UBC's Kev-Bio VLM guides ultrasound landmark detection via clinical questions
UBC-V · hf · 2026-11-04
Researchers at the University of British Columbia released Kev-Bio, a vision-language model for medical ultrasound biometry. It guides ultrasound landmark detection through clinical questions and generates spatial probability maps to derive measurements, aiming to reduce operator variability in applications like prenatal and cardiac exams.
More from Research
- Hobbyist rewrites Qwen 0.6B weights with 39.8M tokens, no gradients, in 3h12m — Altruistic_Heat_9531 · 2026-11-04
- SpecPilot agent boosts speculative decoding from 4.46× to 10.87×, no retraining — pmttyji · 2026-11-04
- Autonomous agent designs target-specific MSAs to improve AlphaFold 3 complex prediction — SJTU · 2026-11-04
- Tencent's LoCo teaches reasoning models when to stop, cutting output length without hurting accuracy — tencent · 2026-11-04
- Muon optimizer makes the first MoE layer harder to load-balance, study finds — Marcos Hernanz · 2026-11-04
- GigaSpeech 3: 39,000 hours of weakly labeled dialectal Arabic speech released for ASR — X-LANCE · 2026-11-04