Vet ai has a truth gap nobody markets
vet ai has a truth gap nobody markets: boehringer’s caninebeat algorithm, trained on 4,000+ annotated dog heart sounds, catches murmurs with >95% sensitivity, yet an nc state trial found experienced clinicians still outperformed an ai stethoscope on the same job.
the frontier of animal diagnostics isn’t model accuracy, it’s knowing when the human ear wins.
Context
Boehringer Ingelheim's CANINEBEAT AI page says the algorithm has over 95 percent sensitivity and specificity for heart murmurs related to structural heart disease. NC State News (Aug 12, 2026) reports that fourth-year veterinary students matched, and experienced clinicians outperformed, an AI-enabled digital stethoscope for murmurs and arrhythmias in cats and dogs.
The study's abstract says it compared the stethoscope with students and experienced clinicians in dogs and cats presenting to a university hospital.
The over 95 percent figure matches the vendor's page, so it is a manufacturer claim. The NC State study matches the clinician comparison; its title says the stethoscope showed moderate murmur detection, and it covered dogs and cats and murmurs plus arrhythmias. Whether the tested stethoscope runs CANINEBEAT, and the 4,000-plus annotated heart sounds, were not seen in the sources read, so unsupported here, not refuted. 'Knowing when the human ear wins' is the author's.
Related work
Watch next
- Check which algorithm the NC State stethoscope used and the 4,000 heart sound training figure.
Sources
- Boehringer Ingelheim: CANINEBEATanimalhealth.boehringer-ingelheim.com
- CANINEBEAT AI overviewcaninebeat.com
- EurekAlert: AI-enabled stethoscope can miss the beat in petseurekalert.org
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 1 October 2026 at 22:47 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 →