2025 · Scientific Reports, 15(1), 23069
Identifying and characterizing suicide decedent subtypes using deep embedded clustering
Belouali, A., Kitchen, C., Zirikly, A., Nestadt, P., Wilcox, H. C., & Kharrazi, H.
Biomedical Informatician · Clinical AI
Anas Belouali.
Clinical AI, machine learning, and real-world health data — building trustworthy systems that hold up in care delivery, not just on a leaderboard.
Adjunct Faculty · Health Informatics & Data Science Program, Georgetown University
Consulting Principal Data Scientist · Hackensack Meridian Health — lung cancer screening & patient decision support
About
Biomedical informatician and health data scientist specializing in clinical AI, machine learning, and real-world health data. PhD from the Johns Hopkins School of Medicine (NLM T15 Fellow), with research focused on using longitudinal health records and interpretable ML to study mental health trajectories and high-risk patient populations. I split my time between research, teaching, and consulting on AI-enabled tools for health systems.
More than a decade of work across academic and health-system settings — electronic health records, predictive modeling, clinical research infrastructure, and responsible AI evaluation. Before Hopkins I led data science at Georgetown's Innovation Center for Biomedical Informatics, building registries for immuno-oncology, NLP pipelines for adverse-event extraction, and a precision-medicine platform integrating multi-omics with clinical records.
Today I split my time between teaching AI for Health Applications at Georgetown and consulting with Hackensack Meridian Health on AI-enabled tools for lung cancer screening and patient decision support. Independent research continues — particularly around suicide risk prediction and longitudinal mental-health analytics from claims data.
A view of the work
Publications, by year and topic
A decade of work — one cell per paper, colored by primary topic.
2026
02
2025
07
2024
02
2023
04
2022
05
2021
03
2020
01
2019
01
2017
01
2016
01
Hover a cell for details · click to open · outlined cells are highlighted work · max 7 papers/year
Selected work
2025 · JAMA Network Open
Identification of temporal condition patterns associated with suicide from claims data using sequence pattern mining
Belouali, A., Kitchen, C., Haroz, E., Lehmann, H., Nestadt, P., Wilcox, H. C., & Kharrazi, H.
2025 · Nature Communications, 16(1), 6274
Towards fair decentralized benchmarking of healthcare AI algorithms with the Federated Tumor Segmentation (FeTS) Challenge
Zenk, M., Baid, U., Pati, S., Linardos, A., Edwards, B., Sheller, M., Foley, P., …, Belouali, A., …, & Yang, H.
2022 · JAMIA Open, 5(2), ooac046
Impact of social determinants of health on improving the LACE index for 30-day unplanned readmission prediction
Belouali, A., Bai, H., Raja, K., Liu, S., Ding, X., & Kharrazi, H.
Now
Current focus
Suicide risk prediction & phenotyping
Using the Maryland Suicide Data Warehouse and large-scale claims data to identify high-risk clinical trajectories, characterize decedent subtypes with deep embedded clustering, and surface temporal condition patterns associated with suicide death.
Digital monitoring & youth mental health
County-level evaluation of digital monitoring tools (e.g., GoGuardian Beacon) used in U.S. K-12 schools to identify students at risk of self-harm, using difference-in-differences and quasi-experimental designs.
Contact
Open to collaborations on mental health informatics, suicide prevention research, real-world evidence, and AI for healthcare.