Gait dataset labeled by encounter, spanning pre-op to 3 months post-TKA
zawthet · x · 2026-11-12
Data-side notes from the AAHKS work: every knee was measured in a real visit, ordered by the surgeon with the clinical reason attached, pre-op and at 3 months—video-derived pose, joint angles, continuous signals, labeled by the encounter rather than by an annotator. The conclusion holds that dynamic coronal alignment is biomechanically distinct and may aid pre-op TKA planning, with patient-reported outcomes layered on next.
More from Research
- Embodied First Exam benchmarks coding agents on 47 robots across 1,156 tasks; best agent hits just 34% — MLL-Lab · 2026-11-12
- Frequency-aware remasking boosts diffusion LLMs by up to 7.33 points on MATH-500 — ucriverside · 2026-11-12
- Remote Labor Index paper on AI automating remote work: 'computer workers should buckle up' — FlorianGallwitz · 2026-11-12
- LMSYS megakernel pushes decoding to 1,109 tok/s on 8× GB300, launches cut 1450→97 — ying11231 · 2026-11-12
- MPEG Diffusion renders subjects hi-res and backgrounds low-res for up to 4.8× faster video generation — xxunhuang · 2026-11-12
- UMAP author recommends PyNNDescent for much faster kNN-graph construction — leland_mcinnes · 2026-11-12