Nassim Oufattole
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Nassim Oufattole has a total of 38 co-authors. Here are the top co-authors by number of shared publications:
- Matthew B. A. McDermott
Department of Biomedical Informatics, Columbia University, New York, NY, USA
3 shared publications - Arkadiusz Sitek
Radiology Department, Massachusetts General Hospital, Boston, MA, USA; Harvard Medical School, Boston, MA, USA
2 shared publications - Ethan Steinberg
Center for Biomedical Informatics Research, Stanford University, Stanford, CA, USA
2 shared publications - Hyewon Jeong
Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, USA
2 shared publications - Justin Xu
Big Data Institute, University of Oxford, Oxford, UK
2 shared publications - Kamilė Stankevičiūtė
Department of Computer Science and Technology, University of Cambridge, Cambridge, UK
2 shared publications
Latest Publications
- 28 May 2026
- Published by Massachusetts Medical Society in NEJM AI
- Vol. 3 (6)
MEDS — An Emerging Data Standard and Ecosystem for Health AI Research
- 28 May 2026
- Published by Massachusetts Medical Society in NEJM AI
- Vol. 3 (6)
MEDS — An Emerging Data Standard and Ecosystem for Health AI Research
- Open Access
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 6243-6244
MEDS: Building Models and Tools in a Reproducible Health AI Ecosystem
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 6243-6244
Open AccessMEDS: Building Models and Tools in a Reproducible Health AI Ecosystem
- Open Access
- 6 January 2025
- Published by Oxford University Press (OUP) in GigaScience
- Vol. 14
Foundation model of electronic medical records for adaptive risk estimation
Background: Hospitals struggle to predict critical outcomes. Traditional early warning systems, like NEWS and MEWS, rely on static variables and fixed thresholds, limiting their adaptability, accuracy, and personalization. Methods: We previously developed the Enhanced Transformer for Health Outcome Simulation (ETHOS), an artificial intelligence (AI) model that tokenizes patient health timelines (PHTs) from electronic health records and uses transformer-based architectures to predict future PHTs. ETHOS is a versatile framework for developing a wide range of applications. In this work, we develop the Adaptive Risk Estimation System (ARES) that leverages ETHOS to compute dynamic, personalized risk probabilities for clinician-defined critical events. ARES also features a personalized explainability module that highlights key clinical factors influencing risk estimates. We evaluated ARES using the MIMIC-IV v2.2 dataset, together with its emergency department extension, and benchmarked performance against both classical early warning systems and contemporary machine learning models. Results: The entire dataset was tokenized, resulting in 285,622 PHTs (63% with at least 1 hospital admission), comprising over 357 million tokens. ETHOS outperformed benchmark models in predicting hospital admissions, intensive care unit admissions, and prolonged stays, achieving superior area under the curve scores. Its risk estimates were robust across demographic subgroups, with calibration curves confirming model reliability. The explainability module provided valuable insights into patient-specific risk factors Conclusions: ARES, powered by ETHOS, advances predictive health care AI by delivering dynamic, real-time, personalized risk estimation with patient-specific explainability. Although our results are promising, the clinical impact remains uncertain. Demonstrating ARES’s true utility in real-world settings will be the focus of our future work.
- 6 January 2025
- Published by Oxford University Press (OUP) in GigaScience
- Vol. 14
Open AccessFoundation model of electronic medical records for adaptive risk estimation
Background: Hospitals struggle to predict critical outcomes. Traditional early warning systems, like NEWS and MEWS, rely on static variables and fixed thresholds, limiting their adaptability, accuracy, and personalization. Methods: We previously developed the Enhanced Transformer for Health Outcome Simulation (ETHOS), an artificial intelligence (AI) model that tokenizes patient health timelines (PHTs) from electronic health records and uses transformer-based architectures to predict future PHTs. ETHOS is a versatile framework for developing a wide range of applications. In this work, we develop the Adaptive Risk Estimation System (ARES) that leverages ETHOS to compute dynamic, personalized risk probabilities for clinician-defined critical events. ARES also features a personalized explainability module that highlights key clinical factors influencing risk estimates. We evaluated ARES using the MIMIC-IV v2.2 dataset, together with its emergency department extension, and benchmarked performance against both classical early warning systems and contemporary machine learning models. Results: The entire dataset was tokenized, resulting in 285,622 PHTs (63% with at least 1 hospital admission), comprising over 357 million tokens. ETHOS outperformed benchmark models in predicting hospital admissions, intensive care unit admissions, and prolonged stays, achieving superior area under the curve scores. Its risk estimates were robust across demographic subgroups, with calibration curves confirming model reliability. The explainability module provided valuable insights into patient-specific risk factors Conclusions: ARES, powered by ETHOS, advances predictive health care AI by delivering dynamic, real-time, personalized risk estimation with patient-specific explainability. Although our results are promising, the clinical impact remains uncertain. Demonstrating ARES’s true utility in real-world settings will be the focus of our future work.