Following is a guest post by Bill Rapp:
With all the focus on making healthcare more efficient, medical imaging procedures like MRIs and CT scans are a natural place to start. That’s why IBM and the Mayo Clinic are working to automate the detection of deadly brain aneurysms.
Traditionally, a patient suspected of having a brain aneurysm due to a stroke, traumatic injury or family history would undergo an invasive test using a catheter, a technique with risks of neurologic complications. To improve detection using a non-invasive magnetic resonance imaging approach, the Mayo Clinic and IBM worked to create automated detection algorithms to help radiologists pinpoint the location of aneurysms in the brain from MRI images.
The new algorithm is 95% effective at identifying likely aneurysms compared with 70% effectiveness for manual interpretation by radiologists. Radiologists still make the final call, but their attention can be focused on the most likely trouble areas, making the reading process faster and more accurate.
The key to this innovation was complete automation of the imaging and detection process, eliminating the need for human guidance to start the detection procedure. All of this must be done in real time with the results delivered to the radiologist in a timely and predictable manner.
This project sets the stage for introducing other automated detection techniques in the future. This framework can be extended to other imaging modalities like CT scans, other body parts such as the liver and kidney and other diseases like cancer. We’re also working on more flexible and affordable ways to deliver this smart technology, for example, through cloud computing.
Fully automated detection, flexibly delivered to the point of care is the future of radiology and other medical procedures and is a great example of smarter healthcare.
Bill Rapp, IBM’s CTO of Healthcare and Life Sciences and co-director of the Medical Imaging Informatics Innovation Center.