ASSET Center Inaugural Seed Grants Will Fund Trustworthy AI Research in Healthcare

by

Illustration credit: Melissa Pappas

Penn Engineering’s newly established ASSET Center aims to make AI-enabled systems more “safe, explainable and trustworthy” by studying the fundamentals of the artificial neural networks that organize and interpret data to solve problems.

ASSET’s first funding collaboration is with Penn’s Perelman School of Medicine (PSOM) and the Penn Institute for Biomedical Informatics (IBI). Together, they have launched a series of seed grants that will fund research at the intersection of AI and healthcare.

Teams featuring faculty members from Penn Engineering, Penn Medicine and the Wharton School applied for these grants, to be funded annually at $100,000. A committee consisting of faculty from both Penn Engineering and PSOM evaluated 18 applications and  judged the proposals based on clinical relevance, AI foundations and potential for impact.

Artificial intelligence and machine learning promise to revolutionize nearly every field, sifting through massive amounts of data to find insights that humans would miss, making faster and more accurate decisions and predictions as a result.

Applying those insights to healthcare could yield life-saving benefits. For example, AI-enabled systems could analyze medical imaging for hard-to-spot tumors, collate multiple streams of disparate patient information for faster diagnoses or more accurately predict the course of disease.

Given the stakes, however, understanding exactly how these technologies arrive at their conclusions is critical. Doctors, nurses and other healthcare providers won’t use such technologies if they don’t trust that their internal logic is sound.

“We are developing techniques that will allow AI-based decision systems to provide both quantifiable guarantees and explanations of their predictions,” says Rajeev Alur, Zisman Family Professor in Computer and Information Science and Director of the ASSET Center. “Transparency and accuracy are key.”

“Development of explainable and trustworthy AI is critical for adoption in the practice of medicine,” adds Marylyn Ritchie, Professor of Genetics and Director of the Penn Institute for Biomedical Informatics. “We are thrilled about this partnership between ASSET and IBI to fund these innovative and exciting projects.”

 Seven projects were selected in the inaugural class, including projects from Dani S. Bassett, J. Peter Skirkanich Professor in the Departments of Bioengineering, Electrical and Systems Engineering, Physics & Astronomy, Neurology, and Psychiatry, and several members of the Penn Bioengineering Graduate Group: Despina Kontos, Matthew J. Wilson Professor of Research Radiology II, Department of Radiology, Penn Medicine and Lyle Ungar, Professor, Department of Computer and Information Science, Penn Engineering; Spyridon Bakas, Assistant Professor, Departments of Pathology and Laboratory Medicine and Radiology, Penn Medicine; and Walter R. Witschey, Associate Professor, Department of Radiology, Penn Medicine.

Optimizing clinical monitoring for delivery room resuscitation using novel interpretable AI

Elizabeth Foglia, Associate Professor, Department of Pediatrics, Penn Medicine and the Children’s Hospital of Philadelphia

Dani S. Bassett, J. Peter Skirkanich Professor, Departments of Bioengineering and Electrical and Systems Engineering, Penn Engineering

 This project will apply a novel interpretable machine learning approach, known as the Distributed Information Bottleneck, to solve pressing problems in identifying and displaying critical information during time-sensitive clinical encounters. This project will develop a framework for the optimal integration of information from multiple physiologic measures that are continuously monitored during delivery room resuscitation. The team’s immediate goal is to detect and display key target respiratory parameters during delivery room resuscitation to prevent acute and chronic lung injury for preterm infants. Because this approach is generalizable to any setting in which complex relations between information-rich variables are predictive of health outcomes, the project will lay the groundwork for future applications to other clinical scenarios.

Read the full list of projects and abstracts in Penn Engineering Today.

Erin Berlew and Rhea Chitalia Receive Solomon R. Pollack Awards for Excellence in Graduate Bioengineering Research

The Solomon R. Pollack Award for Excellence in Graduate Bioengineering Research is given annually to the most deserving Bioengineering graduate students who have successfully completed research that is original and recognized as being at the forefront of their field. This year Penn Bioengineering recognizes the outstanding work of two graduate students in Bioengineering: Erin Berlew and Rhea Chitalia.

Erin Berlew, Ph.D. candidate in Bioengineering

Erin Berlew is a Ph.D. candidate in the lab of Brian Chow, Associate Professor in Bioengineering. She successfully defended her thesis, titled “Single-component optogenetic tools for cytoskeletal rearrangements,” in December 2021. In her research, she used the BcLOV4 optogenetic platform discovered/developed in the Chow lab to control RhoGTPase signaling. Erin earned a B.S. in Chemistry from Haverford College in 2015 and was an Americorps member with City Year Philadelphia from 2015-2016. “Erin is a world-class bioengineering with an uncommon record of productivity gained through her complementary expertise in molecular, cellular, and computational biology,” says Chow. “She embodies everything wonderful, both academically and culturally, about our graduate program and its distinguished history.” Erin’s hobbies outside the lab include spending time with family, reading mystery novels, enjoying Philadelphia, and crossword puzzles. In the future, she hopes to continue to teach for the BE department (she has already taught ENGR 105 and served as a TA for undergraduate and graduate courses) and to conduct further research at Penn.

Rhea Chitalia, Ph.D. candidate in Bioengineering

Rhea Chitalia is a Ph.D. candidate in Bioengineering and a member of the Computational Biomarker Imaging Group (CBIG), advised by Despina Kontos, Matthew J. Wilson Associate Professor of Research Radiology II in the Perelman School of Medicine. Rhea completed her B.S.E. in Biomedical Engineering at Duke University in 2015. Her doctoral research concerns leveraging machine learning, bioinformatics, and computer vision to develop computational imaging biomarkers for improved precision cancer care. In December 2021 she successfully defended her thesis titled “Computational imaging biomarkers for precision medicine: characterizing intratumor heterogeneity in breast cancer.” “It has been such a privilege to mentor Rhea on her dissertation research,” says Kontos. “Rhea has been a star graduate student. Her work has made fundamental contributions in developing computational methods that will allow us to gain important insight into tumor heterogeneity by utilizing a multi-modality imaging approach.” David Mankoff, Matthew J. Wilson Professor of Research Radiology in the Perelman School of Medicine, served as Rhea’s second thesis advisor. “It was a true pleasure for me to work with Rhea and to Chair her BE Thesis Committee,” Mankoff adds. “Rhea’s Ph.D. thesis and thesis presentation was one of the best I have had the chance to be involved with in my graduate mentoring career.” After graduation, Rhea hopes to further precision medicine initiatives through the use of real world, multi-omic data in translational industry settings. She will be joining Invicro as an Imaging Scientist. In her spare time, Rhea enjoys trying new restaurants, reading, and spending time with friends and family.

 

Hao Huang Named AIMBE Fellow

Hao Huang, Ph.D.

Hao Huang, Research Associate Professor of Radiology in the Perelman School of Medicine and member of the Penn Bioengineering Graduate Group, has been named an American Institute for Medical and Biological Engineering (AIMBE) Fellow.

Election to the AIMBE College of Fellows is among the highest professional distinctions accorded to a medical and biological engineer. “The College of Fellows is comprised of the top two percent of medical and biological engineers in the country. The most accomplished and distinguished engineering and medical school chairs, research directors, professors, innovators, and successful entrepreneurs comprise the College of Fellows. AIMBE Fellows are regularly recognized for their contributions in teaching, research, and innovation.”

Huang was “nominated, reviewed, and elected by peers and members of the College of Fellows for contributions to the development and applications of innovative MR methods to study the developing brain.”

A formal induction ceremony will be held during AIMBE’s virtual 2021 Annual Event on March 26. Huang will be inducted along with 174 colleagues who make up the AIMBE Fellow Class of 2021.

Read the full press release.

Week in BioE (April 24, 2018)

Pushing the Limits of Imaging

7T-MRI
An image showing 7-tesla MRI of the human brain

Since the late 1970s with the advent of computed tomography (CT), medical imaging has grown exponentially. Magnetic resonance imaging (MRI) offers some of the clearest pictures of human anatomy and pathology, particularly as the strength of the magnetic field used (measured in units called Teslas) increases. However, MRI machines are expensive, and the costs increase as one uses a machine with higher field strength to ‘see’ the human more closely. Therefore, it is often more useful (and certainly less expensive) to modify existing MRI technology on hand, rather than acquire a new machine.

A recent example is the work of Tamer Ibrahim, PhD, Associate Professor of Bioengineering at the University of Pittsburgh. Dr. Ibrahim used a series of multiple NIH grants to develop a coil system for Pitt’s 7T-MRI — one of only approximately 60 worldwide — enabling it to more accurately image the brain’s white matter. Dr. Ibrahim is interested in seeing how hyperintensity in the white matter is related to depression, which is one of the highest-burden but least-discussed diseases in the world. Called a “tic-tac-toe” radiofrequency coil setting, the device that Dr. Ibrahim created is a network of antennas fitted to the head that minimize concerns such as coil heating and radiofrequency intensity losses, as well as safety concerns.

Dr. Ibrahim has more NIH funding on the way to continue optimizing his device and apply it in other psychiatric and neurological disorders. Rather than purchasing a new MRI machine with higher field strengths to achieve this image quality, Dr. Ibrahim’s coil design can be used on existing machines. One possible outcome is more clinicians using this new coil to study how changes in the brain’s white matter structure occur in a broad range of brain diseases, leading to both earlier detection anfor ad more effective treatment.

Smart Shunt for Hydrocephalus

Hydrocephalus, once more commonly known as “water on the brain,” is a condition marked by abnormal accumulation of cerebrospinal fluid (CSF) in the skull. If unchecked, the accumulation of fluid will create dangerous pressures in the brain that can result in brain damage. Hydrocephalus occurs in one in every 1,000 births, and nearly 400,000 adults in the US suffered at least on episode of hydrocephalus. For both infants and adults, hydrocephalus is often treated surgically with the installation of a shunt to channel the excess CSF out of the cranium. These shunts are simple but effective devices that operate mechanically. However, since they’re entirely mechanical, they fail over time. Being able to determine that such a failure was imminent could allow patients to receive a replacement shunt before complications arise.

To meet this clinical need, a group of scientists at the University of Southern California (USC)  updated existing shunt systems with microsensing technology, creating a “smart shunt” that can tell clinicians how an installed shunt is functioning and alert the clinician that a replacement is needed. The group, including Ellis Fan-Chuin Meng, PhD, Gabilan Distinguished Professorship in Science and Engineering, Dwight C. and Hildagarde E. Baum Chair, and Professor of Biomedical Engineering and Electrical Engineering-Electrophysics, has created a start-up called Senseer to produce these smart shunts.

The shunt currently measures pressure, flow, and occlusion using miniature microelectronics sensors. If device approval comes, the company hopes to move on to developing smart sensors for other organ systems.

DNA-based Drug Testing

Drug and alcohol testing is a controversial topic, partly because of the balance between individual rights to use legal drugs and potential for societal harm if these drugs are abused or if patients transition into illegal drug use and dependence. Inventing technology to determine when, and how much, a person has been drinking or using drugs (including tobacco) would probably increase, rather than decrease, the controversy involved in the topic.

New technology reported recently adds a new element to this discussion. According to Robert Philibert, MD, PhD, Professor of Psychiatry at the University of Iowa and an adjunct faculty member in the Department of Biomedical Engineering, his company’s tests, which rely on epigenetic markers of substance use, could be used, for example, to inform a primary care physician about the actual history of substance use, rather than relying solely on patients’ self-reported use.

Dr. Philibert’s tests are currently pending approval by the Food and Drug Administration. Marketing for the products will begin in the coming weeks.

People and Places

Recognizing the changing priorities in engineering and the growing role of data sciences, Boston University has decided to adapt its curriculum by adding data science requirements for all majors. According to John White, PhD, Chair of the Department of Biomedical Engineering, “Advances in data sciences and computing technology will allow us to make sense of all these data.”

The Biomedical Science Program at Howard Payne University in Brownwood, Texas, has received a $200,000 grant from  the James A. “Buddy” Davidson Charitable Foundation to endow a scholarship in Davidson’s name, as well as to refurbish the program’s Winebrenner Memorial Hall of Science.

Finally, we offer our congratulations this week to James C. Gee, PhD, Professor of Radiologic Science in Radiology at the University of Pennsylvania’s Perelman School of Medicine and a Graduate Group faculty member in Penn’s Department of Bioengineering.  Dr. Gee was named a fellow of the American Institute for Medical and Biological Engineering.