Sequential diagnosis with language models.
AI systems analyzing diagnostic information step-by-step achieved 83% accuracy on complex cases, far exceeding typical physician performance and suggesting potential to augment clinical reasoning.
- Lab
- Preclinical
- Early trials
- Late trials
- In practice
Ravulizumab in IgA Nephropathy - Interim Analysis of a Phase 3 Trial.
A complement-blocking infusion reduced kidney protein loss by 43% more than placebo in IgA nephropathy patients, with fewer serious side effects, offering a new treatment avenue.
Promising but preliminaryCD8 T-cell exhaustion linked with CAR T-cell failure can be mitigated…
Reversing exhaustion in failed CAR T cells through IL2 signaling restoration enabled complete tumor eradication in patient models, pointing toward rescuing stalled immunotherapies.
Promising but preliminaryA novel CD123xVdelta2 bispecific T cell engager induces potent and…
A new bispecific T-cell therapy selectively destroyed blood cancer cells in primates with milder inflammatory reactions than existing approaches, advancing toward human trials.
Promising but preliminaryExamining reliance patterns on AI advice in medical imaging
Showing doctors which specific areas of AI predictions matter most boosted their diagnostic accuracy on brain imaging by 84% compared to AI advice alone.