ArogyaSutra multi-agent framework targets medical AI in Indic languages
Researchers introduce ArogyaSutra, an actor-critic multi-agent framework paired with the ArogyaBodha dataset to improve multimodal medical reasoning across seven major Indian languages.
Score breakdown
The framework and dataset directly extend multimodal medical AI to seven major Indian languages, addressing the lack of equitable AI-driven healthcare assistance in multilingual, low-resource settings like rural India that English-centric MLLMs cannot serve.
- 01ArogyaSutra is an actor-critic-based multi-agent framework for multimodal medical reasoning in Indic languages.
- 02ArogyaBodha is a multilingual multimodal medical QA dataset built from eight heterogeneous sources.
- 03The dataset covers 31 body systems, six imaging modalities, and 21 clinical domains.
The paper identifies a critical gap in AI-driven healthcare: while Multimodal Large Language Models (MLLMs) perform well in general domains, they struggle in specialized, multilingual, and low-resource settings. This is particularly acute in rural India, where patients often pose complex medical queries in native Indic languages and rely on multimodal inputs such as medical images — contexts that English-centric MLLMs are ill-equipped to handle.
The source code and dataset are publicly available at the project's GitHub Pages site.
To address this, the authors introduce two contributions. First, ArogyaBodha, a large-scale multilingual multimodal medical question-answer dataset constructed from eight heterogeneous sources, spanning 31 body systems, six imaging modalities, and 21 clinical domains across English and seven major Indian languages. Second, ArogyaSutra, an actor-critic-based multi-agent framework that integrates tool grounding with dual-memory mechanisms for step-wise, reasoning-aware decision making, and leverages stored actor-critic simulation trajectories for distillation.
Experiments demonstrate that both the dataset and the framework improve multilingual medical reasoning accuracy across all Indic languages evaluated. Ablation studies validate the individual contribution of each component. The source code and dataset are publicly available at the project's GitHub Pages site.
Key facts
- 01ArogyaSutra is an actor-critic-based multi-agent framework for multimodal medical reasoning in Indic languages.
- 02ArogyaBodha is a multilingual multimodal medical QA dataset built from eight heterogeneous sources.
- 03The dataset covers 31 body systems, six imaging modalities, and 21 clinical domains.
- 04ArogyaBodha spans English and seven major Indian languages.
- 05The framework integrates tool grounding with dual-memory mechanisms for step-wise, reasoning-aware decision making.
- 06Actor-critic simulation trajectories are used for distillation.
- 07Source code and dataset are publicly available at the project repository.
Topics
Summary and scoring are generated automatically from the original article. We always link back to the publisher and never republish images or paywalled content. Last processed Jun 13, 2026 · 08:58 UTC. How this works →