Alibaba damo published radar in science on september 18, an expert level medical imaging model that…
alibaba damo published radar in science on september 18, an expert level medical imaging model that reads 146 diseases across 18 abdominal structures and beat many radiologists on accuracy, with weights, code, and the training framework all open. the strongest general radiologist level ai is free.
the bottleneck shifts from building the model to proving it in a clinic.
Context
A EurekAlert and Science press release of 17 September 2026 describes RADAR, Rapid Abdominal Diagnosis with AI and Radiology, a vision-language model for contrast-enhanced abdominal CT trained on 424,911 examinations, which it says substantially outperformed existing AI systems across a wide range of diseases and clinical settings. The Alibaba DAMO repository says it was trained on over 400,000 exams with 15 million anatomy-aware image-text pairs, and includes inference, evaluation and preprocessing code. The South China Morning Post of 18 September 2026 reports an average AUC of 0.913 across 146 clinical findings and 18 abdominal organs, and that it outperformed most radiologists according to the Science study.
The publication and press release are dated 17 September, and 18 September is the SCMP report date. The Science paper was not read, so the reader study design behind the comparison with radiologists is unverified here. SCMP says 146 clinical findings, which are findings, not diseases. The repository lists inference and preprocessing code, its license is listed as Other and was not read, and training framework openness is not established. World's first expert-level generalist is the team's own claim, and the strongest free line is unsupported.
Watch next
- The Science paper's methods and the license text.
Sources
- RADAR press release (EurekAlert, 17 Sep 2026)eurekalert.org
- damo-radar (GitHub)github.com
- Alibaba open-sources medical AI model (SCMP, 18 Sep 2026)scmp.com
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 19 September 2026 at 01:35 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →