Cognitive Computing: Improving X-Rays

Posted on

“What if your X-ray could predict a potential disease months in advance,” writes Laura Lorenzetti for Fortune. This past summer IBM Watson Health created a new partnership with leading medical providers and imaging tech companies to see if cognitive computing can take medical imaging one step further to actually predict the chance of diseases like cancer and heart failure.

Ms. Lorenzetti’s article points out that much of the data gathered from an x-ray or MRI is “unstructured.” It can be difficult for computers to connect the information to patient records in a meaningful way. IBM Watson Health is trying to change that and utilizing its power to connect unstructured data with its massive databases of patient medical history.  Ms. Lorenzetti writes, “the goal is to provide new offerings across various medical environments (a hospital ER or an everyday physician’s office) that can connect systems (medical records, picture archiving, lab results) and deliver cognitive insights to doctors on the spot for better diagnoses.”

One example from the article is the use of mammograms. Not only could Watson connect the image results to the patient’s medical history but it could also “cross-reference against the similar patients within the Watson database.”  These connections could improve a doctor’s  ability to identify early warning signs or risks.

Another example is the use of cognitive computing to help doctors predict which patients are more likely to have a heart attack after reporting chest pain.  Ms. Lorenzetti reports that 2% of patients who visit an ER with chest pain have the early signs of a heart attack missed. By connecting the data dots IBM Watson Health could help doctors identify these signs better.

IBM Watson Health’s shear power to process unstructured data such as medical imaging while also consuming vast amounts of patient data allows for it to draw cognitive insights that will one day improve patient diagnoses and treatments.