IBM and Broad Institute launch project to predict cardiovascular disease

The three-year project aims to produce AI models that can analyze genomic data, health records and biomarkers to predict the onset of heart attacks and other common conditions.

IBM Watson, Harvard, and MIT working on algorithms that predict heart attacks The three-year project aims to produce AI models that can analyze genomic data, health records and biomarkers to predict the onset of heart attacks and other common conditions.

IBM Watson Health on Wednesday announced it's expanding its partnership with the Broad Institute, the research center led by MIT and Harvard, to attempt to build algorithmic models that can help clinicians predict cardiovascular disease. By combining genomic data with clinical data (including existing health records and patients' biomarkers), the project should lead to more robust polygenic scoring, otherwise known as genetic risk scoring.

Over an initial span of three years, researchers will work on building algorithms to identify specific health conditions like cardiac arrest and atrial fibrillation. They'll develop tools that can integrate various kinds of data for modeling, as well as the capability to apply models to patients from different health systems. Lastly, they'll need to ensure the insights gained can be adequately explained to doctors and patients.

The project's insights and tools will be shared with the greater research community, IBM said.  

IBM and the Broad Institute have worked together on similar initiatives. In 2016, they began a five-year, $50 million project using machine learning and genomics to better understand why cancers become resistant to therapies.

IBM Watson Health also announced Wednesday that it's investing $50 million over 10 years in joint research collaborations with Vanderbilt University Medical Center and Brigham and Women's Hospital (the teaching hospital of Harvard Medical School). The collaborations will focus on applying AI to major public health issues, with an initial focus on using natural language processing to make better use of electronic health records.


Must read


IBM also announced that two more health care organizations, Cigna and Sentara Healthcare, are joining the blockchain-based health utlity network that IBM rolled out last month. The network was established in partnership with Aetna, Anthem, Health Care Service Corporation and PNC Bank with the goal of improving interoperability in the health care industry.

Related stories: