Researchers at John’s Hopkins are building a searchable digital library of MRI scans with the goal of giving providers a “Google-like search system that will enhance the way they diagnose and treat young patients with brain disorders”. The cloud-computing project, developed by a team of engineers and radiologists, allows physicians to access thousands of pediatric scans to look for some that resemble their own patient’s image.
Here’s how it works
A doctor reviews a brain scan and notices something abnormal that she can’t quite identify. She can enter the scan into this searchable database to find other similar scans. She can then review the de-identified medical records of patients with scans that are a match to see the diagnosis and treatment the patient received.
“This research is one of the first real applications of ‘Big Data’ analytics, taking medical information from large numbers of patients, removing anything that would identify specific individuals, and then bringing the data into the ‘cloud’ to allow very high-powered analysis,” said Jonathan Lewin, the chairman and radiologist-in-chief of the Johns Hopkins Department of Radiology and Radiological Science.
The database is another example of how computers are being integrated into the healthcare space. Last year, two researchers at Indiana University successfully used artificial intelligence to create computers that can “think like a doctor”. Similarly, IBM’s Watson can analyze symptoms and other test results to come to a diagnosis. It’s a trend that is likely to continue. Just last year, Google announced it was getting into the healthcare space, investing in a venture that combats aging and disease. Although the John’s Hopkins project is comparable to a Google search, it is not connected to the search engine’s efforts. Even with all this computer integration, physicians shouldn’t worry about being replaced any time soon. Instead, they can look forward to computers making them better healthcare providers.
“The scans in our library may help a physician identify a change in the shape of a brain structure that occurs very early in the course of a disease, even before clinical symptoms appear,” said Michael I. Miller, a lead investigator on the project. “That could allow the physician to get an early start on the treatment.”
Right now, the database is only open to physicians in the John’s Hopkins network, and it is only comprised of pediatric brain images. However, researchers say the database could be expanded or replicated elsewhere. They have also begun building a similar MRI brain image library with brain disorders commonly found in elderly patients.