Research
If you are interested in extensions of AI and/or how it can be used to improve human health, you are in the right place. Selected projects have a ⭐ next to them. A full list is also available on Google Scholar.
Topics
-
⭐ Minimizing and quantifying uncertainty in AI-informed decisions: Applications in medicine
Samuel D. Curtis*, Sambit Panda*, Adam Li*, Haoyin Xu, Yuxin Bai, Itsuki Ogihara, Eliza O’Reilly, Yuxuan Wang, Lisa Dobbyn, Maria Popoli, Janine Ptak, Nadine Nehme, Natalie Silliman, Jeanne Tie, Peter Gibbs, Lan T. Ho-Pham, Bich N. H. Tran, Thach S. Tran, Tuan V. Nguyen, Ehsan Irajizad, Michael Goggins, Christopher L. Wolfgang, Tian-Li Wang, Ie-Ming Shih, Amanda Fader, Anne Marie Lennon, Ralph H. Hruban, Chetan Bettegowda, Lucy Gilbert, Kenneth W. Kinzler, Nickolas Papadopoulos, Bert Vogelstein, Joshua T. Vogelstein, Christopher Douville
PNAS, 2025Introduces MIGHT, which helps quantify the amount of predictive information in very high-dimensional data. This was then used to develop and evaluate a biomedical assay to detect cancer early.
-
📄 Fragmentation signatures in cancer patients resemble those of patients with vascular or autoimmune diseases
Samuel D. Curtis, Tingshan Liu, Yuxin Bai, Yuxuan Wang, Sambit Panda, Adam Li, Haoyin Xu, Eliza O’Reilly, Lisa Dobbyn, Maria Popoli, Janine Ptak, Natalie Silliman, Chris Thoburn, Jeanne Tie, Peter Gibbs, Lan T. Ho-Pham, Bich N. H. Tran, Thach S. Tran, Tuan V. Nguyen, Maximilian F. Konig, Michelle Petri, Antony Rosen, Christopher A. Mecoli, Ami A. Shah, Frits Mulder, Nick van Es, PLATO-VTE Study Group, Chetan Bettegowda, Kenneth W. Kinzler, Nickolas Papadopoulos, Joshua T. Vogelstein, Bert Vogelstein, Christopher Douville
PNAS, 2025Shows that there is a shared inflammatory process between cancer and other diseases and thus uncovers a major reason for false positives in early detection tests for cancer.
-
⭐ Generative AI for Biomedical Decisions
Sambit Panda, Christian Cruz
2025A talk giving a basic overview of how to practically get started with generative AI for a biomedical audience. This workshop was done as part of the NIH AIM-AHEAD DICB program.
-
📄 When no answer is better than a wrong answer: a causal perspective on batch effects
Eric W. Bridgeford, Michael Powell, Gregory Kiar, Stephanie Noble, Jaewon Chung, Sambit Panda, Ross Lawrence, Ting Xu, Michael Milham, Brian Caffo, Joshua T. Vogelstein
Imaging Neuroscience, 2025Models batch effects as causal effects, and introduces approaches that leverage causal machinery to mitigate these effects.
-
⭐ Universally Consistent K-Sample Tests via Dependence Measures
Sambit Panda*, Cencheng Shen*, Ronan Perry, Jelle Zorn, Antoine Lutz, Carey E. Priebe, Joshua T. Vogelstein
Statistics & Probability Letters, 2025Introduces the idea that the k-sample testing problem and independence testing problem are equivalent up to a transformation of the data.
-
🎓 Random Forest for Hypothesis Testing: Development and Application to Cancer Detection
Sambit Panda
Johns Hopkins, 2024 -
📝 hyppo: A Multivariate Hypothesis Testing Python Package
Sambit Panda, Satish Palaniappan, Junhao Xiong, Eric W. Bridgeford, Ronak Mehta, Cencheng Shen, Joshua T. Vogelstein
arXiv, 2024Introduces
hyppo, a package that incorporates conventional and novel multivariate hypothesis tests. -
📝 Accurate and efficient data-driven psychiatric assessment using machine learning
Kseniia Konishcheva, Bennett Leventhal, Maki Koyama, Sambit Panda, Joshua T. Vogelstein, Michael Milham, Ariel Lindner*, Arno Klein*
PsyArXiv, 2024Provides a tool for creating a machine learning based scientific assessment using data from the Healthy Brain Network (HBN).
-
📄 FiPhA: an open-source platform for fiber photometry analysis
Matthew F. Bridge, Leslie R. Wilson, Sambit Panda, Korey D. Stevanovic, Ayland C. Letsinger, Sandra McBride, Jesse D. Cushman
Neurophotonics, 2024Introduces
FiPhA, a R package for performing fiber photometry analysis. -
📄 Partial or Complete Loss of Norepinephrine Differentially Alters Contextual Fear and Catecholamine Release Dynamics in Hippocampal CA1
Leslie R. Wilson*, Nicholas W. Plummer*, Irina Y. Evsyukova, Daniela Patino, Casey L. Stewart, Kathleen G. Smith, Kathryn S. Konrad, Sydney A. Fry, Alex L. Deal, Victor W. Kilonzo, Sambit Panda, Natale R. Sciolino, Jesse D. Cushman, Patricia Jensen
Biological Psychiatry: Global Open Science, 2024Investigates the role of norepinephrine (NE), a neurotransmitter, in fear and NE release changes with genotype, sex, etc.
-
📝 Learning sources of variability from high-dimensional observational studies
Eric W. Bridgeford, Jaewon Chung, Brian Gilbert, Sambit Panda, Adam Li, Cencheng Shen, Alexandra Badea, Brian Caffo, Joshua T. Vogelstein
arXiv, 2023Generalizes causal estimators to arbitrary dimensional space and uses this to develop a new test (Causal CDcorr).
-
📝 Simplest Streaming Trees
Haoyin Xu, Jayanta Dey, Sambit Panda, Joshua T. Vogelstein
arXiv, 2023Developed a streaming algorithm for decision trees based on the simplest possible extension of them.
-
📝 Learning Interpretable Characteristic Kernels via Decision Forests
Sambit Panda*, Cencheng Shen*, Joshua T. Vogelstein
arXiv, 2023Demonstrates the kernel derived from random forest is characteristic and develops a hypothesis test based on that fact (KMERF).
-
🖼️ Elucidating Relationships within Neurological Screening Batteries via Random Forest-Based Hypothesis Testing
Sambit Panda, Leslie R. Wilson, Jariatu Stallone, Dalisa Kendricks, Korey Stevanovic, Jesse D. Cushman
2023Applies a random forest based hypothesis test (specifically KMERF) to evaluate the effectiveness of a neurological screening test for mice.
-
⭐ The Chi-Square Test of Distance Correlation
Cencheng Shen, Sambit Panda, Joshua T. Vogelstein
JCGS, 2022Derives an approximation to the p-value of distance correlation that bypasses the permutation test with no significant loss of power.
-
📝 When are Deep Networks really better than Decision Forests at small sample sizes, and how?
Haoyin Xu, Kaleab A. Kinfu, Will LeVine, Sambit Panda, Jayanta Dey, Michael Ainsworth, Yu-Chung Peng, Madi Kusmanov, Florian Engert, Christopher M. White, Joshua T. Vogelstein, Carey E. Priebe
arXiv, 2021Illustrates that forest based methods excel at tabular data classification at small sample sizes while networks excel at larger sample sizes.
-
🎓 Multivariate Independence and k-sample Testing
Sambit Panda
Johns Hopkins, 2020My master’s thesis, which introduces a Python package and a new framework for k-sample testing.
-
📄 Selective and Mechanically Robust Sensors for Electrochemical Measurements of Real-Time Hydrogen Peroxide Dynamics in Vivo
Leslie R. Wilson, Sambit Panda, Andreas C. Schmidt, Leslie A. Sombers
Analytical Chemistry, 2018Developed a sensor that can be used to monitor real-time dynamics of hydrogen peroxide in the brain; we used it to investigate Parkinson’s disease.