hormone release rates at six weeks, one year, two years, and three years. Our model’s hormone release rate was found to stay within a factor of 10 of the published data at all critical time periods. This data is significant in that it has the potential to expedite the hormone modification process. There were several assumptions made in the model’s schematic design as well as material properties and boundary parameters. It is recommended that more in vivo experimentation and data gathering on Implanon implant placement and hormone diffusivity be conducted to improve this model’s accuracy.
  • Item
    Ex-Vivo Chemotherapeutic Drug Treatment of Human Tumor Spheroids
    Herynk, Brad; Liu, Vivian; Sabhaz, Jasmin; Starchenko, Alina (2010-05-21T18:48:22Z)
    Drug testing of microspheroid tumors ex-vivo has proven to mimic micrometastases in the bloodstream of cancer patients1. More recently, cancer treatment has turned to developing individual drug regimens that target specific tumor types with specific drug types and dosages for each patient2. This project models human cancer micrometastases as microspheroids immersed in fluid drug representative of adjuvant therapy. Computer generated results showed half maximal inhibitory concentration (IC50) times of approximately 5 days with initial drug concentration of 0.1 mol/m3 in the extracellular fluid. Data from an experiment performed with a HT-29 microspheroid tumor exposed to varying concentrations of 5’-flourouracil for varying amounts of time were used for comparison. The death concentrations and time were used to validate our computer simulation. Model results show tumor destruction by drug therapy consistent with current experimental results after one week’s worth of treatment. This preliminary model offers a safe and cost-effective alternative to medical testing for better understanding of dosing options for tumor destruction. Our work can now be applied for various tumors of different sizes and cell types to optimize tumor degradation for specific patient treatment options.
  • Item
    Optimization of Transdermal Drug Delivery with Microcapsules
    Jackman, Monica; Kim, Nala; Kim, Sung Min; Wong, Bethany (2010-05-21T18:46:16Z)
    Research in transdermal drug delivery systems has gained much attention in the past thirty years. One of the biggest challenges in developing an effective system however, is getting past the tightly-structured outermost layer of the skin called the stratum corneum. While many different techniques to safely bypass the stratum corneum have been employed, one promising method of transdermal drug delivery is the use of drug-loaded microcapsules. Encapsulating the drug, as opposed to a non-encapsulated topical application, allows the drug to diffuse into hair follicles where drug release can occur in the deeper layers of the skin. COMSOL was used to model the diffusion of drug through a hair follicle and into the skin layers, via microcapsules of inner radius 300nm and outer radius 350nm. The diffusion of microcapsules through the hair follicle, drug through the microcapsule shell, and drug through the skin layers was modeled using three transient diffusion equations that were solved simultaneously. The drug in microcapsules was modeled as emerging from a source, similar to a patch, above the hair follicle. The COMSOL model was verified by comparing the time for the microcapsules to penetrate into the follicle, the time for the drug to diffuse out of the microcapsules, and the drug concentration in the dermis to literature values. It took 60-70 minutes for the microcapsules to be evenly dispersed throughout the follicle, which is supported by literature values.1 The microcapsule released 75% of its encapsulated drug after 1.25 hours, which is comparable to a literature value of 70% release in 2 hours.1 Additionally, the same source showed that drug release by the microcapsule was complete after 4 hours; almost all of the drug had left the microcapsule in 4 hours in our model. The drug concentration in the dermis in our model after 4 hours (0.0005 ug/mL) was lower than literature sources. However, our design only had drug-containing microcapsules applied directly above the hair follicle, when in reality, much more drug would be applied to the skin surface.2 We showed that a slight variation of our design, using a thin layer of drug over the entire skin surface, greatly improves the value of drug concentration in the dermis. A sensitivity analysis was conducted to determine the significance of parameters on our model. The solution from our design led to a slightly high release rate of the drug and a low final drug concentration in the skin, relative to literature values. This indicated that our model would best reflect the behavior of a potent drug requiring a fast release rate and a low dosage in skin. In order to obtain a more accurate solution, the simplifications in our design could be adjusted and the realistic complexities of actual skin could be added. While we only studied diffusion through one hair follicle, studying several hair follicles in a representative region of the skin could lead to changes in the final drug concentration achieved in the skin. Despite the many simplifications of our model, we were able to show that this microcapsule transdermal system is an effective method for bypassing the highly impermeable outer skin layer.
  • Item
    Modeling Brain Cooling Helmets for Ischemia Patients
    Wiechecki, Julie; Ranganath, Neel; McDonough, Brendan; Wind, Jessica (2010-05-21T18:44:04Z)
    In this project, we modeled the effectiveness of a cooling cap designed to lower brain temperatures by approximately 3°C in order to cause temporary hypothermia in the brain and thereby prevent further injury caused by cerebral ischemia. Cerebral ischemia is a condition in which blood flows preferentially through certain blood vessels in the brain and not through others, resulting in certain sections of the brain receiving insufficient blood flow for nutrient uptake and waste removal, potentially resulting in a stroke. Using the modeling program COMSOL Multiphysics, we simulated the brain temperature that results from the use of a cooling helmet consisting of a cap containing flowing coolant. The model incorporates convective flow of the coolant in the cap, and heat conduction through various modeled layers of the head. Our model showed that cooling occurred by the predicted conduction and convection mechanism and our results matched closely with those obtained from the literature, therefore validating our model. Initial results showed appropriate brain cooling; however, damage to the scalp occurred. Time of application and temperature of coolant were then successfully optimized to eliminate scalp damage while maintaining effective brain cooling. Even cooling the brain just a few degrees, as achieved in our model, reduces the extent of brain damage following cerebral ischemia. The use of this model provides insight into the optimal treatment conditions for using the cooling cap in a clinical setting.
  • Item
    Optimizing Combined Laser Treatment for the Removal of Port Wine Stains and Cryogen Spray Cooling to Reduce Thermal Heating at the Skin Surface
    Agustin, Marissa; Jimenez, Rod Brian; Berglund, Caroline; Mookerjee, Adaleena (2010-05-21T18:41:45Z)
    Port wine stains (PWS) are birthmarks caused by the presence of dilated blood vessels, typically 15-55μm in diameter, located in the upper dermis of the skin. Currently, lasers in conjunction with cryogen cooling are the preferred treatment for PWS removal because they can selectively target PWS blood vessels while leaving the surrounding tissue unharmed. In this project, we compared the effectiveness of various combinations of laser, pre-heating and cooling methods for PWS removal. In COMSOL, we implemented pulsed dye laser heating using a finite element model of light diffusion coupled with heat transfer. Our geometry was based on a two-dimensional histological cross-section from a PWS punch biopsy in order to more accurately mimic the vascular anatomy of a PWS. We compared two types of cooling methods, cryogen spray cooling and water contact cooling. In addition, we implemented a preheating step to achieve higher temperatures in deeper blood vessels. We determined effectiveness of the treatment using an Arrhenius thermal damage equation to calculate injury values over the course of the treatment. For a single 2 ms laser pulse without cooling, the blood vessels reached a maximum temperature of 89°C. However, the skin surface temperature reached 65°C indicating that we would need to implement a cooling method in parallel with laser heating. We compared cryogen spray cooling with cold water therapy and found the cryogen spray to be more effective. Cryogen cooling for 100 ms before and during the laser treatment kept the post-laser epidermal temperature below 26°C, while water cooling only brought the post-laser temperature down to 34°C. Because cryogen was the more effective treatment, we used it as our preferred method of cooling for the remainder of our study. We then implemented this cooling method with a ten laser pulse treatment scheme, which elevated temperatures in the blood vessels but did not achieve coagulation temperatures in the deeper blood vessels. The addition of a 40 second preheating step at 60°C effectively increased the temperatures in the deeper blood vessels to desired levels, while keeping damage to the epidermal and dermal layers at a minimum. Our model of light transport and heat generation in the epidermis, dermis and blood vessels verified that it is possible to target blood vessels with laser therapy while inflicting minimal damage to surrounding tissue. Current methods for treating PWS using pulsed dye lasers are limited in that they cannot target blood vessels deeper in the tissue. In our design, we showed that including a preheating step and multiple laser pulses can effectively target blood vessels deeper in the dermis. Our model can be used to select process parameters and different treatment combinations prior to experiments and clinical trials. Simulating laser treatments as well as various pre-cooling and pre-heating methods in COMSOL reduces the need for excess experimentation and potentially decreases the time before new designs are approved for use.
  • Item
    Modeling the Flow and Diffusion of Lidocaine Through Tooth and Gum
    Bernardis, Sarah; Iyer, Shama; Munson, Amy; Pandya, Jui (2010-05-21T18:39:10Z)
    The movement of anesthesia around and through the tooth and gum was modeled in order to design a novel procedure to effectively anesthetize a single tooth with minimal side effects. This model includes the injection of lidocaine into the gum near the tooth, and the ensuing diffusion of anesthesia through the tooth and gum. The geometry of the tooth and gum has been drawn in two dimensions in COMSOL. The Brinkman’s equations were used to model the fluid flow resulting from the initial injection through the porous media, and the mass transfer equation was used to model how the anesthesia flows through the tooth and gum after the completion of the injection through the coupling of the velocities calculated by the Brinkman’s equations. The concentration of anesthesia in this region was calculated, in order to determine the amount of time the tooth is numb and the distance the anesthesia has travelled. The model showed that the optimum procedure for numbing the tooth has a long injection time of five minutes dispensing lidocaine at a slower velocity, specifically 0.018 mol/m3. This is an important process to model because with the longer injection time the patient’s discomfort can be decreased, numbness can be quickly achieved and anesthesia can last for the entirety of an average dental procedure protecting the patient from unnecessary pain.
  • Item
    Controlled release of ciprofloxacin from PLGA-coated contact lenses to treat eye infections
    Greenberg, Seth; Linderman, Stephen; Mogilevskaya, Yevgeniya; Weinschenk, Robert (2010-05-21T18:36:46Z)
    Topical ophthalmic solution, commonly known as eye drops, have long been the most widely-used method for ocular drug delivery. The drug delivery method, however, is characterized pulse delivery to the eye, which results in three consecutive stages: an initial period of overdose, a fairly short period of therapeutic concentration, and an extended period of sub-therapeutic concentration. This problem is further exacerbated by reflex tearing and blinking, which further dilutes and disperses the drop, causing merely 1%-7% of the drug delivered to be absorbed1,2. In recent years, researchers have proposed several designs of drug-eluting contact lenses to address this issue. However, failure to achieve sustained drug release through zero-order release kinetics remains a prevailing problem. An imperative design requirement in an effective drug-eluting contact lens is for the initial burst of release that is characteristic of all mass transfer be followed by a sustained period of nearly constant flux. This allows the drug to remain within a therapeutic concentration range in the eye for a prolonged period of time. Further, the lens must be biocompatible and safe for use. In this study, a contact lens was modeled based on the work of Ciolino et al. to release ciprofloxacin, a common antibiotic for treatment of infections caused by a variety of gram positive and gram negative bacterium, with zero-order release kinetics over many days. COMSOL Multiphysics software was employed to design and stimulate the model within the parameters of commercially available contact lenses. To control for zero-order release, a dual polymer system was used, with an inner polymer film containing ciprofloxacin coated by a transparent polymer commonly used in contact lenses. The former polymer modeled represents poly(lactic-co-glycolic)acid (PLGA), which is biocompatible and has demonstrated ability to control zero-order release kinetics3. This is possible since PLGA degrades with time, maintaining a relatively constant driving force of diffusion despite a finite source of drug. The latter polymer modeled represents pHEMA, which is not biodegradable. By assuming that the radius of the eye is much larger than the contact lens thickness, a two-dimensional, axisymmetric simulation of the system was created with the eye containing the following layers: tear film, epithelium, stroma, and aqueous humor (Figure 1). The model was run with a starting PLGA molecular mass of both 118 kDa and 18 kDa, with initial ciprofloxacin mass of 20mg (1:1 ratio with 118kDa PLGA). Thus, the effect of altering the ratio of drug to PLGA was monitored. The results determined that both 118kDa and 18kDa PLGA show sustained release of ciprofloxacin for one month after a quick initial burst, with the 18kDa PLGA system exhibiting a higher steady state flux. The geometry also proved to be superior, with the design showing a 100% increase in ciprofloxacin concentration in the eye after 50 days over a design with PLGA spanning the width of entire lens. This study thus exhibits prolonged zero-order release kinetics at a therapeutically relevant concentration can be achieved with a contact lens. This prototype can be extended to further applications of ocular drug delivery and have enormous implications in the treatment of eye infections.
  • Item
    Peritoneal Kidney Dialysis
    Basu, Arunabha; Lim, Esther; Zhou, Sherry; Zhu, Charles (2010-05-21T18:33:38Z)
    Chronic kidney disease is a public health problem that afflicts over a tenth of the United States adult population. In this report, we describe the evaluation and analysis of peritoneal dialysis as a method of treatment for chronic kidney disease patients on the computational level. Using COMSOL Multiphysics and Simulation, we modeled the peritoneal cavity and surrounding blood vessels as a 2D slab using a thin-wall assumption and simulated urea mass transfer from the capillary bed through the peritoneal membrane and into the dialysate. Literature values for parameters such as urea diffusivity, bulk flow due to osmotic pressure difference, blood urea concentration, and bodily urea generation were used as modeling parameters. Overall, our results reflected the ability of peritoneal dialysis to adequately remove waste urea from the body. With a drainage/infusion cycle every 5 hours, urea concentration can be maintained at a relatively constant level as peritoneal dialysis removes systemically generated urea. Alternatively, a greater number of shorter peritoneal dialysis sessions removed a significantly higher quantity of urea, resulting in an overall decrease in blood urea concentration. Our sensitivity analysis reflected the significance of certain parameters in peritoneal dialysis and therefore the areas that can be emphasized in such treatment to achieve varying results. Dialysate volume, peritoneal membrane surface area, and bodily urea generation most severely affected post-dialysis urea concentration while urea diffusivity through the capillary bed, peritoneal membrane thickness, and initial urea concentration had little impact.
  • Item
    Modeling of Antibiotic Diffusion from Implanted Absorbable Beads in Localized Injured Tissue
    Grant, Kara; Lawson, Olivia; Polins, Ben; Suen, Leslie (2010-05-21T18:30:47Z)
    Prevention of infection during the healing process in tissue is a major concern and is especially crucial for the health of the patient. In order to combat infection due to bacteria, antibiotics must be deployed in the tissue at a high enough concentration for a long enough period of time in order to be deemed effective. Historically, systemic antibiotics have been administered to diminish the threat of infection, either through intravenous injection or oral ingestion. However, toxicity levels of the drug in the entire body must be taken account, lowering the efficacy of systemic antibiotics. Implanted antibiotic beads provide an attractive alternative by locally administering the antibiotics to the injured soft tissue, allowing higher concentrations of antibiotics to be delivered over a longer period of time, thus increasing the efficacy and safety of the drug. Although these beads must be physically placed into the injured soft tissue, a string of them can easily be implanted as the last step of surgery as a protective measure against future infections. The goal of this project is to model the delivery of vancomycin-impregnated biodegradable antibiotic beads as the vancomycin diffuses into the injured soft tissue. We constructed the model using 2D-axisymmetric geometry in COMSOL. Our bead was comprised of two separate layers, one made of Polylactic Acid containing the initial drug concentration of vancomycin of 12.9 mol/m3, and an outer layer of Polylactide-Polyglycolide Copolymer (PLA-PL:CG), containing no vancomycin. We successfully modeled the shrinking of the biodegradable bead and were able to get results that agreed with other experimental data. These results included collecting data for concentrations of vancomycin in the tissue over time and space. For vancomycin to be successful in fighting infection, the concentration needs to be above the minimum effective level while remaining below the toxic level. Using these requirements, we determined the optimal spacing between multiple beads to be 0.56cm. Our model can be used in place of experimentation to determine the optimal combination of parameters that meet a patient’s needs. This model allows estimation of concentration profiles in the tissue that would otherwise be difficult to gather in an experimental setting. Therefore, our model can be used as a tool for physicians to better prevent infection in traumatic injuries and surgeries.
  • Item
    Modeling Treatment of P. aeruginosa Biofilms in the Lungs Using Aerosolized Tobramycin
    Inzana, Jason; Ishida, Chikara; Rhinehardt, Kristen; Yang, Jennifer (2010-05-21T18:26:53Z)
    The biofilms produced and maintained by Pseudomonas aeruginosa in the lungs of cystic fibrosis patients are difficult to treat and can have fatal effects. Antibiotics are necessary to control and eliminate these bacterial biofilms, but in vivo administration may not be the most effective means. Tobramycin, a commonly used antibiotic for treating cystic fibrosis patients, has been commercially developed into a solution that is inhalable via nebulizer. Inhaling this mist form of the antibiotic will allow administration of higher concentrations at the site of infection. The goal of this study was to develop a model using COMSOL Multiphysics to better understand the distribution of tobramycin to bacterial biofilms in the lungs. Like nearly all medications, tobramycin can become toxic at high concentrations. Since filtration from the blood stream is the only significant mechanism of tobramycin elimination, the kidneys are at the greatest risk for toxicity. Therefore the study focused on the possibility of maintaining safe blood serum concentrations while providing sufficient doses to inhibit the bacteria occupying the lungs. The model showed that the bacteria’s minimum inhibitory concentration was easily achievable throughout the biofilm while keeping the blood serum concentrations at a safe level.