Pain, pleasure and touch: one readout from mice to humans
Can we read pain and pleasure the same way in a mouse and in a person, without touching either? It matters for both: handling stresses animals, while rating scales miss what patients cannot put into words. The aim is one hands-off readout of pain, pleasure and social behaviour that travels across species, from pose, face and movement on camera, with touch delivered by robots and behaviour discovered with as little human bias as possible. Pretrained pose models, behaviour transformers and vision-language models start from open video; the two days end with a prototype and the outline of a translational study. Barely explored so far: how sight and sound reshape touch.
Resources
- Animal recordings
- Sensitive data handling
- Wearables and phone sensing
- Brain signals and imaging
- CalMS21 — mouse social behaviour dataset (Caltech, CC BY 4.0) ↗
- MABe22 — multi-agent behaviour benchmark ↗
- Video / vision models
- Lab equipment
- Speech models
- Language-model endpoints
- Data platform and governance
- DeepLabCut — markerless pose estimation (LGPL-3.0) ↗
- SuperAnimal-TopViewMouse — pretrained mouse pose model (DeepLabCut model zoo) ↗
- SLEAP — multi-animal pose tracking (BSD-3-Clear) ↗
- Keypoint-MoSeq — behaviour motif discovery (non-commercial academic) ↗
- B-SOiD — unsupervised behaviour discovery (GPL-3.0) ↗
- SimBA — supervised behaviour classification (academic use only) ↗
- LISBET — social behaviour transformer, Bellone lab (academic non-commercial) ↗
- QuPath — tissue image analysis (GPL-3.0) ↗
- Cellpose — cell and structure segmentation (BSD-3) ↗
