BME Design Day 2020

Undergraduate Design Team Projects

DT1 HerniAid: Non-invasive, Real-Time Monitoring of Hernia Growth and Development Team Members: Anisha Palasamudrum, Varun Naga, Celina Shih, Ranjani Ramasubramanian, Alan Xu, Stanley Zhu, Eric Guan, Ryan Zhao

Advisors: Sebastien Ladet, PhD; Hien Nguyen, MD; Elizabeth Logsdon, PhD; Medtronic: Jonathan Thomas; Daniel Broom

Abstract: Approximately 5 million adults in the US are currently living with a hernia. A large subset of these patients have ventral hernias, which the average person has a 10% chance of developing in their lifetime. 20% of ventral hernia cases lead to emergent surgery, resulting from abdominal contents being trapped in the hernia (incarcerated), or worse, when blood supply is cut off from the herniated tissue (strangulation). Emergent procedures often require bowel  resection, which has both immediate (site infections, mortality) and long-term (hernia recurrence) consequences. Emergent surgery is also vastly more expensive for both patients and health care providers, costing approximately 10 to 20 times more than an elective surgery. However, there is often a gap between diagnosis and surgery where the surgeon is unable to continuously monitor a patient’s hernia. During this waiting period, there is ample opportunity for the hernia to become incarcerated and eventually strangulate, yet there are no easily accessible methods to detect this worsening, making the chance of emergent procedures much higher. Thus, a method of monitoring at this point of intervention will improve patient outcomes at a lower cost to both patients and health care providers. Our solution aims to track hernia progression non-invasively by utilizing physical metrics to provide surgeons and patients with real-time data to inform their decisions.

Watch Video
Design Team 2UroDeflect: Ureteral Occlusion Device Diverting 100% of Urine from Damaged Urinary System Tissue Team Members: Chris Shallal, Shivam Rastogi, Nicky Zhang, Jay Mandavilli, Ben Treutler, Sahana Kumar, Sean Healy, Nehali Gupta

Advisors: Clifford Weiss, MD; Vishal Kadakia; Collin Shale; Timothy Weihs, PhD; James Wright, MD

Abstract: Urinary diversion via a percutaneous nephrostomy tube (PCN tube) is a surgical procedure that temporarily redirects the natural flow of urine out of the body through decompression, utilizing the pressure gradient created by a pathological obstruction in the urinary system (typically kidney stones or malignancy). These decompressive devices continue to be used in cases of fistulas, leaks, and surgical anastomoses even though there is no obstruction to create the pressure gradient needed to divert urine. This treatment gap results in persistent urine flow and continued damage from continuous contact with urine, resulting in delayed healing of the wound. Hence, there is a demonstrated need for a reliable, complete urinary diversion, which would decrease the frequency of patient returns to the hospital because of insufficient healing. UroDeflect aims to improve temporary urinary diversion by Interventional Radiologists in non-obstructed cases by providing a non-migrating, artificial obstruction to safely divert urine while minimizing tissue damage. Furthermore, UroDeflect is deployed via similar methods as current decompression devices and thus can be easily integrated into the current Interventional Radiology clinical workflow.

Watch Video
Design Team 3 Luminate: A System for Real-Time Position Monitoring of Neonatal Endotracheal Tubes Team Members: Claire State, Matthew Wang, Eli Foster, Sam Zmily, Aidan Aug, Samantha Weed, Ashley Li, Marianna Elia

Advisors: Nicholas Durr, PhD; May Chen, MD; Maide Ozen, MD; Lawrence Nogee, MD; Nicholas Dalesio, MD; Jen Organ, RRT; Lynn Morooney, NNP; Grace Weyand

Abstract: Endotracheal intubation is used in the NICU to provide assisted ventilation for neonates suffering from respiratory disorders. In very premature infants, the margin of error for correct placement of an endotracheal tube (ETT) is less than ± 1⁄2 cm. This contributes to a high occurrence of tube malpositioning, which is associated with adverse events that include unplanned extubation (UPE), where the ETT is dislodged because it is too shallow, and endobronchial intubation, where the ETT is inserted too deep. UPEs increase the risk for severe lifelong complications, such as neurological damage and airway scarring, while endobronchial intubation can result in lung collapse and airway damage. Our team aims to address neonatologists' need to reduce the complications of tube malpositioning by providing continuous monitoring of ETT position throughout a patient’s stay.

Watch Video
Design Team 4 UniStand: A Novel One-Arm-Drive Mobile Stander for Children with Hemiplegia

Team Members: Danielle Vaithilingam, Ivan Vuong, Robert Huang, Jackelyn Navar, Yurie Hong, Nyeli Kratz, Rachel Li, Ashley Koenig

Advisors: Stacy Suskauer, MD; Christopher Joseph, MSPT; Shababa Matin; Scott Paul, MD; Tom Benassi

Abstract: Over 4,000 pediatric patients with cerebral palsy or a traumatic brain injury struggle to control one of their arms and one or both of their legs. These children cannot walk and are unnecessarily confined to wheelchairs, despite being able to stand with assistance. This confinement to a seated position limits their independence and exposes them to social stigma, depression, and anxiety due to their inability to meet their peers at eye level and take part in standing activities. Prolonged sitting also causes severe physiological consequences, including deep vein thrombosis, muscle atrophy and poor respiratory and cardiovascular health – issues that can be alleviated with just a few hours of standing each day.

Although these children would greatly benefit from a method of standing mobility, current options like motorized standing wheelchairs are extremely expensive, unreliable, and fail to provide the physical benefits of manual self-propulsion. Current mechanical standers cannot be propelled with one arm and are therefore unusable by this underserved population.

Team UNIStand is designing and building a mechanical, one-arm-drive mobile stander. Both cost-effective and reliable, the AgileStand is more financially accessible than competing devices. The stander allows users to propel themselves and control the direction of the stander using only their unaffected arm. It also allows users to maintain a prone standing position and put weight on their legs while being comfortably supported by the stander. UNIStand’s AgileStand mitigates the adverse mental and physiological health problems caused by prolonged sitting and enables users to achieve greater independence.

Watch Video
Design Team 5 PneuTech: Increasing Sample Yield and Decreasing the Risk of Pneumothorax

Team Members: Katia Kovrizhkin, Sean Darcy, Ashley Tsang, Jacob Desman, Deborah Weidman, Bibhav Poudel, Tatiana Pereira, Gohta Aihara

Advisors: Robert Liddell, MD; Shababa Matin; Harjit Singh, MD; Brian Holly, MD; Clifford Weiss, MD

Abstract: A percutaneous core needle lung biopsy is a minimally invasive procedure by which a sample of abnormal lung tissue is obtained through a needle inserted directly into the targeted site of the lung using imaging guidance. The most common complication during this procedure is pneumothorax, a collapsed lung, which occurs in ~20% of all patients. With over 100,000 procedures performed in the United States alone, pneumothorax contributes to over $120 million in annual hospital costs. Thus, there is a critical need to improve the safety of percutaneous lung biopsies. The current standard of care utilizes a straight needle guided by computed tomography (CT), making it a challenge to direct the needle to the target site while navigating around critical structures, such as major blood vessels, airways, and ribs. Therefore, additional pleural punctures within a single biopsy or a repeat biopsy may be necessary to obtain an adequate sample, increasing the likelihood of pneumothorax. Our solution addresses these issues via a novel needle system equipped with the ability to access any target site in any location while minimizing the number of pleural punctures. By enabling safer high-yield lung biopsy procedures, our device has the potential to revolutionize the lung biopsy space.

Watch Video
Design Team 6 VestiTech: VestiTech is Determined to Make Early Concussion Screening More Accessible with VR

Team Members: Emily Burnette, Sarah Bortel, Vicky Chen, Nick Gutierrez, Jakob Heinz, Jae Park, Tori Rose, Maggie Wang

Advisors: Michael Schubert, PhD; Mark Shelhamer, ScD; Collin Shale; Grace Weyand; Kathleen Cullen, PhD; Kemar Green, DO; Jorge Otero-Millan, PhD

Abstract: In rural clinics, on the field during sports games, and in the military field after an IED blast, underdiagnosis of mTBI (mild Traumatic Brain Injury) is common. Underdiagnosis can lead to severe long-term injuries like Second Impact Syndrome, where the brain swells rapidly after incurring two mTBIs in a short timeframe, and can also result in unnecessary healthcare costs in the long-term care of secondary symptoms like PTSD. To combat this issue, VestiTech is developing a portable and intuitive device that provides quantifiable data that can be used to screen for an mTBI. Using our clinical mentors’ patented vestibular test, the Vertical Alignment Nulling & Torsional Alignment Nulling (VAN/TAN), in a VR environment, we measure abnormal perception of the relative difference in vertical and torsional alignment between the eyes, a hallmark of mTBI. By measuring these biological properties, we hope to reduce the subjectivity issues of current testing methods and, consequently, the widespread underdiagnosis of mTBI.

Watch Video
Design Team 7 MUSE for Mohs: Optimizing Skin Cancer Treatment with a Rapid, Slide-Free Margin Assessment Tool

Team Members: Hadley VanRenterghem, Afareen Jaleel, Seoyoung Kwon, Shreya Narayan, Sujai Jaipalli, Vinay Ayyappan, Walee Attia, Wally Niu

Advisors: Elise Ng, MD; Nicholas Durr, PhD; Alexander Baras, MD; Greg Mckay; Danaher Corporation (Leica Biosystems): Tracy DeGeer; Martha Davis

Abstract: Mohs micrographic surgery (MMS) is the gold standard treatment for nonmelanoma skin cancer, one of the most common types of cancer. MMS has excellent patient outcomes, because the surgeon iteratively examines the entire surgical margin via histopathology, allowing the tumor to be excised completely. This process is repeated for each stage, leading to scheduling complications, high costs, a strain on hospital resources, and reduced accessibility for patients. Our goal is to acquire images quickly and at a low cost while maintaining the quality needed to inform iterative resections. While standard processing requires the tissue to be frozen and sliced onto slides, our solution will enable unsliced tissue to be imaged by integrating Microscopy with UV Surface Excitation (MUSE) into the Mohs workflow. By streamlining the most inefficient steps of the procedure, we will enable the gold standard of nonmelanoma skin cancer treatment to be accessible to more patients.

Watch Video
Design Team 8 Improving Gastric Enteral Access in Neonates and Infants: Optimization of Pediatric Gastrostomy Tubes to Reduce Tube Displacement

Team Members: Mihika Aedla, Charlotte Cheng, Jocelyn Hsu, Kathy Hu, Jason Qian, Kevin Sompel, Siya Zhang, Anson Zhou

Advisors: Karun Sharma, MD; Elizabeth Logsdon, PhD; Vahe Badalyan, MD; Ginny Gebus; Anthony Ho; Alina Andrews; Children’s National: Karun Sharma; Anthony Ho

Abstract: At any given time, nearly 1.4 million American children require feeding assistance. Of these children, around 117,000 are neonates and infants who experience persistent, severe feeding dysfunction and must be fed through gastrostomy tubes, the current standard for long-term gastric enteral access. However, in this population, 22% of G-tubes are displaced within 3 months of placement, before the recommended replacement period. Tube displacement, ranging from persistent and excessive motion at the insertion site to complete dislodgement, can cause complications such as gastric leakage, skin irritation and infection, and may even necessitate hospital readmission for tube replacement. This is particularly debilitating for these children who are at the most critical stage of development.

Neonates and infants with gastrostomy tubes need a way to minimize the frequency of displacement in order to reduce these complications. G-tube displacement is largely attributed to the failure of the internal retention mechanism and the excessive force applied when managing the tube. Most G-tubes feature a balloon retention mechanism which is prone to leaking or deflation, and in neonates and infants especially, G-tube replacement can be painful for the patient and challenging for parents and caregivers. Our team is targeting this gap by designing a G-tube for neonates and infants that features a secure internal retention mechanism with an easy insertion/removal method optimized for all users. By reducing the risk of tube displacement, we hope to provide neonates and infants with a reliable method of assisted feeding, enabling them to receive the nutrition they need for healthy development.

Watch Video
Design Team 9 Urinefo: Data-Driven Urinary Catheter Monitoring

Team Members: Emily Chang, Clara Kochendoerfer, Rodrigo Lamas, Maya Lapinski, Silu Men, David Shi, Allen Wang, Justin Zhou

Advisors: Stuart Ray, MD; Nicholas Durr, PhD; Disha Mankodi; James Fackler, MD; Kieren Marr, MD; Mason Chen

Abstract: Catheter-Associated Urinary Tract Infection (CA-UTI) is the most common hospital-acquired infection in the world, affecting over 500,000 individuals each year. The condition arises when bacteria are introduced into or fostered within the urinary tract via the insertion and retention of a urinary catheter. Catheterization is a common practice performed in patients with limited mobility or bladder control, such as those in intensive care, geriatrics, or nursing facilities. Unfortunately, this poses certain risks, increasing an individual’s risk of developing bacteriuria, or bacteria in the urine, by 3-10% each day. While some such cases are asymptomatic, the presence of this bacteria increases an individual’s risk of developing infection. This problem is exacerbated by the lack of objective clinical criteria for determining appropriate catheter duration: up to 47% of catheter days are unnecessary, due simply to oversight or convenience. This has resulted in over 300,000 cases of CA-UTI each year that would have been preventable with appropriate catheter use, duration, and care.

Urinefo’s mission is to prevent infections before they occur, while also ensuring utmost patient safety and reducing the burden to both hospitals and healthcare providers. Based on objective measures of urine bacteria count, our automated monitoring system empowers physicians to make informed decisions regarding catheter removal.

Watch Video
Design Team 10 Identifying Source of Pulsatile Tinnitus to Provide Targeted Treatment for Patients

Team Members: Chun Hei Ryan Chan, Tara Foley, David Kavteladze, Min Jung Kim, Alisha Kodibagkar, Albert Lee, Joshua Ni, Eesha Verma

Advisors: Ferdinand Hui, MD; John Carey, MD; Daniel Sun, MD; Shababa Matin; Deepa Galaiya, MD; Ian McLane; Rajat Mittal, PhD

Abstract: Pulsatile tinnitus (PT) is a condition which affects over five million people in the US; these patients hear a constant, rhythmic sound caused by complications in their cranial vasculature. PT severely debilitates the patients’ quality of life, as up to 60 % of PT patients develop major depression or anxiety as a result. PT is a symptom of more serious disorders localized in the head such as intracranial hypertension, fistulas, tumors, and arteriovenous malformations (AVM). During the diagnostic process, patients may undergo tens of thousands of dollars’ worth of medical imaging scans, including MRI, CT, dynamic CTA, and angiography to identify the specific disorder causing PT and its location within the cranium. Despite these efforts, years can pass without a definitive diagnosis since the process is often long, costly, and futile. Many patients never undergo targeted treatment and must simply live with the condition. Additionally, surgeons may operate on an area suspected to cause PT only to discover afterwards that the problem persists, meaning that an incorrect diagnosis was made. We are developing an easily operable, low cost, and noninvasive device that allows clinicians to efficiently diagnose PT in patients. Our device is crucial to identifying the problem early in the patient provider pathway and cutting down on the time and money patients spend on diagnosis, resulting in a more effective treatment of PT.

Watch Video
Design Team 11 EdgeSense: Tumor Cavity Fiducial Marker for Improved Postoperative Breast Cancer Care

Team Members: Pascal Acree, Amee Kapadia, Darsh Patel, Ronak Mahatme, Lucia Zhang, Charlie Almoney, Morgan Kofsky, Brian Park

Advisors: Mehran Habibi, MD; Sara Alcorn, MD; Kristin Broderick, MD; Shababa Matin

Abstract: Of the 250,000 women diagnosed with breast cancer annually in the US, 60% are treated with breast conserving surgery (BCS): a lumpectomy followed by targeted radiotherapy. Around 22,500 BCS patients further undergo oncoplastic reconstruction of the breast following lumpectomy to improve cosmetic outcomes. After the procedure, the surgical margins (edges) of the specimen are sent to pathology, guiding radiation planning. However, during oncoplastic reconstruction, the margins and tissue around the tumor bed are moved, making it nearly impossible for radiation oncologists to precisely delineate the cavity when planning follow-upradiotherapy targeted to the breast. Imprecise delineation of the tumor bed compromises radiotherapy efficacy as it leads to the radiation oncologist either having to increase dosage to the cavity or increase number of therapy sessions required, exacerbating the side effects and costs of treatment. However, if the irradiation dosage or quantity of dose fractions administered is not increased, rates of cancer recurrence increase. EdgeSense has developed a novel fiducial marker that provides accuracy and reliability for post-operative breast cancer radiation therapy, bringing confidence to radiation oncologists and comfort to patients.

Watch Video
Design Team 12 Thrombotect: A Wearable Temperature Sensor for Continual Deep Vein Thrombosis Monitoring

Team Members: Shayan Hemmati, Timothy Bedard, Sneha Batheja, Sarah Kulkarni, Ayon Mitra, Aryaman Shodhan, Isabel Vanderzee, Bill Zhang

Advisors: Adam Sapirstein, MD; Collin Shale; Elliott Haut, MD; Michael Streiff, MD; Steven Tropello, MD; David Maher

Abstract: Deep vein thrombosis (DVT) is the formation of a blood clot in a major vein (usually in the lower limbs). The most severe complication of DVT is pulmonary embolism (PE). This occurs when a thrombus detaches from the vein and migrates to the pulmonary circulation. DVT has a prevalence as high as 10% in ICU patients, and the associated costs of DVTs are tens of millions of dollars per year in the US. While all ICU patients are at increased risk for DVT, there is no consensus standard for diagnostic ultrasound screening. Instead, in most ICUs, clinicians make screening decisions based on their estimates of DVT probability in individual patients. This may lead to late or missed diagnoses along with unnecessary testing. Our team has developed a wearable, temperature-based screening tool that will alert clinicians about the probability of DVT. Thrombotect will give physicians a continual method to monitor ICU patients for the probability of DVT formation. This information can be used to pursue diagnostic ultrasound and treat DVT when necessary. This paradigm should reduce time to treatment, mortality rates, and costs of care.

Watch Video
Design Team 13 Auditus: Improving User-Centered Hearing Technology for Older Adults with Dementia Team Members: Shubhayu Bhattacharyay, Eshan Joshi, Annie Liang, Ben Straus, Cindy SeYeon Choi, Alex Rovalino, Mohan Peddada, Rumani Kafle Advisors: Carrie Nieman, MD; Shababa Matin; Kunal Parikh, PhD; Kevin Franck, PhD; Tilak Ratnanather, DPhil

Abstract: Within the last decade, the world’s leading dementia investigators have uncovered a decisive relationship between age-related hearing loss and dementia, identifying hearing loss as the leading modifiable risk factor for incident dementia and cognitive decline. However, no hearing loss solution in the market addresses the unique physiological, psycho-behavioral, and socioeconomic needs of dementia patients with hearing loss. As a result, despite validated evidence that hearing treatment can improve long-term outcomes for dementia patients, the rate of hearing technology use among dementia patients remains among the worst after correcting for age, race, family income, and geographical location. By August 2020, the FDA Re-authorization Act of 2017 will authorize the over the counter (OTC) sale of hearing devices intended to treat mild-to-moderate hearing loss. We recognize this as a unique opportunity to introduce the first direct-to-consumer solution targeted for older patients with dementia, bypassing stifling market incentives and augmenting access by an underserved population. In working with Access HEARS, a non-profit founded by our clinical advisor, we directly assess the needs of this target population, interview patients, caretakers, and health care providers in our design process, and conduct proof-of-concept trials.

Watch Video
Design Team 14 HydroGene: Driving Gene Delivery from Pipette to Patient

Team Members: Siddharth Arun, John John, Kisha Patel, Pranav Reddy, Shreya Sharma, Betul Celiker, Joshua Devier, Ana Rosu

Advisors: Florin Selaru, MD; Vivek Kumbhari, PhD; Yervant Ichkhanian, MD; Shababa Matin; Jordan Green, PhD; Robert Kruse, MD; Yuting Huang, MD; Shayan Roychoudhury; Tom Benassi

Abstract: Worldwide, over 4M patients above the age of 25 are affected by liver-based monogenic hereditary disorders. Hemophilia A, affecting roughly 20K males in the United States, is a burdensome liver-based genetic condition. The current standard of care for these patients is a thrice-weekly infusion of blood clot supplements to manage chronic bleeds, resulting in annual medical costs over $300K. Hospitalizations for acute bleeds have resulted in an additional $1M annually upon the patient. Gene therapy has been proposed as a potential cure for these costly and difficult-to-manage conditions, as it can replace a missing or defective gene and treat the disease. Traditionally, gene therapy is delivered through the bloodstream via adeno-associated viral vectors (AAVs). However, AAVs can trigger an immune response, as many patients have preexisting antibodies that fight AAVs. Additionally, due to its small genome, AAVs require complex and expensive bulk manufacturing systems. An alternative to viral delivery, HydroGene’s novel non-viral system integrates into a common endoscopy procedure that accesses liver cells via the biliary ducts, a delivery site well supported in literature. While traditional gene therapy lacks the ability to target specific tissues, HydroGene provides a localized delivery that maximizes safety and efficacy and minimizes procedural costs. Additionally, HydroGene’s technology is versatile and applicable to an array of monogenic disorders and compatible with a wide range of therapeutics, including DNA, RNA, and protein-based medicines. With a simpler, safer method of gene delivery, HydroGene aims to accelerate the progress and expand the reach of gene therapy research.

Watch Video
Pediatric IV Infiltration: Improving the Standard for Early Detection of IV Infiltration in Neonates

Team Members: Kavya Anjur, Taj El-Khalili, Laboni Hassan, Sneha Kamada, Benjy Monteagudo, Rebecca Mosier, Sophia Triantis, Jack Walters

Advisors: Robin Yang, MD, DDS; Collin Shale; Hillary Jenny, MD; Richard Redett, MD; Stephanie Morgenstern, RN; Jim Fackler, MD; Karina Frank

Abstract: Every year in the neonatal intensive care unit (NICU) between 95,000 to 116,000 patients in the United States will experience an IV infiltration. IV infiltration occurs when the fluid administered leaks out of its intended path in the vein and into surrounding tissue. Failure to detect this promptly can often lead to damaging effects such as necrosis and compartment syndrome which increase the length ofhospital stay and cost of care.  Currently, nurses monitor the IV site every few hours for symptoms of swelling, blanching, and change in temperature. Nurses are often not able to monitor the IV site frequently enough to catch these early symptoms that present themselves within minutes of infiltration. NICU nurses need a highly sensitive way to rapidly detect IV infiltration in order to minimize time before intervention. While effective in the adult population, current solutions are too expensive for use on all patients and not designed for neonate anatomy. Our solution is a continuous monitoring device to alert exactly when early symptoms of infiltration occur. This low-cost system is designed specifically for neonates and improves nurse workflow by reducing the need for IV site checks every few minutes.

Watch Video
Design Team 16 Endoscopic Endonasal Skull Base Surgery: A Surgical Instrument that Deskills Endoscopic Endonasal Skull Base Reconstruction

Team Members: Tejasvi Desai, Arshdeep Singh, Daniel Lewis, Diane Lee, Kaitlyn Calabresi, Mahita Varanasi, Ria Jha, Tianyu Wang

Advisors: Nicholas Rowan, MD; Collin Shale; Debraj Mukherjee, MD; Omar Ahmed, MD; Tom Benassi; David Maher

Abstract: Endoscopic endonasal skull base surgery (EESBS) is a minimally-invasive surgical procedure used to treat pathologies of the skull base, the interface of the sinonasal passages and the brain. Roughly 20,000 EESBS cases are performed in the US annually. During EESBS, the pathology is accessed by making a defect in the dura mater, a membrane between the sinonasal and intracranial space, which may cause cerebrospinal fluid (CSF) leakage during the procedure. Repair of the dural defect with grafts is a critical step, as a postoperative CSF leak may result in reoperation, a 10% to 40% risk of meningitis, or even death. Thus, EESBS is only as effective as its reconstruction. However, reconstruction is challenging and time-intensive due to limited surgical tools and anatomical variables such as pressure from CSF pulsations. Thus, surgeons need a method that reduces the technical difficulty of reconstruction. Our solution is a surgical instrument that allows surgeons to simultaneously tuck multiple sides of the graft underneath the dura during reconstruction, thereby reducing operating time and the risk of postoperative CSF leaks.

Watch Video
Design Team 17 PostOptiX: Assessing Blood Perfusion Post Free Flap Breast Reconstruction

Team Members: Zach Zarubin, Kayla Gee, Josh Krachman, Jessica Shen, Siddharth Krishnan, Matthew Kunkler, Quinn Hauck, Aydin Turkay

Advisors: Abhishek Rege, PhD; Jay Kandukuri, PhD; Gedge Rosson, MD; Damon Cooney, MD; Elizabeth Logsdon, PhD; Nicholas Durr, PhD; Vasoptic Medical: Abhishek Rege

Abstract: Free flap breast reconstruction is a procedure in which a plastic surgeon reconstructs the breast by completely removing tissue from one body part and transplanting this “free flap” to the chest. Around 18,000 of these operations are performed annually, and immediate breast reconstruction after mastectomy is increasing by approximately 5% each year. However, this procedure incurs the risk of cell death (necrosis) as a result of failing to reestablish perfusion throughout the flap, necessitating postoperative monitoring. Current methods employed to monitor blood reperfusion are invasive, do not measure the entire flap, or cannot make quantitative predictions of a region's likelihood to necrose. Our solution is to provide a Laser Speckle Contrast Imaging device that addresses these limitations, in an effort to improve and streamline the industry’s standard for preempting postoperative free-flap necrosis.

Watch Video


MSE Student Design Projects

Oculy Oculy: An Implantable Drug Delivery System for the Treatment of Diabetic Macular Edema

Team Members: Brooke Hayley, Niranjanaa Jeeva, Disha Mankodi, Natsuha Omori, Shayan Roychoudhury, Noah Yang

Advisors: Kunal Parikh, PhD; Soumyadipta Acharya, MD, PhD; R. D. Ravindran, DO; R. Kim, DO; Chitaranjan Misra, DO; Pradeep Ramulu, MD, PhD; David Green; Namratha Potharaj

Abstract: Around the world, the prevalence of diabetes is increasing at staggering rates. Over 422 million individuals are struggling with this condition and its comorbidities globally, and this number is projected to increase to 642 million by 2040. The burden of diabetes disproportionately affects low and middle-income countries (LMIC) because of major challenges in healthcare infrastructure. In India, there are over 70 million diabetic patients and this number is rapidly increasing.  One manifestation of diabetes is diabetic macular edema (DME), the fluid build-up in the back of the eye caused by the leakage of highly permeable retinal blood vessels characteristic of diabetic patients. This condition causes visual impairment in 31 million people worldwide. Many patients with DME are prescribed monthly intravitreal injections. However, while visiting Aravind Eye Hospital, the world’s largest high-volume eye care system, our team saw that on average patients were only receiving two injections per eye per year. Many patients are not able to receive the prescribed monthly injections because of the recurring cost of the medication and travel required. Thus, patients need an effective, long-lasting treatment for diabetic macular edema in order to impede disease progression and prevent vision loss. Oculy meets this need with our device, an ocular implant that slowly releases medication to the back of the eye over an extended period of time, thus improving clinical outcomes for patients who cannot visit the hospital as frequently. Oculy reduces preventable blindness by increasing access to treatment for chronic eye diseases globally.

Watch Video
revIVe: Pulsatile Flushing Syringe which Ensures Compliance to Nursing Standards

Team Members: Brooke Hayley, Niranjanaa Jeeva, Disha Makodi, Natsuha Omori, Shayan Roychoudhury, Noah Yang

Advisors: Clifford R. Weiss, MD; Youseph Yazdi, PhD; Ami Kumordzie, MiKaela Olsen

Abstract: Over five million central venous catheters (CVC) are inserted in the United States every year, to allow direct delivery of medication into the blood. Up to 15% of these catheters can become occluded, often due to blood clots clogging the lumen of the catheter. Once occlusions are detected, they must be cleared immediately to prevent downstream complications. These clots can delay treatment and become an incubator for bacteria, increasing the risk for Central Line-Associated Bloodstream Infection (CLABSI), which leads to 28,000 deaths in the US per year. Patients with occluded catheters are nearly three times more likely to develop CLABSI. In the hospital, inpatients with occlusions are treated with time-consuming and expensive declotting agents If the occlusion is not treatable with the agent, patients must have their CVC replaced, increasing their length of stay at the hospital. To prevent catheters from getting occluded, the gold standard as stated in the Infusion Nurse Society policy is to flush the catheter with saline using a start-stop method, where 10 short boluses of 1 mL are delivered with a syringe, interrupted by brief pauses. Several studies have shown that pulsatile flushing is at least twice as effective in clearing the inside of the catheter. However, interviews and observations at the Johns Hopkins Medical Institute have revealed that nurse compliance with pulsatile flushing is as low as 6%. Therefore, nurses need an easier and quicker way to introduce pulsatile flow in lines in order to prevent thrombotic occlusion of central venous catheters. RevIVe presents PulseFlush, a disposable syringe that makes sure that the gold standard of flushing is met, regardless of how busy or experienced the nurse is. Nurses can push the syringe continuously, while the device creates the pulsatile flow profile. Consistent pulsatile flushing will break up any debris inside the catheter, preventing clots from forming and bacteria from growing inside. Nurses no longer have to actively push and pause the plunger of the syringe, making it easier for them to comply with the standard of care.

Watch Video
NEC Check NEC Check: Enabling Earlier Detection of Neonatal Necrotizing Enterocolitis

Team Members: Alina Andrews, David Maher, Elizabeth Russo, Madeline Wartan, David Zarrin

Advisors: David Hackam, MD; May Chen, MD; Mark Kovler, MD; Emily Dunn, MD; Youseph Yazdi, PhD; Soumyadipta Acharya, PhD; Aditya Polsani, MS; Joshua de Souza, MSE; Medtronic, Minimally Invasive Therapies Group (Boulder, CO)

Abstract: Necrotizing Enterocolitis (NEC) is an inflammatory disease that causes bowel necrosis in premature infants. It is the number one cause of death in infants weighing less than 1500 grams and the number one surgical emergency among all patients in the Neonatal Intensive Care Unit (NICU). 380,000 infants are born prematurely annually, with over 90,000 at elevated risk for developing NEC. This disease can onset acutely and severely, causing seemingly healthy infants to undergo surgery within 24 to 48 hours. Current diagnostic methods, which include physical examinations and abdominal X-rays, are not sensitive and specific enough to facilitate early detection of NEC. Physical examinations consist of manual abdominal girth measurements and auscultation. These assessments are both subjective and infrequent, as they occur every three to four hours. X-ray, the diagnostic gold standard, is only effective at diagnosing late-stage NEC. These limitations result in a large percentage of the 5,400 infants diagnosed with definite NEC each year to require surgical intervention since medical treatment was initiated too late. Surgical intervention carries a 30% mortality rate that results in over 800 infants dying each year. Our solution for earlier NEC detection is a novel four-quadrant abdominal patch sending output data to an algorithm that is capable of objectively and continuously monitoring the earliest physiological and clinical indicators of the disease: reduced bowel sounds and altered gut perfusion. This proprietary algorithm will continuously isolate bowel sounds from ambient and environmental noise in neonatal critical care settings to estimate gastrointestinal motility levels. A separate algorithm will analyze temporal perfusion changes by trending regional oxygen saturation (rSO2) values in real-time. In combination, these objective metrics of gut perfusion and gastrointestinal motility will guide medical management of neonates. Integrating NEC Check into the current standard of care for this patient population can quickly alert providers of changes in abdominal health.

Watch Video
Safe-T Safe-T: Improving the Safety of Anesthesia Care Through Alleviation of Upper-Airway Obstruction

Team Members: Alina Andrews, David Maher, Elizabeth Russo, Madeline Wartan, David Zarrin

Advisors: Jake Abernathy, MD; Akbar Herekar, MD; Sarah Coppola, PhD; Prashanth Reddy, MBBS; Naresh Pagidimarry, MS; Youseph Yazdi, PhD; Soumyadipta Acharya, PhD; Aditya Polsani, MS; Joshua de Souza, MSE; Medtronic, Minimally Invasive Therapies Group (Boulder, CO)

Abstract: In a study of global access to surgical care, the Lancet Commission found that more than 4.8 billion people in the world do not have access to surgery. In India, this results in 90% of the population lacking access to safe surgical care. In order to achieve safe, affordable, and timely surgical care, patients also need access to reliable anesthesia care. Urban hospitals in India often deal with high patient volume which limits the amount of attention each anesthetist can reasonably give each patient. Accordingly, physician attention is a limited resource in these clinical settings. Maintaining a patent airway is essential for adequate oxygenation and remains a significant demand for the attention of the anesthesia team on a daily basis. Failure to maintain a patent airway can be life-threatening to patients. Glossoptosis is the downward displacement of the tongue that is often caused by loss of muscle tone. This complication often presents when escalating sedatives are administered to comfort patients during cases using regional anesthesia and can lead to complete obstruction of the upper airway. The primary clinical response involves three, two-handed maneuvers performed by the anesthetist: head elevation, chin lift, and jaw thrust. Although this approach effectively maintains the patency of the airway, it does not allow the provider to attend to other responsibilities until the risk of airway compromise is alleviated. In some cases, after five to 10 minutes of performing the jaw thrust, anesthetists will elect to intubate the patient and transfer to general anesthesia. Our solution to relieve this burden on anesthesia providers is a medical device that can keep the patient in the jaw thrust position. Although competitors in this space focus on replicating one of the three maneuvers performed by the anesthetist, Safe-T aims to prevent upper airway obstruction by designing a device that mimics all three of these maneuvers.

Watch Video
NeMo: Development of a Neonatal Monitoring System for Mothers in Rural Uganda

Team Members: Ryan Bell, Brielle Cenci, Kaveri Das, Grace Weyand, Hannah Sennik, Sneha Shivkumar

Advisors: Soumyadipta Acharya, PhD; Azadeh Farzin, MD; Christopher Golden, MD; Alain Labrique, PhD; Youseph Yazdi, PhD; Aditya Polsani

Abstract: Approximately 2.7 million newborns die each year. 99% of these deaths occur in low-resource settings, and 75% of these deaths occur during the first week of life. In Uganda, community health workers, or CHWs, visit the homes of newborns to assess their health during the first week of life. The CHWs check for seven danger signs that have been established by the World Health Organization (WHO). There are four qualitative signs: difficulty feeding, convulsions, chest indrawing, and movement only when stimulated; and three quantitative signs: respiratory rate > 60bpm, and temperature > 37.5°C or < 35.5°C. Identification of even one of these danger signs is indicative of severe neonatal illness, which means that the neonate should be taken to a health facility. The problem is that there are not enough CHWs to visit every newborn, which means that many neonatal illnesses go unnoticed until it is too late. Our solution is the NeMo (neonatal monitoring) system, which empowers mothers to assess their own baby’s health at home. This system aims to reduce the number of preventable neonatal deaths by improving the frequency and quality of neonatal health assessments during the first week of life. The NeMo system consists of an audiobook that guides the mother in checking for the qualitative danger signs and a wearable sensor that checks for the qualitative danger signs. The mother will use the system every day of the newborn’s first week of life, allowing for more frequent assessment of the neonate compared to the current standard of care.

Watch Video
OSMO: Methods of Reducing Fluid Overload in Patients with End Stage Renal Disease

Team Members: Ryan Bell, Brielle Cenci, Kaveri Das, Grace Weyand, Hannah Sennik, Sneha Shivkumar

Advisors: Chirag Parikh, MD; John Sperati, MD; Ashish Nimgaonkar, MD; Suzanne Przybyla; Pradeep Mehra, MD; Aditya Polsani; Youseph Yazdi, PhD; Soumyadipta Acharya, PhD

Abstract: Osmo is focused on reducing fluid overload (FO), which is defined as excess fluid in the blood, for patients diagnosed with end stage renal disease (ESRD). Patients with ESRD have limited to no kidney function and must receive renal replacement therapy (RRT). The most common therapy is in-center hemodialysis (ICHD) where patients visit a clinic three times a week for a machine to filter the blood of toxins and excess fluid. Healthy kidneys are constantly filtering blood, so the shift to a three-time per week filtration schedule has a severe impact on toxin and fluid clearance efficiencies. Fluid levels in the body fluctuate depending on fluid intake and output, and most patients diagnosed with ESRD have limited to no ability to eliminate these fluids through urination. The average patient on ICHD has 1.5L of excess fluid on their body between dialysis sessions. The symptoms of excess fluid manifest as difficulty breathing, high blood pressure, and swollen extremities. Severe cases, or patients presenting with FO above 1.3L, result in a significant increase in risk of cardiovascular mortality. Of the nearly 450,000 patients on ICHD in the United States, there are about 320,000 hospitalizations due to FO each year.  The Osmo team has developed both a hydrogel-based solution and an intra-abdominal implant solution to alleviate the buildup of fluid between ICHD sessions.

Watch Video
SubTract Medical SubTract Medical

Team Members: Ebenezer Armah, Zack Buono, Sophia Diaz, Karina Frank, Kiley Gersch, Monet Slinowsky

Advisors: Bashar Safar, MD; Chady Atallah, MD

Abstract: SubTract Medical aims to develop a novel, device-based approach to treating complex perianal fistulas. Perianal fistulas are a debilitating physical condition that affects 140,000 individuals annually in the U.S. alone. This condition presents in patients as a chronic inflammatory tract between the anal canal or rectum and the surrounding perianal skin. This condition can result in devastating symptoms such as constant drainage, chronic pain, psychological and social distress, and the inability to sit. If left untreated, fistulas can dramatically worsen, putting patients at risk of fecal incontinence or sepsis. Anorectal fistulas are classified as either simple or complex based on their anatomy and/or patient comorbidities. Factors that can cause a fistula to be deemed complex include its involvement of a significant portion of the anal sphincter muscle, comorbidities such as Inflammatory Bowel Disease, or the existence of multiple branches and/or openings. Simple fistulas are relatively easy to treat, but complex perianal fistulas pose a greater clinical challenge, with current solutions succeeding only 40-70% of the time. Our solution aims to deskill a technically challenging procedure that can be used to repair complex perianal fistulas which has shown promising results when performed in the hands of skilled surgeons. By closing the internal opening of the fistula simply and effectively without compromising the anal sphincter muscles, our solutions support the body’s natural healing process while simultaneously preserving patients’ fecal continence. By deskilling a prohibitively complex procedure, our device makes complex perianal fistula repair achievable to surgeons of all skill levels, thus making the solution accessible to a wider and more complex patient population.

Watch Video
ByteSight Technologies ByteSight Technologies

Team Members: Ebenezer Armah, Zack Buono, Sophia Diaz, Karina Frank, Kiley Gersch, Monet Slinowsk

Advisors: Jennifer Stevenson, PhD; Yvonne Linton, PhD; Mohamed Bayoh, PhD

Abstract: ByteSight Technologies aims to decrease the global malaria burden via an innovative computer vision tool that rapidly identifies the dangerous mosquitoes that carry and transmit malaria. Malaria is one of the most devastating vector-borne diseases, resulting in approximately 228 million cases and 405,000 deaths per year. The most effective way to reduce malaria cases is to prevent them before they occur. Vector control programs can accomplish this task by identifying and eliminating the malaria-transmitting mosquitoes before they can spread disease. Effective vector control operation is dependent on accurate and timely vector surveillance, or the monitoring of the density and distribution of these dangerous mosquitoes. Although the thousands of mosquito species in existence may look the same to the naked eye, only 40 of these are capable of actually transmitting malaria. All vector control implementation decisions depend on knowing the type, location, density, and behaviors of these vectors. Therefore, in order to effectively deploy the correct interventions directed at mosquito populations in specific locations, these programs must have accurate, species-specific mosquito demographic data. Current vector surveillance practices are time- and labor-intensive, requiring highly trained entomologists to visually identify individual specimens manually. Entomologists’ average identification rate is approximately one mosquito per 20 minutes, with an average identification accuracy of ~66%. Our computer vision-based technology addresses the current gap in surveillance by improving identification accuracy to well over 90%, and by identifying individual specimens on the order of seconds, thereby increasing the availability of key vector surveillance data needed for effective vector control.

Watch Video


Advanced Design Projects

SymMEDtry SymMEDtry: Minimizing Limb Length Discrepancy in Total Hip Arthroplasty Procedures

Team Members: Hannah Takasuka, Aditi Jithendra, Jerry Yan, Robert Li, Evan Bender, Akash Chaurrasia, Katie McCarren, Claire State

Advisors: Amir Manbachi, PhD; Julius Oni, MD; Adam Levin, MD; Robert Sterling, MD; Jeff Siewardsen, PhD; Megan Callanan

Abstract: Over 300,000 total hip arthroplasties (THAs), commonly known as total hip replacements, are performed in the US annually. As many as 30% of these procedures have outcomes associated with limb length discrepancy (LLD), where one leg is shorter than the other. Severe LLDs (over 1.5 cm) result in patients experiencing an abnormal gait, hip instability, and joint pain. Thus, severe LLDs generally result in a revision procedure. Over 30,000 THA revision procedures are performed each year in the US due to LLD, equating to almost $360,000,000 dollars of unnecessary healthcare expenditures annually. Existing solutions do not adequately address this problem, so orthopedic surgeons need a better method for assessing LLD during THA procedures. SymMEDtry is developing a device that uses a computer vision algorithm to provide intraoperative feedback on changes in leg length, enabling surgeons to minimize LLD at its onset.

Watch Video
PediaCORE PediaCORE: Providing Low-Cost and Accessible Physical Therapy Alternatives to Children with Cerebral Palsy

Team Members: Samiksha Ramesh, Akaash Sanyal, Nicholas Maritato, Teya Bergamaschi, Joshua Park, Maggie Li, Sundari Parise, Brian Gabriele, Everett Xu, Katharine Lee, Ricky Cheng

Advisors: Amy Bastian, PhD; Elizabeth Logsdon, PhD; Cari Sullivan, PT; Keith Slifer, PhD; Michael Browne, PhD

Abstract: Cerebral Palsy (CP) is the most common motor disability in the United States, with over 500,000 currently reported cases and 10,000 new diagnoses every year. Children with CP suffer neurological detriment which impairs muscle control, resulting in poor trunk stability. Trunk stabilityis an important prerequisite to many everyday functions, including sitting, standing, and walking, so children with CP experience a significantly reduced quality of life due to their condition. The standard of care for these patients is in-clinic physical therapy (PT), during which a therapist tailors the treatment to the child, adapting the exercises as the child progresses. However, there are several difficulties inherent to providing PT for young children with CP. The exercises mandated by PT often cause discomfort and fatigue. As a result, young children often do not cooperate with therapy. There is a need for a solution that can improve comfort and enjoyment to better motivate therapy. Virtual reality (VR) technology is an exciting and rapidly growing area of PT research. VR offers an entertaining and immersive experience that mitigates the discomfort caused by PT. Our solution mimics the aspects of VR that make it successful in PT while remaining safe for children to use. Our solution is a unique combination of novel software and hardware. We utilize a large, immersive, and interactive display along with highly customizable software to create an engaging and immersive therapy environment. Our software includes a variety of games that imitate specific PT exercises, each of which can be customized to appeal to children of varying interests and ages.
CottInSight CottInSight: Minimizing Cotton Ball Retention in Neurosurgical Procedures

Team Members: Raphael Bechtold, Benjamin Garlow, William Zhu, Renee Liu, Alexandra Szwec, Arushi Tandon

Advisors: Judy Huang, MD; Amir Manbachi, PhD; Collin Shale; Thomas Benassi; Micah Belzberg; Noah Gorelick; Brian Hwang; George Coles; Betty Tyler, Ian Suk, Henry Brem, MD

Abstract: Cotton balls are used in neurosurgical procedures to assist with hemostasis and improve vision within the operative field. Although the surgeon can reshape pieces of cotton for multiple intraoperative uses, this customizability and scale also places them at perpetual risk of being lost, as blood-soaked cotton balls are visually similar to raw brain tissue. Retained surgical cotton can induce potentially life-threatening immunologic responses, impair postoperative imaging, lead to a textiloma or misdiagnosis, and/or require reoperation. This study investigated three imaging modalities (optical, acoustic, and radiographic) to find the most effective method of identifying foreign bodies during neurosurgery.

First, we examined the use of dyes to increase contrast between cotton and surrounding parenchyma (optical approach). Second, we explored the ability to distinguish surgical cotton on or below the tissue surface from brain parenchyma using ultrasound imaging (acoustic approach). Lastly, we analyzed the ability of radiography to differentiate between brain parenchyma and cotton.

Our preliminary testing demonstrated that dark-colored cotton is significantly more identifiable than white cotton on the surface level. Additional testing revealed that cotton has noticeable different acoustic characteristics (eg, speed of sound, absorption) from neural tissue, allowing for enhanced contrast in applied ultrasound imaging. Radiography, however, did not present sufficient contrast, demanding further examination. These solutions have the potential to significantly reduce the possibility of intraoperative cotton retention both on and below the surface of the brain, while still providing surgeons with traditional cotton material properties without affecting the surgical workflow.

Watch Video
ObstetriCare ObstetriCare: Obstetric Fistula Management in Low-Resource Settings

Team Members: Erika Bhadra, Kevin Gorman, Pranavi Pallinti

Advisors: Chi Chiung Grace Chen, MD; Laura Keyser, DPT, MPH; Elizabeth Logsdon, PhD; Soumyadipta Acharya, MD, PhD; Tom Benassi; Sarah Lee; Bailey Surtees; Namratha Potharaj

Abstract: More than 2 million women in Sub-Saharan Africa and Asia suffer from an obstetric fistula, which is an abnormal hole that develops between the genital and urinary tracts as the result of prolonged, obstructed labor with the absence of timely effective medical care. The primary symptom of this condition is urinary incontinence, as urine is no longer contained in the bladder and instead flows freely through the vaginal canal. While urinary incontinence has many physical effects such as odor, skin irritation, and even nerve damage, this condition extends far beyond physical effects. These women are unable to perform daily life tasks, are ostracized by their families and communities, and often have increased rates of depression and suicide. Surgical treatment exists, but remains limited due to increased demand, a lack of trained surgeons and facilities, and cost barriers. Currently, humanitarian organizations are focusing their efforts on the front end by raising funds to go towards prevention efforts and surgical treatments. As those who remain untreated continue to experience severe urinary incontinence, an affordable, easy-to-use device thatdiscreetly contains and redirects unwanted leakage of fluids would temporarily mitigate the social distress associated with obstetric fistula until patients can afford to present for treatment. Our team is developing this solution with a specific need in mind: Women with obstetric fistulas in low- resource settings need a method to minimize uncontrolled urine flow to alleviate social distress caused by physical symptoms. Like many other organizations in this field, ObstetriCare supports aholistic approach to care. While our solution is management-based, we  acknowledge the need for both prevention and treatment efforts as well, and therefore hope to partner with fistula hospitals and humanitarian organizations to effectively distribute our product and get it into the hands of the end user. Though we hope that we can live in a world one day where obstetric fistulas no longer exist, for now our team is dedicated to easing the burden of this devastating condition.


Rehabilitation Engineering Design Projects


Team Members: Lucy Bian, Bailey Robertson, Jesse Wu, Zhiwei Yue

Advisors: Scott Paul, MD; Brian Morris; Brooke Hayley

Abstract: Osteoarthritis is a degenerative joint disease that affects over 30 million adults across the United States and is typically characterized by pain or aching in the joints, stiffness, decreased range of motion, or even muscle weakness. These impairments, when involving the hand and wrist, result in increased difficulty and pain when attempting to open food packaging such as vacuum sealed jars. Jar opening devices are available at kitchen stores because able bodied persons with the impairments associated with arthritis can have trouble opening jars. However, the current jar opening devices on the market are often insufficient to address the needs of persons with hand and wrist osteoarthritis. Existing devices can be too difficult for these persons to grip to open them, and don't offer a way to hold and stabilize the jar during the opening process. The HYDRiA aims to stabilize jars of various sizes, while maximizing efficiency of jar opening with minimal grip and torque from the user. Yet, the solution is completely body powered, can stabilize the jar without the user holding it, and not occupy too much counter space in the design kitchen.

Watch Video
TRACK-tile TRACK-tile

Team Members: Margaret Jukes, Nikhil Murty, Kalyn Younger, Valerie Zawicki

Advisors: Scott Paul, MD; Brian Morris; Elizabeth Russo

Abstract: Conditions involving the central nervous system can result in movement disorders that causeimpairments such as involuntary muscle spasms and other unwanted movements. In children,cerebral palsy (CP) is one of the most common conditions that result in movement disorders.The deficits in coordination of motor control prevent the normal acquisition of motor skills duringcritical periods of development in childhood. Fine motor control is a crucial component forlearning the skills needed for self-care and other activities of daily living.  Motor learningrequires frequent, correct repetition. We believe in using game therapy to augment the time thatchildren spend with rehabilitation professionals, enabling their further improvement in theabsence of the therapist’s absence. Current solutions focused on reducing impairments in earlyfine motor development do not explore the use of games in assisting routine clinical therapysessions. Children ages 3-5 with upper-limb motor dysfunction who have difficulty with finemotor control of their fingers need an at home method to help reduce motor impairments inisolating finger movements. Our team developed a game-based, biofeedback system to engageyoung children to perform the frequency of repetitive motor training necessary to reduce theirfine motor impairments.

Watch Video
StairWell StairWell

Team Members: Tre Crittendon, Varun Kedia, SJ Burris, Matthew Garza

Advisors: Scott Paul, MD; Brian Morris; Brooke Hayley

Abstract: More than 1.8 million individuals in the U.S. use walkers to assist with balance, stability, and pain management during walking. However, there is no current way to safely use the support of these walkers when ascending and descending stairs. Instead, walker users who encounter stairs need to rely on another person to provide support and balance, fold their walker up and carry it with them up the stairs, or, where available, use stair or floor lifts. If none of these options is viable, they are forced to altogether avoid areas accessible only by stairs. Our solution will allow walker users to seamlessly transition from walking on flat ground, to ascending or descending a staircase and back to flat ground. Our team has developed a design that allows walker users to easily adjust the height of the legs of the walker, allowing for stability while both ascending and descending stairs of variable height steps.

Watch Video
ClipZip ClipZip

Team Members: Matthew Lee, Anish Thyagarajan, Evelyn McChesney, Jackelyn Navar

Advisors: Scott Paul, MD; Brian Morris; Elizabeth Russo

Abstract: In the United States alone, 6.6 million people over the age of 20 are stroke survivors (Mozaffarian et al., 2016). Many of these persons deal with some form of chronic motor dysfunction in one of their upper limbs (Hatem et al., 2016) making activities of daily living difficult to perform independently. Among this population, the ability to operate dressing closures, specifically zippers, has been shown to be particularly difficult. The most difficult step is fastening together the two halves of the zipper (Green et al., 2011). Many assistive devices and technologies have been developed to help these individuals, however the solutions require the purchase of new clothing, costly alterations, and some devices are not even effective. Therefore, there is a need to develop a solution that allows for patients to zipper their existing clothing through an attachment or tool that integrates with clothing seamlessly and allows users to complete the task independently and quickly. Our device improves the ability to, with one hand, bring the sides of the jacket together, maintain this proximal distance, and allow for threading of the insertion pin into the slider. Ultimately, we hope that our device will allow persons with impairments in upper limb coordination to wear their own clothes, zip them quickly and efficiently and, most importantly, allow them to regain some freedom and independence.

Watch Video


Precision Care Medicine Design Projects

Team Unicorn Team Unicorn: Pre-Diagnosing ADHD from Diagnostic Questionnaires: Calculating Personalized ADHD Risk Scores from Patient Data

Team Members: Sindhu Banerjee, Indranuj Gangan, Aamna Lawrence, Stephen Li, Bhagyashree Maity

Advisors: Luther Kalb, PhD; Sridevi Sarma, PhD; Raimond Winslow, PhD; Joseph Greenstein, PhD; Hieu Nguyen, Han Kim

Abstract: ADHD is the most common pediatric psychiatric condition and has an estimated $42 billion annual cost. However, due to the complex nature of ADHD and its overlapping symptomology with other psychiatric diseases, it can be difficult to diagnose--taking providers up to 8 hours to diagnose one patient. Today there is no single conclusive test for identifying ADHD. Multi-modal evaluation methods, including clinical evaluation of the child’s motor skills, verbal skills, and diagnostic questionnaires, are commonly used.

Our team has utilized features derived from these diagnostic questionnaires and performed statistical modeling to generate individual patient risk scores for ADHD. Using machine learning methods, we have begun to identify the features most significantly associated with ADHD with the aim of aiding clinicians in diagnosing ADHD more efficiently during their clinical assessments and enabling a more personalized patient evaluation.

View Poster
Team Shiba Team Shiba: Predicting Hemorrhagic and Thrombotic Complications in Pediatric Patients on Extracorporeal Membrane Oxygenation Support (ECMO)

Team Members: Chunming Gu, Vy Tran, Yuqi Kang, Xuemin Zhu, Paige Epler

Advisors: Melania Bembea, MD; Sridevi Sarma, PhD; Joseph Greenstein, PhD

Abstract: ECMO is an advanced method of life support for pediatric patients with cardiac and/or pulmonary failure. In a recent multicenter study of pediatric ECMO, the cumulative proportions of hemorrhagic and thrombotic events were 70% and 37.5%, respectively. Early prediction of these complications is important for patient recovery. Using multivariable logistic regression models and random forest classification, our model attempts to determine the most relevant demographic, physiologic, laboratory, and transfusion-related risk factors for hemorrhagic and thrombotic complications in ECMO patients, using a time-series data set. In this study, we analyze electronic health record data from 169 pediatric patients (age 0-18 years) admitted to a single center between 2011–2018. This study demonstrates that comprehensive health data can be used to establish an effective prediction model for ECMO-related complications.

View Poster
Team Polar Bear Team Polar Bear: Predicting Respiratory Decompensation in Ventilated ICU Patients

Team Members: Zina Kurian, Akanksha Girish, Yvette Tan, Michael Young, Beini Hu

Advisors: Jim Fackler, MD; Jules Bergmann, MD, Raimond Winslow, PhD; Joseph Greenstein, PhD

Abstract: Mechanical ventilation is a critical and often life-saving treatment required by patients with respiratory failure. However, during mechanical ventilation, a patient is at risk for developing ventilator associated complications (VAC), requiring further escalation of respiratory support. In fact, 5-10% of patients on a ventilator will develop a complication. This complication can significantly increase a patient’s mortality risk and increases the risk of other comorbidity complications. Predicting respiratory decompensation early would allow early intervention and potentially improve patient outcomes, but currently there is no reliable method of predicting respiratory decompensation. Our group is developing a predictive model using physiological time series and electronic health record data, utilizing the Physionet MIMIC-III database. The cohort consists of n=7734 patients receiving extended mechanical ventilation, of which n=1733 experience a VAC and n=6001 do not. We will build a generalized linear model to predict a time-varying risk score of a patient’s risk of declining respiratory status. Our model with EHR data alone resulted in an AUC of 0.69 and accuracy of 0.79. The results highlight the value of identifying potential predictive features and the computational need to explore further model types. We anticipate that this clinically significant early warning of decompensation could trigger interventions to aid in clinical decision-making for improved patient outcome.

View Poster
Team Mola Mola Team Mola Mola: Advanced Risk Stratification and Prediction of Venous Thromboembolism in Critically Ill Patients

Team Members: Jinrui Liu, Akilan Meiyappan, Bronte Wen, Elizabeth Wu

Advisors: Adam Sapirstein, MD; Nauder Faraday, MD; Sachidanand Hebbar, PhD; Raimond Winslow, PhD; Joseph Greenstein, PhD; Hieu Trung Nguyen; Hanbiehn Kim

Abstract: Venous thromboembolism (VTE) is a preventable condition that includes both pulmonary embolism (PE) and deep venous thrombosis (DVT). VTE is responsible for more than 100,000 deaths a year in the U.S. Diagnosing VTE is particularly problematic in the intensive care setting. In spite of awareness and prophylaxis, studies show a VTE incidence of up to 10% in medical-surgical ICUs. Existing clinical scoring systems have poor sensitivity and rely on patient reported symptoms and routine vital signs measurements. Furthermore, treatment or prophylaxis with heparin carries significant risks, making accurate risk assessment crucial to balancing the risk-benefit ratio of intervention. Using high frequency physiologic data and information about underlying disease states, we are developing a novel risk assessment algorithm using machine learning methods to develop a more accurate, sensitive, and specific model for predicting VTE. Such a model could provide clinicians with more timely and reliable information to prompt definitive testing while decreasing unnecessary testing.

View Poster
Team GiraffeTeam Giraffe: Predicting Acute Kidney Injury after Elevated Risk Non-Cardiac Surgery: Acute Kidney Injury and Kidney Failure Prediction through Machine Learning Algorithms  Team Members: Jiayuan Kong, Ran Sui, Yinghe Sun, Tony Wei, Evan Yu

Advisors: Lee Andrew Goeddel, MD; Raimond Winslow, PhD; Joseph Greenstein, PhD; Han Kim; Hieu Nguyen

Abstract: Acute kidney injury (AKI) occurs when decreased renal perfusion or toxins cause glomerular or renal tubular damage. While often somewhat reversible renal injury can progress to renal failure requiring mechanical replacement of kidney function. AKI is associated with longer hospital stays and costs, higher incidence of chronic kidney disease, and increased hospitality mortality rates by up to ten fold. More than 10 million people worldwide suffer from AKI each year, of which between 30 to 40% occurs post-surgery.

We have data from a retrospective cohort of 10,123 surgery patients cared for at the Johns Hopkins Hospital from the aQI (anesthesia quality improvement) dataset. Features from aQI include demographic data such as age and race, time series data such as blood pressure and heart rate, and other parameters such as medications taken, lab results, and comorbidities.

We propose to create a machine learning-based platform to give real-time risk assessment of AKI, which would cut down on human resources costs, give clinicians the opportunity for early intervention, improve patient outcomes, and decrease health care costs.

View Poster
Team Cool Monkey Team Cool Monkey: Hypoxemia Prediction in the ICU: Predicting Hypoxemia--Low Blood Oxygen--to Enable Timely Intervention by Clinicians

Team Members: Zhenzhen Wang, Wen Shi, Chaoran Chen, Stephen Kyranakis, Ananya Swaminathan

Advisors: Timothy Ruchti, PhD; James Fackler, MD; Jules Bergmann, MD; Raimond Winslow, PhD; Joseph Greenstein, PhD; Hanbiehn Kim; Hieu Nguyen

Abstract: Hypoxemia is defined as an abnormally low level of oxygen in arterial blood. It is commonly diagnosed among general Intensive Care Unit (ICU) patients. Recent research has shown that moderate and severe hypoxemia are independently associated with higher mortality rates and an increase in the length of stay. Despite the severity and prevalence of hypoxemia in the ICU, current practices cannot accurately predict hypoxemic events and therefore do not provide clinicians enough time to prepare proper interventions that minimize patient mortality. If hypoxemia could be predicted 10 to 30 minutes in advance, clinicians and physicians would have sufficient time to administer necessary interventions. Our team proposes a machine learning-based hypoxemia alert system that predicts the onset of hypoxemia in the near future using patient vital signs and demographic data. This system could improve the efficiency and timeliness of treatment, thus improving ICU patient outcomes.

View Poster
Team Blue Jays Team Blue Jays: Delirium Prediction and Endotype Identification Using a Large Multi-Center ICU Database Team Members: Teya Bergamaschi, Kirby Gong, Joanna Guo, Ryan Lu, Akaash Sanyal

Advisors: Robert Stevens, MD; Jose Suarez, MD; Raimond Winslow, PhD; Joseph Greenstein, PhD; Sridevi Sarma, PhD; Hanbiehn Kim; Hieu Nguyen

Abstract: Delirium is an acute onset of global brain dysfunction. It is extremely prevalent in hospitals, particularly in ICUs, affecting up to 80% of critically ill patients and costing the American health care system $164 billion annually. Delirium is independently associated with poor health outcomes in patients and is often under-recognized and misdiagnosed. As 30-40% of delirium cases are considered preventable, a high performing prediction model would allow early intervention strategies to reduce patient risks of delirium. Additionally, given the variability in root cause and clinical trajectory, clustering delirium into endotypes can provide further insight into treatment options based on the underlying pathology.

We have developed a prediction model that uses the first 24 hours of data to estimate the risk of delirium during the ICU stay. Our preliminary model outperforms PRE-DELIRIC, the current gold-standard delirium prediction algorithm. We are extending this to develop a model with a real-time risk score for delirium. We are also using clustering to analyze the outcomes and features of patients, which may yield a better understanding of delirium physiology and trajectory.

View Poster
Team Alpaca Team Alpaca: AI-Driven Prognostication of Neurological Trauma Patients in the ICU: Leveraging ICU Data to Predict Short- and Long-Term Outcomes for Patients with Traumatic Brain Injury

Team Members: Anil Palepu, Jenna Ballard, Robert Li, Aditya Murali, Samiksha Ramesh

Advisors: Robert Stevens, MD; Jose Suarez, MD; Sridevi Sarma, PhD; Hanbiehn Kim; Hieu Nguyen; Joseph Greenstein, PhD

Abstract: Traumatic brain injury (TBI) is a leading cause of mortality and permanent neurological disability in the world, with over 50 million cases reported annually worldwide. Among TBI patients admitted to the Intensive Care Unit (ICU), 66% will die or suffer moderate-to-severe long-term neurological detriment within 6 months of discharge due to the lack of a reliable way to preempt patient decline. However, current models to predict a patient’s neurological recovery after TBI do not incorporate physiological time series data collected in the ICU. The aims of this project are two-fold: first, we would like to leverage features extracted from large, high-dimensional, and high-resolution data in a large, drivemulticenter ICU database to predict a patient’s neurological condition and survival at the end of their ICU stay. Second, we aim to determine to identify TBI disease subtypes and map them to specific recovery trajectories. These aims will be accomplished using both supervised and unsupervised machine learning algorithms.

View Poster


NeuroData Design Projects

Team Tealeaf Projection-Based Split Criteria for Multivariate Regression with Random Forest

Team Members: Vivek Gopalakrishnan, Jennifer Heiko, Suyeon Ju, Morgan Sanchez, Celina Shih

Advisors: Joshua Volegstein, PhD; Benjamin Pedigo; Jaewon Chung

Abstract: Random Forest (RF) is an interpretable and robust machine learning algorithm for classification and regression. However, existing RF methods are not well-equipped for multivariate regression tasks (predicting multiple continuous outputs from multiple inputs) due to the inherent challenges of variance in high-dimensional space. To address this issue, we introduce two projection-based split criteria: axis projection and oblique projection. For axis projection, rather than computing mean squared error (MSE) over all predictors and all samples, at each split, MSE is computed on a predictor chosen at random. The oblique projection split criterion splits based on MSE of a linear combination of predictors using weights 1.0 and -1.0 and 0.0. These new split criteria outperform all existing split criteria in the Scikit-Learn implementation of Random Forest (MSE, mean absolute error (MAE), and Friedman MSE) in several nonlinear simulation settings.
Team mvlearn mvlearn: The First Broad Multi-View Learning Package for Python

Team Members: Richard Guo, Ronan Perry, Gavin Mischler, Theodore Lee, Alexander Chang, Arman Koul, Cameron Franz

Advisors: Joshua Vogelstein, PhD; Benjamin Pedigo; Jaewon Chung

Abstract: In many datasets, there are multiple measurement modalities of the same subject, i.e. multiple X matrices (views) for the same class label vector y. For example, a set of diseased and healthy patients in a neuroimaging study may undergo both CT and MRI scans. Traditional methods for inference and analysis are often poorly suited to account for multiple views of the same subject as they cannot account for complementing views that hold different statistical properties. While single-view methods are consolidated in well-documented packages such as scikit-learn, there is no equivalent for multi-view methods. In this package, we provide a well-documented and tested collection of utilities and algorithms designed for the processing and analysis of Multiview data sets. Our Python package is pip-installable, open source, and free to use.
Team Mouse Multiscale Comparative Connectomics: From High-Dimensional Neuroimaging Data to Biological Insights

Team Members: Vivek Gopalakrishnan

Advisors: Joshua Vogelstein, PhD; Carey Priebe, PhD

Abstract: The connectome is a comprehensive map of the structural and functional connections of the brain, derived from multiple neuroimaging modalities. This information-rich representation of the brain can be used to better understand previously intractable neurological disorders. However, existing methods to analyze connectomes are limited: either they can only operate on a single connectome or, if they operate on a population of connectomes, they lack a statistical model which makes their validity unclear. To fill this need, we introduce a set of algorithms that enable the discovery of significant biological components at multiple levels of the connectome. We validated these algorithms using high-resolution connectome data collected from well-validated mouse models of autism. Armed with these algorithms, researchers can interrogate existing connectome data sets and discover new biologically relevant markers to aid the diagnosis and treatment of disease.
Team M2G M2G: Pipeline for Connectome Estimation

Team Members: Chaunkai Luo, Fang Cai, Zhen Hu, Jaylen Kang, Wilson Tang, Xuemin Zhu

Advisors: Joshua Vogelstein, PhD; Ross Lawrence; Alex Loftus

Abstract: Connectomes are brain mappings that allow us to understand the variability in brain connectivity and brain function. Currently, there is both a lack of standardization in and high barrier of expertise required for generating connectomes from diffusion Magnetic Resonance Imaging(dMRI) data. To address these issues, the team is continuing development on the M2G package as well as generating an open source collection of connectome datasets using already publicly available dMRI datasets. The tool is designed to be an easy to use, one click solution for a variety of human connectome estimation needs. We hope this tool and these datasets will help researchers push our understanding of the human brain further. The package is open source, free to use, and available on pip-install today.
Graph Statistics in the Brain Graph Statistics in the Brain: Bridging the Gap Between Graph Theory and the Brain

Team Members: Sambit Panda, Jayanta Dey, Ali Saad-Eldin, Gun Kang, Shan Qiu, Casey Weiner

Advisors: Joshua Vogelstein, PhD; Jaewon Chung; Benjamin Pedigo

Abstract: Inherent variability within a single network or  populations of networks is an increasingly desirable phenomenon to characterize. Implications of this can be used to improve diagnosis or discovery of possible causes for neurological disorders, improving artificial intelligence by modeling it on how the brain operates, and understanding how network distinctions among brains can manifest into phenotypic differences. To approach this problem, a connectome (a connected graph - directed or undirected - that models connections) can be constructed to show relationships between brain regions. The GraSPy python package has been created to tackle the problem of how to analyze these constructed connectomes. In our project, we added to the functionality of GraSPy with new tools for classifying graphs for multiple outcomes, graph matching, covariate assisted spectral embedding, & network classification using signal-subgraphs. We venture to ask such questions as ‘Does our brain connectivity influence our phenotypes?’ and ‘Do phenotypes embed something in common in terms of brain connectomes?’ If the answer is yes, then the question arises to what extent they are similar. Can we measure the similarities between two connectomes?
Brainlit Brainlit: A Python Package to Visualize and Analyze High-Resolution, Volumetric Brain Data

Team Members: Bijan Varjavand, Ryan Lu, Matthew Figdore, Alex Fiallos, Xi (Stanley) Wang, Victor Wang, Jenny Trieu, Sanika Phatak

Advisors: Joshua Vogelstein, PhD; Thomas Athey; Jaewon Chung; Benjamin Pedigo

Abstract: Understanding the structure of the brain can lead to advances in treating neurological disorders. Our team is developing Brainlit, a Python-based software library devoted to processing, analyzing, and conducting inference on large volumetric datasets. Research in fields such as neuroscience is trending more and more towards extremely big data. For example, recent imaging advances such as serial two-photon microscopy can image entire brains with submicron spatial resolution, sufficient to resolve and trace individual axons (Winnubst et al.). These huge terabyte-sized datasets come with myriad problems such as storage, access, and parallelization of algorithms. The Brainlit package provides easy-to-use utilities and algorithms for handling, visualizing, and analyzing large data. Brainlit is fully open-source and can be downloaded from the Python Package Index (PyPi). The documentation and all releases are available at


Independent Design Projects

Sun Jay Yoo sojourner: Design and Testing of a High-Performance Software Pipeline for Single Molecule Tracking, from Data Processing to Analysis

Team Members: Sun Jay Yoo

Advisors: Carl Wu, PhD; Amir Manbachi, PhD; Sarah Lee

Abstract: sojourner is a single-molecule tracking analysis software and graphical user interface (GUI) built as a R package. In order to allow users to derive causal relationships from single-molecule kinetics and build mechanistic cellular models from super-resolution fluorescent microscopy, sojourner aims to provide a full suite of analytic tools that turns tracking data into meaningful statistical results.

sojourner can first be used to read and process raw particle output from various localization software into a versatile format that can be further refined, saved, and analyzed. sojourner currently supports the Diatrack, MosaicSuite, and SLIMFAST software. Users are then granted tools such as trimming, filtering, linking, merging, and masking to remove noise and selectively choose target data. All track data can be dynamically visualized individually or as time-stacked videos to qualitatively assess macroscopic cellular dynamics. sojourner also supports a multitude of quantitative statistical tools to analyze single-molecule data effectively and accurately. Calculations such as the mean square displacement and the diffusion coefficient are supported along with other tools such as hidden Markov models, cumulative distribution function fits, normal distribution fits, dwell time, and kernel density. sojourner takes full advantage of high-performance, multi-cluster computing units and optimizes intensive operations for parallel computation for maximum efficiency in both the package and the GUI.

As an integrated software, sojourner aims to unify the many steps in the single-molecule tracking pipeline and provide a simple method for statistical analysis.

Watch Video
Shanelle Mendes A Gene Delivery Method for Treating Blindness in Neovascular Age-Related Macular Degeneration: Designing a Novel Polymer for Gene Delivery to Retinal Cells

Team Members: Shanelle Mendes

Advisors: Jordan Green, PhD; David Wilson, PhD; Yuan Rui; Peter Campochiaro, MD; Amir Manbachi, PhD; Sarah Lee

Abstract: Age-related macular degeneration (AMD) is one of the major causes of blindness among those over 50 in the United States, and affects over 170 million people worldwide. Over 1 million people in the United States suffer from neovascular AMD (NVAMD), which causes blindness at a higher rate than other forms of AMD. Existing therapies for NVAMD include antibody therapeutics like Avastin and Lucentis, which aim to eliminate the overgrowth of blood vessels in the retina through inhibition of vascular endothelial growth factor A (VEGFA). However, these therapies involve monthly intravitreal injections, which can be both time-consuming and uncomfortable for patients. An alternative approach to direct administration of antibody therapeutics is to deliver nucleic acids coding for VEGFA inhibitors. This gene delivery approach would allow for a more sustained production of VEGFA inhibitors, thus avoiding repeated treatments. In 2017, the FDA approved Luxturna, a retinal gene delivery drug. However, Luxturna costs $425,000 per eye. Furthermore, recombinant adeno-associated virus (rAAV) vectors like Luxturna have several drawbacks including a limited gene cargo capacity and risk of pre-existing immunity in patients. Non-viral, polymer-based vectors for gene delivery have much larger gene cargo capacities and have a lower chance of inducing an immune response. Here, we have designed a novel, non-toxic, and highly efficient, non-viral gene delivery vehicle for delivery to retinal cells.

Watch Video
Sanjana Eranki Designing a Bioadhesive Hydrogel for Drug Delivery to the GI Tract

Team Members: Sanjana Eranki

Advisors: Joshua Doloff, PhD; Gina Wang; Amir Manbachi, PhD; Sarah Lee

Abstract: Oral administration is currently the most common form of drug delivery to the gastrointestinal (GI) tract. This, however, presents numerous challenges as shown by the failure of efficient drug absorption. Thelimitations of oral drug delivery include low stability and solubility as the drug is passed through the body, which contributes to reduced permeability across intestinal membranes. Even when encapsulated into carrier systems, controlled release through the lining of the GI tract is difficult to accomplish due to the inability of these vehicles to penetrate the intestinal membrane barrier. Drug delivery through the outer lining of the lower GI tract presents a route for increased bioavailability and prolonged retention time, allowing greater amounts of the drug to reach systemic circulation. This method also provides the greatest contact of the drug with the absorption surface, improving the chances of permeability. Additionally, administration through the outer lining bypasses the harsh enzymes of the stomach. In order to achieve more targeted drug delivery to the lower GI tract, we aim to design a bioadhesive hydrogel that can encapsulate drug crystals for controlled release with high retention time. Because of their non-toxic quality, hydrogels are promising vehicles for drug delivery. Made out of modified natural materials, this adhesive hydrogel can be controlled to precise locations on the lower GI tract. Thus, this system reduces harm to surrounding tissue in the body while overcoming challenges presented by traditional methods of delivery.

Watch Video
Sandeep Kambhampati SpineSight: Using Doppler Ultrasound to Monitor Spinal Cord Blood Flow in Spinal Cord Injury Patients

Team Members: Sandeep Kambhampati

Advisors: Nicholas Theodore, MD; Amir Manbachi, PhD; Youseph Yazdi, PhD; Ian Suk; Sarah Lee

Abstract: Spinal cord injury (SCI) affects approximately 2.5 million people worldwide. Ensuring the recurrence of blood flow to the spinal cord post-injury can greatly improve patient recovery outcomes. Doppler ultrasound has been shown to be a promising imaging modality for monitoring spinal cord blood flow post injury; however, methods to quantify perfusion in real time from ultrasound data are lacking. To this end, a perfusion quantification algorithm was developed so physicians can quickly and reliably monitor SCI blood flow using Doppler ultrasound in order to manage patient recovery.

Watch Video
Michael LanNovel Conical Conduits for Size-Mismatched Nerves: Designing Nerve Guides to Reduce Secondary Pain After Peripheral Nerve Injuries Team Members: Michael Lan

Advisors: Hai-Quan Mao, PhD; Amir Manbachi, PhD; Alison Wong, MD; Sarah Lee

Abstract: Peripheral nerve injuries often result in excess axonal growth (neuromas) from the severed proximal nerve stump, and these neuromas can cause symptomatic pain. Although some surgical treatments exist to prevent neuroma formation, one of the most successful methods, Targeted Muscle Reinnervation, is limited due to the size mismatch between the severed proximal nerve stump and a donor nerve with which the stump will synapse. This project aims to design a conical conduit from electrospun polycaprolactone nanofibers with a nanofiber-hydrogel composite and chondroitin-sulfate proteoglycan gel filling to limit axonal growth and direct the proximal nerve endings to the donor nerve. This will account for the size-mismatch property between the two nerves during Targeted Muscle Reinnervation and prevent neuroma growth, relieving the patient of symptomatic pain after peripheral nerve injury.

Watch Video
Marion PangImproving Prostate Cancer Outcomes by Designing Nanoparticles with Enhanced Functionality: Nanoparticles that are More than the Sum of its Parts  Team Members: Marion Pang

Advisors: Hai-Quan Mao, PhD; Donald Zack, MD; Yizong Hu; Cynthia Ann Berlinicke; Amir Manbachi, PhD; Sarah Lee

Abstract: Prostate cancer is a highly prevalent and fatal disease – it is estimated that 1 in every 6 men would be diagnosed with the disease in their lifetime. The disease is the most common non-cutaneous cancer accounting for >1.3 million diagnoses and >10% of deaths globally in 2018. However, current forms of treatment such as radical prostatectomy and radiation therapy result in undesirable side effects such as urinary issues, erectile dysfunction and lifelong incontinence. As an alternative, applications of nanotechnology have enabled the development of nanoparticle (NP) therapeutic carriers suited for the delivery of chemotherapeutics in prostate cancer therapy. This treatment offers patients a safe and effective targeted therapy option that would preserve critical structures within the prostate, while avoiding undesired side effects. Here, we present a method to synthesize enhanced, non-toxic, biodegradable NPs to target and improve the efficiency of gene delivery in order to better treat prostate cancer by developing a high-throughput screening method to screen through large libraries of functionalized particles.

Watch Video
Kirby LeoA Bioadhesive for Diced Cartilage Glue Graft Applications: Developing an Accessible and Stiff Bioadhesive for Cartilage Grafts in Facial Plastics Team Members: Kirby Leo

Advisors: Patrick Byrne, MD; Nicholas Durr, PhD; Laura Ensign-Hodges, PhD; Andrew Lee, MD; Warren Grayson, PhD; Amir Manbachi, PhD; Sarah Lee

Abstract: Cartilage damage and imperfections in the facial region, especially the nose, drastically affect patient quality of life. Rhinoplasties are performed by facial plastic surgeons using cartilage autografts or allografts to augment the complex 3D nasal structure or contour. Before implantation, cartilage is formed into the desired graft geometry through various techniques of reforming the tissue. Dicing cartilage into <1mm pieces has recently been a technique of interest for nasal contour purposes due to the inherently formable tissue material it produces and markedly lower rate of postoperative complication. Despite these clinical benefits, adoption of the dicing technique to create cartilage grafts is challenged by the fragility of the resulting graft. Bioadhesives, such as fibrin sealants like TISSEEL, are commonly used to hold diced cartilage pieces together, but major limitations with current adhesives include cost and variable results. We propose a novel bioadhesive from accessible autologous and polymeric reagents. Results from materials testing indicate that our bioadhesive has stiffness-tuning properties that may be beneficial for creating various types of rhinoplasty grafts in the operating room and open opportunities to expand processed cartilage techniques to more complex applications.

Watch Video