Tyler O'Neal, Staff Editor HEALTH

Tandon-led initiative will help healthcare practitioners adopt new data-intensive technologies quicker, easier

A multi-disciplinary, multi-school team of researchers are reimagining the way that new inclusive healthcare technologies are put into work.

How will rapidly expanding health data-intensive technologies affect the future of healthcare work? A team of NYU researchers from the schools of engineering, medicine and business led by Professor Oded Nov of NYU Tandon are conducting a broad investigation into how to best bring inclusive tech into the clinic, empowering healthcare workers to take advantage of data-driven research and improve health outcomes for patients. 

The problem that the team is taking on is the disconnect between the status quo for healthcare practice that nurses, physician assistants, allied health staff, etc. are already familiar with, and the ways these practices are altered by advanced technologies. Particularly troublesome is the new reliance on big data, which in such vast quantities can burden practitioners who are not used to working with it.

The team recently received a $2.5 million National Science Foundation (NSF) grant to pursue research focused on the growth of data-intensive technologies in healthcare, including telehealth and artificial intelligence (AI) based tools. The new grant reflects a continued investment of the NSF in NYU’s digital health work initiative.

“The new grant will help us further develop our NYU-wide research program on digital health work as an interdisciplinary research domain that brings together technological, organizational and medical innovations toward a healthy and resilient society, and an inclusive healthcare workforce,” said Nov.

The project’s approach centers on alleviating misalignment between current healthcare work and data-intensive technologies, focusing on three areas:

  • Co-developing tools and generalizable design principles with users that lower the barriers to technology integration for healthcare workers 
  • Empowering individuals within healthcare systems who have diverse roles to adopt and use the tools and improve their skills 
  • Enabling patient-centered healthcare that promotes autonomy and strengthens clinician-patient concordance 

While new technologies are constantly being developed, the hardest part to making sure they work is the “last mile” — a socio-technical challenge that involves getting the right technologies matched with the right interfaces into the hands of diverse healthcare workers, and creating alignments between workflows, organizations, and technologies. 

For example, a nurse may have access to technology that allows them to remotely monitor the vitals of a home-bound patient, over long periods of time. Packaging the tracked data and presenting it interpretably in the context of the nurse’s workflow could be helpful in identifying and solving potential health problems before they escalate, and empower an increasingly diverse and overburdened non-physician healthcare professionals.

The new grant is part of the NSF Future of Work at the Human-Technology Frontier, one of the foundation’s 10 Big Ideas that covers evolving technologies that are actively shaping the lives of workers and how people in turn can shape those technologies, especially in the world of work, The initiative brings together NSF research communities to conduct basic scientific research on the interaction of humans, society, and technology that will help shape the future of work to increase opportunities for workers and productivity for the American economy.

The team, comprising researchers with varied skills and expertises, includes investigators Devin Mann of NYU Grossman School of Medicine, and Batia Wiesenfeld of NYU Stern, bringing together multiple schools into one project. The full team includes Rumi Chunhara, Maurizio Porfiri, and Graham Dove from the Tandon School of Engineering; Antoinette Schoenthaler, Joseph Ravenell, Katharine Lawrence, Olugbenga Ogedegbe, and Yin Aphinyanaphongs from the Grossman School of Medicine; and John-Ross Rizzo, who has appointment at both schools.

Penn Medicine wins $6 million to advance understanding of human genome function in health, disease

Tyler O'Neal, Staff Editor HEALTH

The National Institutes of Health (NIH) has selected Penn Medicine as one of 25 award recipients across 30 sites in the United States to serve as Impact of Genomic Variation on Function (IGVF) investigators, with the goal of better understanding how genetic differences impact how human genes function, and how these variations influence human health and disease. Funded by the NIH’s National Human Genome Research Institute (NHGRI), Penn Medicine will be awarded more than $1.2 million per year, with a contract that is expected to be supported for five years, totaling more than $6 million in funding for this research.

Whole-genome sequences among people are more than 99.9 percent identical, it’s the 0.1 percent of differences, alternate orders of the As, Cs, Gs, and Ts that make up DNA, combined with environment and lifestyle, that shape a person’s overall physical features and disease risk. Researchers have identified millions of human genomic variants that differ across the world, including thousands associated with the disease. With results from this new research and advanced computer modeling, Penn and other IGVF consortium investigators aim to identify which variants in the genome are relevant for health and disease, with major implications for physicians and their patients.  

“A fundamental question in biology is to understand how genetic variation affects genome function to influence human health and diseases,” said Hao Wu, Ph.D., an assistant professor of Genetics in the Perelman School of Medicine at the University of Pennsylvania, who will serve as the Penn site’s principal investigator. “With the IGVF award, we can leverage the brainpower of Penn’s experts in human genetics, single-cell sequencing, and functional genomics to decode how genetic variants may contribute to how genes are regulated, how cells function, and ultimately, human diseases. This is a terrific opportunity for collaboration with researchers across departments and institutions.”

Penn Medicine site researchers will focus on the dynamic processes of generating human heart and brain cells, investigating the impact of genomic variants from ethnically diverse populations on the regulatory networks that control gene expression. This effort will generate a detailed functional map of genetic and epigenetic landscapes in early lineages of human heart and brain cells, allowing researchers to dive further into how genes and non-coding regulatory sequences play a role in human congenital heart or brain diseases.

The IGVF consortium plans to develop a catalog of the results and approaches used in their studies and share this information through a web portal to assist the research community with future projects. Since there are thousands of genomic variants associated with disease, and it is not possible to manipulate each variant individually and in each biological setting, consortium researchers will also develop computational modeling approaches to predict the impact of variants on genome function.

The Penn Medicine site will be co-led by Hongjun Song, Ph.D., a professor of Neuroscience at the Perelman School of Medicine at the University of Pennsylvania, and Sreeram Kannan, Ph.D., an assistant professor of Electrical & Computer Engineering at the University of Washington, Seattle. Other researchers from Penn include Sarah Tishkoff, Ph.D., the David and Lyn Silfen University Professor of Genetics and Biology, Guo-li Ming, MD, Ph.D., the Perelman Professor of Neuroscience, Kiran Musunuru, MD, Ph.D., MPH, a professor of Medicine, Junwei Shi, Ph.D., an assistant professor of Cancer Biology, Ziyue Gao, Ph.D., an assistant professor of Genetics, and Wenli Yang, Ph.D., a research assistant professor of Medicine.

Duke, Pitt researchers demo training enormous AI models in health care while protecting data privacy

Tyler O'Neal, Staff Editor HEALTH

The new platform draws data from multiple institutions while protecting the privacy

Researchers at Duke University and the University of Pittsburgh have developed a platform that allows multiple hospitals and research centers to share private patient data securely to better train machine learning models. The technology could provide single institutions access to advanced predictive tools they could never develop on their own to both advance research and improve patient outcomes.

Called “LEARNER,” researchers summarized the platform’s development at the National Science Foundation’s Convergence Accelerator Expo 2021, an event that shares the program’s research portfolio in an exhibition format, like a big science fair. The new LEARNER platform trains AI models with sensitive health care data by only sharing the internal weightings and workings of the algorithm, keeping all of the data inherently secure in the user’s database.

“AI has incredible potential to improve health data analysis and diagnosis, but it requires a vast amount of data to reach a standard that is acceptable to use in real-life decisions,” said Helen Li, the Clare Boothe Luce Professor of Electrical and Computer Engineering at Duke. “And whenever you talk about health care data, there’s always a high level of privacy concerns. LEARNER allows health data from many sources to be used to train an AI model without actually sharing any of the sensitive data.”

When a machine learning algorithm is trained, it compares the decisions it arrives at to the correct answers, attempts to tweak its inner workings to fix the errors, and repeats the process over and over again until it is no longer improving. These tweaks to its inner workings are referred to as weight parameters.

LEARNER is based on a concept called “Federated learning.” In this setup, a single AI model is housed in a central cloud that is provided to users in multiple locations. Each location runs the AI model with its own data and produces a new set of weight parameters, which is in turn sent back to the cloud. The central AI model then uses all of the new weight parameters to develop a single updated algorithm. The process is repeated until the AI model is as good as it can get.

Because only the weight parameters and not the actual data are being shared with the cloud, this technique sidesteps any concerns about data privacy, but the final trained model still represents data from all the entities involved.

“The original information remains hidden on local computers,” explained Li. “For a large model, the process typically requires about 50-100 rounds of training between the local entities and the cloud, which sounds like it might take a long time, but in fact only takes a matter of hours.”

Built in collaboration with Heng Huang, the John A. Jurenko Endowed Professor at the University of Pittsburgh, the LEARNER prototype demonstrated its usefulness in single-cell multi-omic data and electronic health records. In the former, researchers showed that LEARNER could use scRNA sequencing data to predict the protein markers for associating mRNA sequences with protein production. In the latter, they were able to use health record data to predict the probability of heart failure patients being readmitted within 30 days of being released.

But if all goes according to plan, that’s only the beginning. The researchers are developing a user-friendly interface to encourage researchers and doctors to use the platform. Not only would this help LEARNER develop new and better AI health models, but the platform also would eventually provide users with hundreds, if not thousands, of pre-trained AI models that they could use in their own laboratories and hospitals.

“We hope LEARNER will be a platform for health experts who want to take advantage of AI but maybe don’t know a lot about AI themselves,” said Li. “We also hope it will help AI researchers who want to delve into health care and biomedical fields collaborate with one another on large-scale projects.”

Li and her colleagues are in talks with a North Carolina-based AI company to continue to develop and potentially commercialize the LEARNER platform.