The Department of Veterans Affairs and precision medicine startup Flow Health have inked a 5-year deal to apply deep learning to medical decision-making. The goal of the partnership is to understand why some people are susceptible to certain diseases and to use artificial intelligence to select the most appropriate treatments for them.
The collaboration will see the pair integrating vast amounts of data to learn how each gene variant in the genome affects phenotype, Flow Health said in a statement. They will use this knowledge to determine an individual’s disease risk as well as to make more precise diagnoses and recommend treatments.
“Developing artificial intelligence which can automatically identify the best diagnostic and treatment pathways will assist clinicians in delivering precision medicine to every veteran," said Dr. Robert Rowley, Flow Health chief medical officer, in the statement. But to get to the point where AI can personalize care, deep learning models require “huge amounts” of data.
Flow Health aims to integrate more than 30 petabytes–30 million gigabytes–of data from 22 million veterans over 20 years into a “knowledge graph” of medicine and genomics, according to the company. Its precision medicine platform combines structured and unstructured data from a range of sources into one system and organizes it around relationships with people, places and events, the company said.
This is just the VA’s latest step in precision medicine. The agency is participating in the Precision Medicine Initiative by carrying out the Million Veterans Project, an effort to understand how genes affect health. Data is being collected from veterans receiving care from the VA healthcare system and will be anonymized for further research into diseases and conditions including diabetes, cancer and post-traumatic stress disorder.
The VA also teamed up with IBM earlier this year, in a project where it will use the IBM's Watson for Genomics technology to scale up its cancer precision medicine program.