Low-quality slit-lamp images can undermine automated keratitis diagnosis. Jiang, Song, Yao et al. report a two-stage AI framework designed to recover image quality and preserve diagnostic performance.

A large multi-omics analysis identifies SCG2 in tumor-associated macrophages as a key mediator of 5-FU resistance in colorectal cancer.

Transcript INTRO Welcome to Medical AI Digest Podcast. Today, a paper from NPJ digital medicine. MAIN The paper is titled “Physicians and artificial intelligence diverge in evaluating large…

Wearable inertial sensors and Topological Data Analysis identified interpretable gait phenotypes across neuromotor conditions. In stroke, the resulting subgroups tracked with clinically meaningful differences in mobility and gait…

A refined AI-assisted ASPECTS approach improved acute ischemic stroke imaging assessment and prognostic prediction across DWI, CT, and prospective real-world cohorts.

Transcript INTRO Welcome to Medical AI Digest Podcast. Today, a paper from NPJ digital medicine. MAIN Today we’re looking at “Whole body CT attenuation and volume charts from…

BCGNet shows that physiological knowledge learned from PSG can strengthen contactless sleep monitoring with an under-pillow ballistocardiography device.

COMPASS is a generalizable AI model that predicts immune checkpoint inhibitor response from bulk tumor transcriptomes across multiple cancers and therapies.

KGRD is a knowledge-graph-augmented AI framework designed to support diagnosis and counselling in paediatric rare genetic disorders. In benchmark testing, it outperformed baseline LLM approaches, especially when only…

Transcript INTRO Welcome to Medical AI Digest Podcast. Today, a paper from Annals of surgery. MAIN The paper is titled “The Ambulatory Surgery Center Paradox: Why 60% of…