AccelRare, an AI-powered pre-diagnosis tool for rare diseases that was recently launched in India by Sanofi, aims to help doctors identify conditions that often go undiagnosed for years by generating a list of possible diagnoses in minutes. The web-based platform is designed for healthcare professionals and can generate five possible rare disease diagnoses in five to ten minutes based on a patient’s symptoms, medical history and laboratory findings. Sanofi says the platform covers 310 rare diseases and has a diagnostic reliability of around 88-89%.The launch comes at a time when AI is increasingly finding applications across healthcare, from medical imaging and clinical documentation to disease screening. In India, companies ranging from startups to global technology firms are exploring AI-based diagnostic tools, while hospitals are beginning to pilot large language models in clinical workflows.Rare diseases present a particularly difficult challenge. Although an estimated 72-96 million Indians are believed to be living with one, each individual disease affects a relatively small number of patients, making diagnosis difficult even for experienced physicians. According to the Indian Organisation for Rare Diseases, 43% of doctors in India have never encountered a rare disease patient.”The average patient takes six to eight years to receive a confirmed diagnosis,” said Dr. Mayur Shah, Head of Rare Diseases Business at Sanofi India, during a media interaction. “Most rare diseases present with symptoms that overlap with common illnesses, which is why patients often undergo years of investigations before reaching the right diagnosis.”Unlike consumer-facing AI chatbots, AccelRare is not intended to diagnose patients directly or replace clinical judgement. Instead, it functions as a clinical decision-support tool for doctors.Healthcare professionals enter symptoms, examination findings and laboratory results into the platform, after which the algorithm generates five possible rare disease diagnoses ranked by probability. Doctors must still order confirmatory tests before arriving at a final diagnosis.One of the first questions surrounding any AI system in healthcare is reliability, particularly given the well-documented issue of hallucinations in generative AI models.Addressing this, Shah said the platform does not operate as a general-purpose large language model. Instead, it uses a score-based algorithm built on MedVir, a medical knowledge base of peer-reviewed literature, and has been validated by 67 rare disease experts across 13 medical networks. Because it produces differential diagnoses rather than definitive answers, the company argues the risk profile differs from that of conversational AI systems.Dr. Milan Choksey, Medical Lead for Rare Diseases at Sanofi India, added that the algorithm continues to improve as more clinical cases are processed through the system, though he stressed that it is intended to augment, not replace, a physician’s clinical assessment.The platform is a proprietary product developed by MedVir, with Sanofi acting as its distribution partner in India. The company said the tool is available free of cost to healthcare professionals and does not follow a revenue model. It has also been registered in India as a medical device.One concern around AI models trained largely on Western datasets is whether they perform equally well in India. Shah acknowledged that the underlying knowledge base currently draws more heavily from Western populations but said the model is designed to improve as more Indian doctors use it. He also noted that the platform relies primarily on clinical symptoms and examination findings rather than patient images or demographic identifiers, making it less susceptible to regional variations.The company has also chosen not to collect personally identifiable patient information. According to Sanofi, the tool anonymises all cases, a decision it says was taken to strengthen patient privacy while making the platform easier to deploy across healthcare settings.Sanofi is also engaging with government doctors as part of its rollout, although the company said it is not currently working with the government on deploying the platform. It hopes wider adoption among public healthcare professionals could eventually pave the way for broader institutional use.Since its launch in May, around 600 clinical questionnaires have been initiated on the platform, with roughly 300 completed assessments and about 29% repeat usage, according to the company. While still early, Sanofi says the figures indicate that doctors are returning to the tool after initial use.For Sanofi, the launch also reflects a broader shift in how pharmaceutical companies are approaching AI. Rather than limiting its use to drug discovery or clinical trials, the company is beginning to deploy AI earlier in the patient journey, where identifying a disease sooner could have a significant impact on treatment outcomes.Add 
as a Reliable and Trusted News Source
Add Now!
Read More News on
AI in rare disease diagnosisAccelRareSanofiIndia rare disease diagnosis
(Catch all the Business News, Breaking News, and Latest News Updates on The Economic Times.)
…moreless
Read More News on
AI in rare disease diagnosisAccelRareSanofiIndia rare disease diagnosis
(Catch all the Business News, Breaking News, and Latest News Updates on The Economic Times.)
Subscribe to The Economic Times Prime and read the ET ePaper online.
…moreless
