GE Healthcare and Partners HealthCare announced today a 10-year collaboration to integrate artificial intelligence into the complete patient care continuum.

The plan will be implemented and executed through the newly formed Massachusetts General Hospital and Brigham and Women’s Hospital Center for Clinical Data Science, both of which are members of the Partners health system.

“The fundamental thing that’s changing in medicine, above everything else, is that we are literally inundating our clinicians with data of all kinds as medicine become more sophisticated,” Dr. David Torchiana, CEO of Partners HealthCare, said in a roundtable discussion. “Rather than enhancing the quality of care, sometimes it seems to complicate and render care more difficult and more confusing.”
Their initial focus will be on developing applications based on deep learning technology for the diagnostic imaging field. These applications will help clinicians make stroke prognoses, identify fractures in the emergency room, track tumor response to therapy and diagnose cancer using ultrasound data.

A lumbar spine MR exam can generate up to 300 images for a single spinal injury patient, and radiologists may have to review prior scans and notes in the EHR to make a diagnosis. A deep learning application would be able to rapidly analyze the data to determine the most important images for the radiologist to read.

In the future, Partners HealthCare and GE plan to develop products for other medical specialties such as molecular pathology, genomics and population health. That has the potential to help reduce unnecessary biopsies, streamline clinical workflows and lessen the amount of time clinicians spend performing administrative tasks, so they can have more time with patients.

They also plan to co-develop an open platform that they and third-party developers can prototype, validate and allows them to share applications with hospitals and clinics globally.

Torchiana said that the most important aspect of this partnership is translating these applications into commercial-grade solutions. That’s where GE comes in, since the company has a presence in virtually every hospital in the U.S.

“[We] have to be able to get this technology into the front lines and into care delivery,” said Torchiana. “The commercialization part of this is what’s going to enable this to change people’s lives.”

But translating these applications into clinical practice is not going to be an easy feat. Dr. Mark Michalski, executive director of the MGH & BWH Center for Clinical Data Science, said the major challenge is that it requires a large amount of computation and clinical data.

Over the past five years, Michalski and his team have deployed an entire infrastructure of partners, known as eCare, to gain access to a considerable amount of data. He referred to it as “a challenge but an opportunity.”

“We have seen some incredible things happening in machine learning and artificial intelligence with self-driving cars and smartphones,” said Dr. Keith Dreyer, chief data science officer at MGH and BWH. What this partnership is about, is taking that technology and translating it effectively so we can get excited about it within health care.”

Business Affairs Homepage