NEJM AI is a new journal on medical artificial intelligence and machine learning from NEJM Group, the publisher of @NEJM.
- Tweets1.7K
- Following330
- Followers13.7K
- Likes1.1K
ALT Promotional image for episode 47 of the NEJM AI Grand Rounds podcast featuring Dr. Michal Nedoszytko. The left side includes the title: "Abridge’s Dr. Michal Nedoszytko on Physicians Who Build Software." A photo of Dr. Nedoszytko is on the right.
ALT A quote from a Perspective published in NEJM AI reads as follows: "The barrier to building has collapsed, but the barrier to building safely has not. For the first time, the people who understand patient care best may have the means to shape technology that helps patients directly.” The Perspective is titled “Vibe Coding versus Clinical Reality” and the author is Michal Nedoszytko, M.D., Ph.D. The background color is a blue gradient and includes the NEJM AI logo.
ALT Four panels show brain MRI of patient with SHH medulloblastoma before and 3 months after Pretreatment T1-weighted postcontrast (Panel A) and FLAIR postcontrast (Panel B) images show a heterogeneously enhancing cerebellar mass. Follow-up T1-weighted postcontrast (Panel C) and FLAIR postcontrast (Panel D) images demonstrate substantial interval decrease in lesion size with minimal residual enhancement and persistent FLAIR hyperintensity.
ALT The image is a diagram of the canonical and noncanonical SHH pathway signaling related to SMO inhibitor noncandidacy.
ALT The image is page 1 of the Perspective "Clinical AI in 2025–2026 — Six Themes from the SAIL Year in Review" published in NEJM AI. An orange banner at the bottom encourages readers to access the full Perspective at ai.nejm.org.
ALT Four panels show brain MRI of patient with SHH medulloblastoma before and 3 months after Pretreatment T1-weighted postcontrast (Panel A) and FLAIR postcontrast (Panel B) images show a heterogeneously enhancing cerebellar mass. Follow-up T1-weighted postcontrast (Panel C) and FLAIR postcontrast (Panel D) images demonstrate substantial interval decrease in lesion size with minimal residual enhancement and persistent FLAIR hyperintensity.
ALT A table showing themes in clinical AI for 2025–2026 from SAIL year in review. The three columns are "Theme," "Examples," and "Looking Ahead." Themes include "Evaluation: moving up the hierarchy of evidence," "Clinical workflow integration," and "Rise of direct-to-consumer AI."
ALT The image is a flowchart titled "Figure 1. Study Pipeline." The chart shows how a training dataset (n=2359) was collected from a multicenter registry in Japan and evaluated using an independent dataset (n=964) derived from a separate registry conducted at six other hospitals in Japan.
ALT Chart comparing AI auscultation models. The left graph is "Full-site model" with positions 2RSB, 2LSB, 4LSB, and 5LMCL. The right graph is "Two-positions ensemble model" with positions 2LSB and 4LSB. Both graphs display AUROC metrics.
ALT A quote from an Original Article published in NEJM AI reads as follows: “AI-generated layperson summaries provided before outpatient neurosurgical consultations were associated with better patient-reported interaction quality and autonomy support and with shorter consultations.” The article is titled “Effect of Previsit AI-Generated Document Summaries on Consultation Time and Patient–Physician Communication — The AI-INFOCARE Randomized Controlled Trial” and the authors are E. Benker-Hasani et al. The background color is a blue gradient and includes the NEJM AI logo
ALT A bar chart showing the total cost of recommended care and the primary outcome, relative to the treating physician (ratio of means), for six representative current-generation systems under the standard prompt (red bars), the cost-aware prompt (blue), and the parsimonious-but-safe prompt (orange). The systems are GPT-5.5, Claude Opus 4.8, Gemini 3.1 Pro, Grok 4.3, Qwen 3.7, and DeepSeek R1. A dashed line marks the treating physician’s cost.
ALT Image from NEJM Catalyst featuring a health care professional with a patient. A holographic display shows a dialogue about symptoms and questions like "What are your symptoms?" and "When did you first notice it?" The title reads "Ambient AI as a Tool for Returning Time to Care."
ALT A quote from a Perspective published in NEJM AI reads as follows: "The danger is not that trainees will use AI. The danger is that they may never learn what it feels like to reason before AI intervenes.” The Perspective is titled “No Mastery without the Machine — Medical Education in the Age of Clinical AI” and the authors are P. Jarrett et al. The background color is an orange-yellow gradient and includes the NEJM AI logo.
ALT The image is a flowchart titled "Figure 1. Study Pipeline." The chart shows how a training dataset (n=2359) was collected from a multicenter registry in Japan and evaluated using an independent dataset (n=964) derived from a separate registry conducted at six other hospitals in Japan.
ALT Illustration of an AI-generated, layperson-friendly summary of medical findings provided before outpatient consultations. On the left, a physician in a white coat and stethoscope speaks and gestures, set against a faded table of clinical trial statistics and medical terms. On the right, a patient in a hospital gown smiles and gestures back. A blue line connects a glowing pen icon with sparkles to a document icon labeled “Discharge Report Summary."