Reading patterns in glucose over time. ↗
GluFormer researchers trained a model on continuous glucose readings to study metabolic patterns and prediction.
Experimental · Weizmann Institute / Pheno.AI / NVIDIAHealth & biology
Medicine, molecules, human ability. A place to discover developments quickly, then stay with the questions that matter to you.
See the work.
Newest source publications first. These are source dates, not live activity.
An MRI becomes a spatial view. ↗
SimBioSys TumorSight Viz turns breast MRI images into three-dimensional views for surgical planning, according to NVIDIA’s report.
Emerging · SimBioSys / NVIDIAMore room to read cancer genomes. ↗
Wellcome Sanger Institute explored accelerated genome analysis using NVIDIA Parabricks and GPU systems.
Emerging · Wellcome Sanger Institute / NVIDIAA way back to conversation. ↗
A man with ALS used an experimental brain–computer interface to turn attempted speech into conversation.
Emerging · UC DavisSeeing molecules together. ↗
AlphaFold 3 expanded structural prediction to complexes of proteins and other biomolecules.
Demonstrated · Google DeepMind & Isomorphic LabsA surgical ecosystem tests an edge. ↗
J&J MedTech and NVIDIA announced an MOU to test AI capabilities for a connected digital surgery ecosystem.
Emerging · Johnson & Johnson MedTech / NVIDIALooking for new antibiotic families. ↗
Researchers used explainable deep learning to identify a structural class of antibiotic candidates.
Experimental · MIT & collaboratorsDesigning proteins with purpose. ↗
RFdiffusion enabled researchers to generate proteins with specified structural and functional goals.
Demonstrated · University of Washington & collaboratorsA new lead against resistant bacteria. ↗
Scientists identified abaucin, a compound active against Acinetobacter baumannii, with help from deep learning.
Experimental · McMaster University & MITReconnecting intention and movement. ↗
A brain–spine interface enabled a participant with spinal cord injury to control standing and walking.
Emerging · EPFL, CHUV & collaboratorsA broader map of human proteins. ↗
Researchers applied AlphaFold to a broad collection of human proteins.
Demonstrated · DeepMind & EMBL-EBISeeing the shape of life. ↗
Researchers used AlphaFold to predict protein structures with accuracy competitive with experimental structures on many CASP14 targets. Bright's interactive p53 view uses a separate, later AlphaFold DB prediction: human p53 AF-P04637-F1 model v6, created on 2025-08-01.
Demonstrated · DeepMind