Research Teams Awarded Funding to Support Postdoctoral Health AI Research (PAIR) Fellows

The Rutgers Center for Biomedical Informatics & Health Artificial Intelligence (BMIHAI) is proud to announce the recipients of funding for multidisciplinary projects to support Postdoctoral Health AI Research (PAIR) Fellows. The four research teams have been selected to receive support to hire postdoctoral fellows who will lead projects and benefit from multi-disciplinary mentorship from experts in health AI-related fields.

“This mechanism aims to support a new generation of postdoctoral fellows who will be trained and mentored by multi-disciplinary teams of experts in cutting-edge, cross-disciplinary biomedical informatics and health AI-related fields,” said Leslie Lenert, the director of BMIHAI and a professor in the Department of Medicine at Robert Wood Johnson Medical School (RWJMS).

The research teams receiving PAIR Fellow funding are:

Project 1 Lead Mentor:
• Ellen Francis, an assistant professor of epidemiology at Rutgers School of Public Health

Co-Mentors:
• Priyadarshini Kachroo, an assistant professor of biomedical and health informatics at the Rutgers School of Health Professions
• Stephanie Shiau, an associate professor of epidemiology at Rutgers School of Public Health

According to the researchers, their project, “Supervised Epigenomic and Metabolomic Integration to Characterize Cardiometabolic Programming in HIV-Exposed Uninfected Infants,” examines how prenatal exposure to HIV, despite the absence of infection in the infant, may influence early cardiometabolic health.

“By integrating epigenomic and metabolomic data collected at birth, the study will identify molecular pathways associated with cardiometabolic health at 12 months of age,” they said. “The findings may provide new insights into how early-life biological changes influence lifelong health and inform future strategies to promote healthy development.”

Project 2 Lead Mentor:
• Anat Kreimer, an associate professor in the department of biochemistry and molecular biology at RWJMS

Co-Mentors:
• James Millonig, an associate professor in the department of neuroscience and cell biology at RWJMS
• Mahmudur Hera, a postdoctoral fellow at the Rutgers Center for Advanced Biotechnology and Medicine
• Arif S. W. Kusuma, a postdoctoral fellow at the Rutgers Center for Advanced Biotechnology and Medicine

“Deciphering how genetic variation alters the function of regulatory elements remains a central challenge in genomics, as gene regulation underlies nearly all biological processes,” the research team said. “The project, “Novel Computational Genomics Methods to Decipher Context-Specific Mechanisms of Variation in Gene Regulatory Networks,” aims to advance understanding of how these genetic variants disrupt context-specific gene regulatory networks (GRNs) which are crucial for understanding disease susceptibility,” they said.

The team proposes to develop innovative methods for the reconstruction, analysis and prediction of GRNs, their architecture and components, and making these methods easily accessible to the research community. “Our framework will provide a generalizable framework for translating large-scale genomic data into mechanistic insights, advancing our understanding of the gene regulatory code, disease etiology, and applications in precision medicine,” the researchers said.

Project 3 Lead Mentor:
• Soha Saleh, an assistant professor in the Department of Rehabilitation and Movement Sciences in the Rutgers School of Health Professions

Co-Mentors:
• Waheed U. Bajwa, a professor in the Department of Electrical and Computer Engineering and the Department of Statistics
• David Zald, the Henry Rutgers Professor of Psychiatry and director of the Rutgers Center of Advanced Human Brain Imaging Research
• Vikram Bhise, an associate professor of pediatrics and neurology at RWJMS
• Sergei Adamovich, a professor in the department of biomedical engineering at New Jersey Institute of Technology
• Fares Yahya Al-Shargie, 2026 PAIR Fellow and Research Associate II at Rutgers

According to the researchers, their study, “Integrating Brain Connectivity and Machine Learning to Predict Movement Recovery After Stroke,” leverages machine learning and computational biomarker discovery approaches to identify EEG-based brain network biomarkers predictive of post-stroke functional recovery.

“By integrating longitudinal brain connectivity measures with clinical assessments collected within one month and at 4 and 6 months post-stroke, the project will develop and validate predictive models that enable individualized prognosis, stratify patients by recovery potential, and support precision rehabilitation strategies,” they said.

Project 4 Lead Mentor:
• Linden Parkes, an assistant professor of psychiatry at Rutgers Center of Advanced Human Brain Imaging Research

Co-Mentors:
• Ahmad Beyh, 2026 PAIR Fellow and Postdoctoral Associate at Rutgers
• Waheed U. Bajwa, a professor in the Department of Electrical and Computer Engineering and the Department of Statistics
• Stephen Hanson, a professor of psychology at Rutgers-Newark

According to the research team, schizophrenia and related psychotic disorders emerge gradually as the brain develops, and they involve subtle problems in how brain regions are wired together and how they signal to one another. A likely contributor lies in the brain’s microstructure: the fine-scale organization of brain tissue that shapes how efficiently neural signals travel between regions.

This project, “Using biophysical recurrent neural networks to characterize and chart the developmental emergence of abnormal brain dynamics in psychosis,” builds AI-based models of the brain, trained on each person’s own brain-imaging maps of this microstructure, that can simulate how brain wiring and activity arise from underlying biology.

The researchers said, “The goal is to pinpoint which disruptions actually cause the problems seen in patients and to forecast how a person’s symptoms are likely to unfold over time, ultimately supporting earlier, more biologically grounded identification and treatment of these disorders.”

BMIHAI, based within IFH, serves as a catalyst for transformative research by harnessing the power of AI to transform the way research is conducted and by uniting health-related educational, training and research efforts involving data science under one umbrella.