2018
Global Health Innovations
-
THE TEAM
Student Team: Ailon Haileyesus, Ben Ostrander, Christine Diaz, Katie Solley, Yueheng “Dillon” Dou nemo Empowering Mothers
Advisors: Soumyadipta Acharya, PhD; Azadeh Farzin, MD; Christopher Golden, MD; Alain Labrique, PhD; Neha Goel; Youseph Yazdi, PhD
Sponsors: Bill and Melinda Gates FoundationABSTRACT
Approximately 2.9 million newborns die each year, with 75 percent of these deaths occurring during the first week of life. One approach to tackle this problem has been to send village health teams (VHTs) to the home to assess the neonate during this high-risk period. The VHT assesses the neonate based on the WHO guidelines: difficulty feeding, convulsions, chest indrawing, movement only when stimulated, respiratory rate greater than 60bpm, and temperature greater than 37.5°C or less than 35.5°C. Identification of even one of these danger signs is indicative of severe neonatal illness, and result in referral of the neonate to a facility. However, the limited number of village health teams and infrequency of visits create a bottleneck, and the timing of proper assessment aligning with the onset of symptoms is happenstance. During the first week when the VHT is not present, sick neonates may be identified too late to impact survival. By improving the frequency and quality of neonate assessment during the first seven days of life, our innovation aims to tackle preventable neonatal deaths caused by delayed identification of illness.
NeMo (neonatal monitoring) aims to task-shift neonatal assessment from WHTS to mothers, and provide tools for them to identify neonatal illness in the first week of life. The team has developed the NeMo system which consists of a smartphone application to guide the mother through the assessment based on the evidence-based WHO guidelines and an ultra low-cost, disposable sensor that measures the respiratory rate and temperature of the neonate. Our team has redesigned the NeMo app to be more user-friendly and incorporated back-end functionality to process sensor data. The sensors are embedded in the band, and an aux cord is used to connect the band to the smartphone. The mother will use the system every day for the first week of the neonate’s life, allowina for more freauent assessment of the neonate compared to the current standard of care. The team has conducted usability studies in rural villages of Iganga, Uganda to observe how the mothers interfaced with our system and to ensure a user-centric approach. Each of these studies resulted in informed design modifications and insights that the VHTs are equipped to successfully transfer their neonatal assessment roles to mothers. We have also significantly developed the sensor technology. The respiratory rate sensor can detect between 20-60 breaths per minute with a tolerance of +/ – 1 breath and the temperature sensor can detect with precision up to 0.1 degrees Celsius.
-
THE TEAM
Student Team: Conor Reid, Regina Cho, Jacob Caldwell, Jessica Powers, Emma Headley
Advisors:Soumyadipta Acharya, MD, PhD; Youseph Yazdi, PhD;
Jay Pasricha, MD; Ashish Nimgaonkar, MD; Mouen Khashab, MD; Vivek Kumbhari, MD; Nick Durr, PhD; Chad Weiler, PhD; Dawnielle Farrar-Gaines, PhD; Amol Bapaye, MD, DM; Aditya Polsani, BDS, MS; Sundeep Lakhtakia MD, DM; Mohan Ramchandani, MD, DM; Nageshwar Reddy MD, DM; Naresh Pagidimarry
Sponsors: Boston Scientific Endoscopy – Laura Christakis, Barry Weitzner, Tradd McCraryABSTRACT
With a population of 1.25 billion, 70 percent of patients living in rural villages, and a physician-patient ratio of 1:1674, there are distinct challenges for patient access to appropriate care in the Indian healthcare system. In partnership with Boston Scientific Endoscopy, last year’s team analyzed both the technical and systemic challenges associated with Endoscopic Retrograde Cholangiopancreatography (ERCP) in India in order to identify innovation targets that may increase patient access to ERCP in developing nations. Based on insights from over 40 interviews with physicians in both India and the Johns Hopkins Medical Institutions, they determined that a solution to facilitate the process of selective cannulation of the common bile duct during ERCP may have the highest clinical impact as it would reduce the training barriers for ERCP and increase the safety of the procedure. This year, we have focused on two mitigations to address the issue of selective cannulation: guidewire manipulation and visualization of the common bile duct.
ERCP entails navigating an endoscope into the duodenum and gaining access into hepatobiliary tree in order to treat various diseases. At the start of the procedure, the physician is tasked with threading a small guidewire (0.035″ diameter) through a muscular sphincter to then enter the pancreaticobiliary tree in a process called
“selective cannulation.” Once access has been established, physicians are able to perform therapies such as retrieving obstructing gallstones, expanding tissue strictures, or placing a stent to relieve obstruction from malignancies. The most feared complication of this procedure is pancreatitis, which is a life-threatening condition that occurs in five percent of cases1 and is in part due to multiple failed cannulation attempts leading to tissue irritation. The high degree of technical skill and precision required for this procedure creates a steep learning curve, and it has been cited to require up to 400 cases for a physician to achieve an 80 percent success rate in selective cannulation.Our team is developing solutions to deskill the process of selective cannulation by addressing the two greatest challenges that our stakeholders have cited for performing
ERCP: (1) visualization of the orientation of the common bile duct behind the papillary opening, and (2) increased control of the accessories (guidewire and sphincterotome) so as to increase their ability to enter the common bile duct. Proof-of-concept has been developed through benchtop, ex-vivo, and porcine models for the solutions allowing the team to make informed design decisions. With promising feedback from end-users, the team believes that the solutions being developed will positively impact therapeutic endoscopy by reducing training barriers for physicians and making safe ERCP more universally accessible for patients in the developing world.
-
THE TEAM
Student Team:Jewell Brey, Tristan Ford, Margaret Glancey, MSPH, Adam Goodwin, Collyn Heier, Laura Scavo
Advisors:Soumyadipta Acharya, MD, PhD; Sylvia Hinrichsen, MD, PhD; Kimber Ashman, MSE; Angela Harris, PhD; Yvonne Linton, PhD; Teja Maruvada, MSE; Benedict Pagac; Austin Reiter, PhD; Chandrakant Ruparelia, MBBS, MPH; Alexandra Spring, MS
Sponsors: United States Agency for International Development (USAID)ABSTRACT
Despite public health systems making great strides toward reducing mosquito-borne disease, mosquitoes remain the deadliest animal in the world and are responsible for several recent disease outbreaks, including the 2015/16 Zika epidemic in Brazil. Mosquito surveillance, the regular monitoring of mosquito populations throughout a season, is critical to guide effective methods of mosquito control. However, mosquito surveillance remains logistically challenging and highly human-resource intensive.
VectorWeb, an innovative, intelligent mosquito surveillance system, has the potential to transform the field. We have developed a suite of technologies for rapid vector identification to automate the vector surveillance process.
This provides existing surveillance efforts with real time data at no additional labor cost. Using our core software platform capable of counting and identifying mosquitoes, we have created VectorEyes, a system of commercial mosquito traps outfitted with cameras and image transmission capabilities over local cellular networks. Images are sent to a cloud-based server, where imaged mosquitoes are counted by an algorithm and the data logged on our proprietary dashboard, known as VectorWeb. We also provide VectorSnap, a low-cost cell phone app that allows users to submit images of mosquitoes for automated identification in the absence of a smart trap.
Preliminary user testing has been conducted with municipal health workers in Brazil, confirming usability of the VectorEyes hardware and providing insights for further development. When tested on field-collected mosquitoes, our novel algorithm currently achieves an average of 74.3% +/- 4.3% sensitivity and 97.1% +/- 0.6% specificity across eight mosquito species, including the primary disease-carrying vectors in Brazil. A survey of 1,000 Brazilian citizens conducted by the Boston Consulting Group confirmed need and market viability for the VectorWeb system.
-
THE TEAM
Student Team: Andrew Eisenthal, Shashwat Gupta, Sabrina Liu, Edward Ruppel, Wade Schutte
Clinical Mentors:
Advisors: Youseph Yazdi, PhD; Hari Tandri, MD; Hans Lee, MD; Diana Yu, MD; Fabrizio Assis, MD; Zoar Engelman, Phd; Anisha Bapna; Howard Levin, MD; Mark Gelfand; Jie Wang, MDABSTRACT
The Coride/Symap team investigated novel device solutions to change the lives of cardiac, pulmonary, and renal patients that have failed drug treatment. Utilizing China as a target market, the team traveled to multiple cities throughout China to interview physicians, medical device companies, and regulatory officials to look at the possibility of starting a venture based in China. This included visiting four top-tier hospitals to find suggestions for potential targets and isolate the patient care and referral pathway. Visiting various staged medical device companies provided insight into the market landscape, current trends, and strategy on how to navigate the Chinese regulatory system.
Throughout the year, the team researched and developed solutions for treating the therapeutic target. The devices and targets have been continuously tested in animal studies for preclinical validation. Coridea and Symap aim to continue development of the proposed devices with an eye towards commercialization through the target markets.
Advanced Healthcare Innovations
-
THE TEAM
Student Team: Conor Reid, Regina Cho, Adam Goodwin, Jessica Powers, Collyn Heier
Advisors:Soumyadipta Acharya, MD, PhD; Youseph Yazdi, PhD; Clifford Weiss, MD; Christopher VandenBussche, MD, PhD; David Feller-Kopman, MD; Wotek Mydlarz, MD; Austin Reiter, PhDABSTRACT
Fine needle aspirations (NAs) are a minimally invasive biopsy method and a staple in the screening process for almost all suspected cancers. Over 2.2M FNA biopsies occur each year, but in up to 54% of cases, these biopsies have low diagnostic yield because an inadequate volume of target cells are obtained. The resulting uncertainty leads to repeat biopsies and even tissue excision. To address this problem, a separately reimbursed service exists called Rapid On-Site Evaluation (ROSE), in which a pathologist examines the sample under a microscope in the OR and verifies that it contains adequate cell volume while the patient is still on the operating table. Rapid on-site evaluation can lead to an 80% reduction in repeat biopsies, but only large academic centers with high procedure volume and personnel can afford it. Telepathology products that attempt to solve this problem have significant technical and usability issues that hinder adoption, including requiring trained operators in the OR and being limited by poor image resolution. Thus, there is an opportunity to provide the value of ROSE for those who currently cannot utilize it.
To address this opportunity, CytoSage is developing PathCom-a device that automates ROSE so its benefit can be realized by those who don’t have a trained pathologist to offer the service. PathCom consists of an imaging system that digitizes slides of FNA biopsy samples, analyzes their cellular content, and determines if the sample is “adequate,” or has sufficient cellularity. PathCom addresses critical usability and technical shortcomings of existing telepathology devices by removing the need for a trained specialist in the OR, imaging at high magnification, and integrating smoothly with procedural workflow. PathCom will take an input of cellular aspirate directly from the proceduralist. The system will involve a mechanical system that will prepare and stain a slide from this aspirate, eliminating the need for a technologist or trained professional to be present for slide preparation, which has been identified as a key usability necessity. The system will then scan and analyze the slide, searching for areas of interest which may indicate adequacy either due to presence of abnormal cells or a large number of lymphocytes, depending upon location. When adequacy (or inadequacy) has been determined, the determination will be communicated to the proceduralist via audio/ visual cues, allowing them to make informed decisions as to whether or not to take more sample from a given biopsy location. With PathCom, the benefits of ROSE can be offered to patients and proceduralists who are not currently able to access it.
-
THE TEAM
Student Team: Ailon Haileyesus, Benjamin Ostrander, Christine Diaz, Katie Solley, Yueheng “Dillon” Dou
Advisors: Kunal Parikh, PhD; Soumydipta Acharya, MD, PhD; Youseph Yazdi, PhD, MBA; Sudeep Pramanik, MD, MBA; Michael Boland, MD; Laura Kline; George ColesABSTRACT
Eyedea is focused on redefining corneal transplantation and restoring vision around the globe. In the U.S. alone, 4.7 million Americans have a disease called Fuchs’ Dystrophy, which slowly robs people of their vision. This disease can be treated by corneal transplantation, yet only two percent of Americans with Fuchs’ receive a transplant. In the last six years, a procedure called Descemet’s Membrane Endothelial Keratoplasty (DMEK) has gained traction and improved outcomes. Although this is by far the best procedure for treating Fuchs’ Dystrophy, it is still performed relatively rarely.
A key reason DMEK is not performed as often is the processing of donor tissue at eye banks by eye bank technicians. Current processing techniques rely on manual tools insufficient for the task; this process is technically challenging, inconsistent, and has significant room for error, typically taking up to 60 minutes.
Eyedea’s device streamlines tissue preparation by standardizing the most difficult steps in the process with a one-time-use device allowing technicians to repeatedly and reliably process the tissue for surgeons. Our technique reproduces the gold standard processing method and allows anyone to be able to do the task. The device has been tested on both rabbit and human eye models, providing key insights that allowed for iteration of the device. Initial testing has proven conceptual verification for multiple solutions as we are creating a multi-device system for eye banks. Through our partnership with KeraLink, the team is planning to conduct usability studies to further validate that this device will fit into the work flow.
-
THE TEAM
Student Team: Andrew Eisenthal, Edward Ruppel, Sabrina Liu, Shashwat Gupta, Wade Schutte
Advisors: Ashish Nimgaonkar, MD; Cliff Weiss, MD; Robert Liddell, MD;
Soumyadipta Acharya, MD, PhD; Hai-Quan Mao, PhD; Chris Jeffers, PhD; Brandon Craft; Paul FearisABSTRACT
For patients with a suspicious pulmonary lesion found on chest radiography, further information is required to properly drive management.
For many patients, the next step involves obtaining a tissue specimen, often times via CT-guided percutaneous transthoracic needle biopsy.
Approximately 400,000 CT-guided lung biopsies are performed each year in the U.S. Furthermore, there is an estimated 1.1 million CT-guided lung biopsies performed globally due to the rising incidence of lung cancer from smoking/pollution and better imaging. This is typically an outpatient procedure, performed either by an interventional radiologist or by a standard diagnostic radiologist. The number one complication from the procedure is a collapsed lung (pneumothorax) which occurs when air leaks in the space between the chest wall and the lung. Side effects can include labored breathing and chest pain. This complication occurs up to 30 percent of the time and turns a same-day outpatient procedure into a multi-day hospitalization in which the patient is connected to a chest drain and vacuum to withdraw the excess air.
With the cost of treating a pneumothorax ranging up to $11,000, this places a large financial burden on the healthcare industry. There is only one FDA approved solution which only attempts to occlude needle tract after the biopsy has occurred, resulting in a mere 50 percent efficacy.
However, because up to 75 percent of pneumothoraces occur during the procedure, a solution to this problem must block all sources of air before the biopsy occurs, including inside the lung and the outside environment.
Furthermore, the current solution has barriers to adoption such as ease-of-use, workflow integration, addition of time, and cost.
With the current shortcomings, PneumoNix Medical proposes a simple, cost-effective solution which can effectively block all sources of air before the biopsy occurs without adding time to the procedure. The device consists of a smart needle delivery system coupled with a proprietary biosealant. The device will improve on the efficacy of the current FDA approved solution while ensuring trust and putting control back into the physician’s hands, allowing patients to undergo this procedure without risk of complications.
-
THE TEAM
Student Team:Jewell Brev, Jacob Caldwell, Tristan Ford, Emma Headley, Laura Scavo
Advisors:Soumyadipta Acharya, MD, MSE, PhD; Youseph Yazdi, MS, PhD, MBA; Lisa Maragakis, MD, MPH; Simon Mathews, MD; Vinciya Pandian, MSN, CRNP;
Austin Reiter, MA, PhD; Joshua Budman, MSE; Neha Goel, MS; Bashar Safar, MBBSABSTRACT
Each year there are over 150,000 surgical site infections (SSI) in the United States alone. 60 percent of these SSis appear after discharge, but patients are unprepared to accurately monitor their own surgical sites when they leave the hospital. This results in patients either not calling their providers when they should, or taking alternative action that doesn’t result in proper follow-up. Even when patients do call their providers to describe a concerning symptom, one-in-three SSis will not be identified due to inadequate information, resulting in the patient not being brought in for evaluation. This delay in treatment can lead to poor patient outcomes and an increased risk of readmission, the cost of which is not reimbursed and costs hospitals $3.3 billion annually.
SurgiSight is developing a surgical wound management system to facilitate post-discharge communication between patient and provider, and to inform clinical decisions through relevant sign and symptom measurement. SurgiSight is a system comprising of an easy-to-use, patient-facing application, our specialized processing algorithms, and an EMR compatible provider report. The mobile app allows the patient to take pictures of their surgical site to capture visual signs, such as redness, as well as answer survey questions that capture systemic responses and symptoms, such as pain and fever. The collected images will then be measured using our insight-driven algorithms to provide a quantitative evaluation that will track the surgical site over time. This will be done by deploying unsupervised learning techniques to effectively analyze incision sites for spreading redness, swelling, and dehiscence. These results will then be transferred to the EMR system used by the hospital. On the provider side, measurements will be shown in conjunction with the patient’s images and survey responses. Additionally, rather than review this data each time it is collected, providers will be alerted of a concerning increase in any one symptom. Otherwise, the report is available for them to review at any time in the EMR system.
Our team has validated the need for our product through interviews with over 40 key stakeholders as part of the National Science Foundation (NSF) I-Corps program.
This includes 13 surgeons, 12 nurses, and 12 managers or administrators within the hospital system. We have also reached back out to various stakeholders during the development process as part of ongoing validation of our prototype. These stakeholders have emohasized need for comparable sian and smotom development over time, clear usable information from patients, and improved patient-clinician communication, which are addressed by our solution. Moving forward we will be collecting feedback from patients to validate our design, gathering wound images to improve our algorithm, and developing a usable mobile app platform for patient use.
