NEJM AI
LightNews LightNews
banner
ai.nejm.org
NEJM AI
@ai.nejm.org
1.3K followers 91 following 450 posts
NEJM AI, a new monthly journal from the publisher of @nejm.org, explores the cutting-edge applications of artificial intelligence and machine learning in clinical medicine. Online at ai.nejm.org.
Posts Replies Media Videos
Pinned
ai.nejm.org
NEJM AI @ai.nejm.org · 20d
In the latest episode of the AI Grand Rounds podcast, Dr. Marinka Zitnik of Harvard Medical School describes how AI-driven drug repurposing offers a way to work within existing constraints while still opening new therapeutic possibilities. Full episode: nejm.ai/ep38

#MedSky
AI Grand Rounds Episode 38 Bridging AI and Biology to Tackle Medicine’s Hardest Problems with Dr. Marinka Zitnik Photo of Dr. Zitnik
ai.nejm.org
A new study demonstrates the successful integration of a real-time delayed cerebral ischemia prediction algorithm into a live intensive care unit environment using a silent validation methodology. Learn more: nejm.ai/4r8KL7R

#AI #MedSky #MLSky
Figure 2. Overview of the ICU Cockpit Visual Interface.
February 10, 2026 at 6:15 PM
ai.nejm.org
Case Study by G. Kuling et al.: Assessment of Short-Answer Questions by ChatGPT in a Medical School Course nejm.ai/4jJFYY9
February 10, 2026 at 2:30 PM
ai.nejm.org
An editorial by Dr. Rebecca Sternschein explores the utility of multiple-choice questions and short-answer questions for assessment in medical education and whether a large language model can adequately grade short-answer questions. Read the editorial: nejm.ai/49GROh1

#AI #MedSky #MLSky
Editorial Kuling et al. take an important step in the exploration of how LLMs may enable wider use of [short answer questions], allowing a powerful formative assessment tool for clinically relevant reasoning skills to be a key aspect of the future of medical education. “Grading LLMs on the Ability to Grade” by Rebecca Sternschein, M.D., M.H.P.E.
February 10, 2026 at 2:30 PM
ai.nejm.org
Models advanced. Biology lagged. On NEJM AI Grand Rounds, Dr. Marinka Zitnik explains why collaboration matters as much as computation. Listen to the full episode: nejm.ai/ep38

#AI #MedSky #MLSky
February 9, 2026 at 1:31 PM
ai.nejm.org
Original Article by G. Starke et al.: Machine Learning–Based Patient Preference Prediction: A Proof of Concept nejm.ai/3IAIYrD
February 6, 2026 at 2:26 PM
ai.nejm.org
Response by G. Starke et al.: Machine Learning Should Enrich, Not Replace, Human Deliberation in End-of-Life Decisions nejm.ai/45rBeQW

Letter by Ari Nahum, MD: Model Performance Convergence Highlights Data Limitations in a Patient Preference Predictor nejm.ai/4qNgrzV
February 6, 2026 at 2:26 PM
ai.nejm.org
In a letter, Teva D. Brender, MD, and Alexander K. Smith, MD, comment on the Original Article “Machine Learning–Based Patient Preference Prediction: A Proof of Concept.” nejm.ai/3NC4IFN

@alexsmithmd.bsky.social #AI #MedSky #MLSky
Page 1 of the letter "Machine Learning Can Assist Surrogate Decision-Makers." Read the full letter at ai.nejm.org.
February 6, 2026 at 2:26 PM
ai.nejm.org
Parsa et al. propose the use of health AI target product profiles, which specify the performance thresholds an AI tool must meet to produce benefit within a specific care setting, accounting for workflow, capacity, and utility trade-offs. Learn more: nejm.ai/4qBoHTn

#AI #MedSky #MLSky
“Meaningful evaluation of AI in health care must move beyond only diagnostic accuracy performance metrics and must adopt a proactive approach to identify the true needs in systems where AI can advance care.” Perspective “Designing Clinically Useful AI: A Blueprint for Impact” by S. Parsa et al.
February 5, 2026 at 2:30 PM
ai.nejm.org
Editorial by Rebecca Sternschein, MD, MHPE: Grading LLMs on the Ability to Grade nejm.ai/49GROh1
February 4, 2026 at 5:31 PM
ai.nejm.org
A new study found that while grading short-answer responses from first-year medical students, GPT-4o achieved moderate agreement with human graders, suggesting that AI-assisted grading could help assess open-ended questions in medical education. Learn more: nejm.ai/4jJFYY9

#AI #MedSky #MLSky
Figure 1. Study Flow Diagram for Human and GPT-4o Grading of Short-Answer Responses.
February 4, 2026 at 5:31 PM
ai.nejm.org
Rare diseases are small in numbers — but huge in impact. On NEJM AI Grand Rounds, Dr. Marinka Zitnik explains why #AI trained across thousands of diseases could help change what’s possible. Hear more from Dr. Zitnik: nejm.ai/ep38

#MedSky #MLSky
February 4, 2026 at 2:16 PM
ai.nejm.org
A new Perspective by Venkatesh L. Murthy, MD, PhD, and Chirag J. Patel, PhD, explores what constitutes a high-value application in medical artificial intelligence. Read the full Perspective: nejm.ai/4jIBLUD

@venkmurthy.bsky.social #AI #MedSky #MLSky
“Regardless of one’s feelings about the potential of medical AI, a shift from intellectually interesting problems with relatively niche clinical applications, which dominated the first phase of this field, to solving high-value medical problems ... is overdue.” Perspective “The Paradoxical Challenge of High-Value Medical Artificial Intelligence” by Venkatesh L. Murthy, M.D., Ph.D., and Chirag J. Patel, Ph.D.
February 3, 2026 at 6:30 PM
ai.nejm.org
Editorial by David Blumenthal, MD, MPP: Evolving Regulatory Science nejm.ai/4biPa3o
February 3, 2026 at 2:04 PM
ai.nejm.org
Policy Corner by R. Scheufen Tieghi et al.: The Promise of Animal Testing Alternatives at the U.S. Food and Drug Administration and National Institutes of Health nejm.ai/4svkOB6
February 3, 2026 at 2:04 PM
ai.nejm.org
Editorial by Chirag J. Patel, PhD: Planning for an AlphaFold for Drug Safety Assessment: Benchmarking and Data Infrastructure as Prerequisites nejm.ai/44SwDHo

#AI #MedSky #MLSky
Page 1 of the editorial "Planning for an AlphaFold for Drug Safety Assessment: Benchmarking and Data Infrastructure as Prerequisites" Bottom: Read the full editorial at ai.nejm.org.
February 3, 2026 at 2:04 PM
ai.nejm.org
A new study introduces TARGET-AI, a multimodal framework that integrates longitudinal EHR embeddings with contrastively pretrained ECG image representations to estimate pretest disease probability and guide targeted screening for structural heart disease. nejm.ai/4pU5qeI
Futuristic digital illustration of a cube with a heart icon at its center, surrounded by floating medical records and EKG graphs. Original Article TARGET-AI: A Foundational Approach for the Targeted Deployment of Artificial Intelligence Electrocardiography in the Electronic Health Record E.K. Oikonomou and Others
February 2, 2026 at 6:15 PM
ai.nejm.org
Datasets, Benchmarks, and Protocols by P. Chung et al.: Verifying Facts in Patient Care Documents Generated by Large Language Models Using Electronic Health Records nejm.ai/4sm0Sk1

#AI #MedSky #MLSky
Figure 1. VeriFact Evaluation of a Large Language Model–Generated Clinical Narrative. Figure 2. VeriFact System Architecture. Figure 3. VeriFact Performance for Large Language Model–Generated Brief Hospital Course Narratives Evaluated Using Atomic Claim Propositions.
February 2, 2026 at 1:30 PM
ai.nejm.org
Editorial by David Blumenthal, MD, MPP: Evolving Regulatory Science nejm.ai/4biPa3o

Editorial by Chirag J. Patel, PhD: Planning for an AlphaFold for Drug Safety Assessment: Benchmarking and Data Infrastructure as Prerequisites nejm.ai/44SwDHo
January 30, 2026 at 2:26 PM
ai.nejm.org
Policy Corner by R. Scheufen Tieghi et al.: The Promise of Animal Testing Alternatives at the U.S. Food and Drug Administration and National Institutes of Health nejm.ai/4svkOB6

#AI #MedSky #MLSky
Table 1. Example Validation Criteria for New Approach Methodologies.
January 30, 2026 at 2:26 PM
ai.nejm.org
Case Study by D. van de Sande et al.: Operationalization of Artificial Intelligence to Assist in Surgical Discharge: A Feasibility Case Study nejm.ai/3YJ1MJT

#AI #MedSky #MLSky
Figure 1. Distribution of Surgery Duration. Figure 2. Confusion Matrix for Postoperative Day Two Discharge Prediction. Figure 3. Model Performance on the 114 Patients.
January 29, 2026 at 2:15 PM
ai.nejm.org
Editorial by Chirag J. Patel, PhD: Planning for an AlphaFold for Drug Safety Assessment: Benchmarking and Data Infrastructure as Prerequisites nejm.ai/44SwDHo
January 28, 2026 at 2:17 PM
ai.nejm.org
Policy Corner by R. Scheufen Tieghi et al.: The Promise of Animal Testing Alternatives at the U.S. Food and Drug Administration and National Institutes of Health nejm.ai/4svkOB6
January 28, 2026 at 2:17 PM
ai.nejm.org
Editorial by David Blumenthal, MD, MPP: Evolving Regulatory Science nejm.ai/4biPa3o

@davidblumenthal.bsky.social #AI #MedSky #MLSky
Page 1 of the editorial "Evolving Regulatory Science" Bottom: Read the full editorial at ai.nejm.org.
January 28, 2026 at 2:17 PM
ai.nejm.org
Twelve to 13 years. Several billion dollars. On NEJM AI Grand Rounds, Dr. Marinka Zitnik breaks down the real cost of traditional drug development — and why alternatives matter. Listen to the full episode: nejm.ai/ep38

#AI #MedSky #MLSky
January 27, 2026 at 5:31 PM
ai.nejm.org
Assessing Generative AI Chatbots for Alcohol Misuse Support: A Longitudinal Simulation Study nejm.ai/4bi0wF2
January 27, 2026 at 1:30 PM