- Fractal introduces Vaidya.ai, a groundbreaking multimodal medical LLM for diagnosis, treatment, and advice.
- Trained on vast datasets, Vaidya.ai understands multiple Indian languages, including Hindi, Kannada, Tamil, and Telugu.
- The AI is not intended to replace doctors but to provide reliable healthcare information.
- Currently, Vaidya.ai is an AI research project, not a commercial product.
- Vaidya.ai demonstrates an 83% accuracy rate, ranking within the top 500 based on PG NEET exam evaluations.
- Development involved LLM and VLM architectures validated by over two decades of medical exam data.
- The platform encourages consultation with healthcare professionals for medical advice.
Main AI News:
Fractal, a leading AI firm, has introduced Vaidya.ai, India’s first multimodal medical Large Language Model (LLM). Positioned as a general-purpose assistant, Vaidya.ai is designed to aid users with diagnostics, treatment options, and medical guidance.
Leveraging open-source models with parameters ranging between 30 to 70 billion, Vaidya.ai has undergone extensive training on over 650,000 images and 200,000 text inputs. This robust training enables the AI to understand and process multiple languages, including Hindi, Kannada, Tamil, and Telugu, according to co-founder and CEO Srikanth Velamakanni.
Velamakanni emphasized that while AI advancements are significant, they are not poised to replace medical professionals in the foreseeable future. “It’s not even in the foreseeable future that we will have a situation where we don’t need doctors, and AI may tell you what to do. AI is far from that. The problem we are solving is very different. It’s about aiding consumers in accessing healthcare information which is more reliable than any other sources of information,” he said.
In addition to Fractal’s existing product, Qure.ai —which is known for its ability to detect conditions such as tuberculosis, lung cancer, heart failure, and stroke—Vaidya.ai holds promise as a unique enterprise solution. However, Velamakanni clarified that, as of now, Vaidya.ai remains a research-focused project, not yet available as a commercial product.
Regarding the AI’s accuracy, Velamakanni noted that Vaidya.ai’s performance is evaluated against the highly competitive PG NEET exam questions. Suraj Amonkar, Fractal’s Chief AI Research and Platforms Officer, added that the development process involved training models with LLM and VLM architectures and validating them against over 20 years of PG NEET and PG Medical examination papers.
Conclusion:Â
The launch of Vaidya.ai by Fractal signifies a significant step forward in integrating AI into the healthcare sector. By providing a reliable source of medical information in multiple Indian languages, it addresses a critical need in the market for accessible, trustworthy healthcare assistance. While it remains in the research phase, the accuracy and potential of Vaidya.ai suggest it could become a valuable tool in the healthcare ecosystem, complementing human expertise rather than replacing it. This development could set a precedent for further advancements in AI-driven healthcare solutions, opening up new opportunities and challenges in the market.
