What if Biopharma gets access to billions of patient records sitting in the vaults of Hospitals and Clinics? Will this accelerate clinical trials? Bring drugs to market faster? In this post, I want to highlight a story of how a Health Care Provider and an AI Tech company joined hands to solve this complex problem.
The challenge with Biopharma research has always been access to real world patient data. They have all been locked in the clinic and hospital vaults AKA Epic’s and Cerner’s of the world. With additional layers of security in the name of HIPAA, Patient privacy and data residency, these are some of the hard to get data.
Mayo Clinic is a nonprofit medical center focused on both Health care and research. Mayo sees more than 1 million patients a year and they have 10 million patient records on hand. Mayo wanted to use this enormous wealth of patient data to improve care and help the research community. However this is no easy task, given the regulatory challenges including HIPAA, Patient privacy, data residency, etc.
Mayo joined hands with nference, an AI Tech company to solve this problem. Nference helped Mayo deidentify the patient data and host them in a secure containerized cloud platform. On top of this, they also worked with third party audit firms to certify that the data is properly deidentified and is secure and compliant with HIPAA and other regulations.
👉 The next part of this puzzle is - How to extend the insights from this platform to the larger Biopharma research community?
nference solved this problem by acting as a conduit that connects Mayo’s clinical platform with the Biopharma industry. Nference designed a solution that allows multiple entities to build a common artificial intelligence model without sharing raw data using their Nferx platform. Thus Biopharma can access only the insights from the platform, and there are strict measures in place to prevent reidentification of data.
⚡"The combination of curated, deidentified clinical data and advanced analytics technology will accelerate scientific discovery and development of new therapeutics" says Venky Soundararajan, Chief Scientific Officer, Nference Inc.⚡
💡💡💡 Food for thought - In my previous post, I highlighted Model as a Service (MaaS) and how it will have a revolutionary impact similar to SaaS. What if Clinics around the world sitting on enormous patient data build specialized ML models trained on real world data? For example, a Biopharma getting into a new therapeutic area such as oncology can start from the model instead of starting from scratch. A true win-win for the industry.
This post is inspired by the HLTH 2022 keynote by John Halamka, M.D., M.S., the president of Mayo Clinic. Sources will be updated in comments.
Image generated using #DALLE AI
Subash Rajavel is the founder and Chief editor at xGenom. He has 15 plus years of experience in Health care and Life Sciences. His primary focus are Patient centric solutions, Digital Health and Next-gen Clinical Trial solutions.
You can reach him at subash@xgenom.com
Comments