FreedomIntelligence/BodyVista
Learning ResourcesA browser-based 3D anatomy and physiology explorer for students, teachers, and self-directed learners.
Project Overview
BodyVista is a 3D interactive platform for learning human anatomy and physiology. It places 2,900 anatomical structures, 12 body systems, 140 dynamic demonstrations, and six microscopic topics in the browser so learners can inspect spatial relationships, peel back anatomical layers, and connect organ structure with function. The interface is available in Chinese, English, and Japanese, with layouts for desktop and mobile.
The project suits students, teachers, science communicators, and independent learners who want an explorable reference rather than static diagrams. Core interactions run locally in the browser, and the site does not require a self-hosted backend or database.
Learning Modules
- Human Anatomy lets you explore structures by system and by seven body regions, with selection, progressive hiding, multi-select, transparency, and structure focusing.
- Life in Motion provides 133 system demonstrations, five whole-body demonstrations, and two oral and dental demonstrations, with playback, progress, and speed controls.
- Microscopic Exploration covers six topics: alveolar gas exchange, the nephron, sarcomere contraction, neural synapses, intestinal villi, and the vessel wall.
- Anatomy Challenge presents five-question rounds on structure identification, finding structures in 3D, and microscopic knowledge, with answer feedback and results.
The 3D viewer also supports rotation, panning, zooming, perspective and orthographic projection, and transverse, sagittal, and coronal sections. Drawing, erasing, labelling, and measurement tools are included, as are structure search, explanations, related structures, references, and FDI tooth numbers for dental content. An AI-assisted learning option opens ChatGPT or DeepSeek with a prompt built from the current structure and selected language; this is an external link and does not require an API key inside the project.
Quick Start
You can try the hosted live demo or the Hugging Face Space without installing anything. To run the project locally, install Node.js 22.12 or later and npm, then clone the repository and start Vite:
git clone https://github.com/FreedomIntelligence/BodyVista.git
cd BodyVista
npm ci
npm run dev
Open the local URL printed in the terminal. The source entry point must be served through Vite; opening index.html directly from the filesystem will not work. To check tests and produce a production build:
npm test
npm run build
npm run preview
npm run build checks translation and oral-module content coverage before generating the site in dist/. Models and thumbnails are included in the repository, so cloning and initial loading may take some time; keep assets/, thumbs/, and dental/models/ in place. For deployment, the repository publishes through GitHub Actions, with the Pages source set to GitHub Actions; pushes to main install dependencies with Node.js 22, run tests and content checks, build the site and model assets, and publish dist/ (deployment instructions).
Constraints and Licensing
The project requires Node.js 22.12 or later for local development and uses Vite for serving and building. Models and thumbnails are committed to the repository, so a full clone is relatively large and initial loading can take longer than a lightweight documentation project.
The content is intended for teaching and learning. Animations, microscopic scenes, and some oral structures include schematic representations and are not intended for clinical diagnosis or treatment decisions. Third-party models and educational content carry their own licences, including CC BY-SA 2.1 Japan and CC BY-SA 4.0 for some dental resources. Per-structure provenance is recorded in catalog.json, and some records are marked NC exceptions; research only; their provenance and permitted public distribution scope remain to be verified. The MIT licence covers only code and documentation that the project has the right to release under it, not third-party models or separately licensed content (licensing details).
HotRepo's take
The repository was recorded in HotRepo's new-project list for the JavaScript language on 2026-11-03, with 22 stars gained over a growth window from 2026-11-02 18:45 to 2026-11-03 05:45 UTC, reaching 122 stars and 26 forks. No all-language daily rank was recorded. That early snapshot supports a statement about attention during the launch window, not a long-term adoption trend. What matters for readers is the scope of the content itself: a browser-delivered anatomy atlas with demonstrations, quizzes, and sectioning, plus explicit provenance and licence caveats for third-party models. The NC exceptions; research only marker is a concrete reason to check asset rights before reusing or publicly redistributing the models, even though the project's own code is MIT-licensed.
Written by AI from the project's README, updated . The project's own docs are the source of truth.