By Shahram Ebadollahi
In IBM Watson’s early days, the cognitive computer was a whiz at words. It was designed to ingest vast amounts of documents and Web pages, understand words and their context, and answer free-form questions from people–offering up responses ranked by its confidence in their accuracy.
These days, we’re adding a wide variety of other types of data to Watson’s repertoire, perhaps most significantly, images–including photos, medical images and videos. Simply put, we’re teaching Watson to “see.”
A watershed moment in our effort to expand Watson’s visual capabilities comes today: we’ve announced our intention of acquiring Merge Healthcare Incorporated, a leading provider of medical image handling and processing systems. It addresses radiology, cardiology, orthopedics eye care and other medical fields. The planned acquisition is subject to regulatory review and Merge shareholder approval and is anticipated to close later this year.
After the deal closes, we’ll begin to bring Watson imaging technologies to Merge’s portfolio of products so medical professionals have the opportunity to get help from Watson when analyzing X-rays, angiograms, electrocardiograms and other images to spot anomalies and identify treatment options.
The new capabilities are aimed at enabling medical researchers and clinicians to better understand diseases and to help them design therapies that address the needs of individual patients. We also expect it to bolster doctors’ and researchers’ ability to run cohort analytics using Watson, gaining insights from patients with similar issues.
It’s all part of our core vision for IBM Watson Health. We provide cognitive computing technologies that generate insights drawn from medical data at a massive scale. Our purpose is to enable medical professionals and patients everywhere to make better-informed decisions, which can lead to better health outcomes. The goal is nothing less than to transform the way healthcare is practiced and life science research is conducted.
The acquisition of Merge is particularly gratifying to me because of my professional background. I got my PhD at Columbia University in multimedia content analysis and retrieval–with a focus on cardiology images. Since then, I have worked on a variety of image analytics projects at IBM Research and led our effort to develop a research strategy for drawing insights from electronic medical records.
By bringing Merge into the Watson family, we aim to meld one of the most significant forms of electronic medical data–images–into the electronic medical record using one of the most powerful new analytics technologies: cognitive computing. Medical images are used by clinicians for a wide variety of diagnoses and health-monitoring scenarios, so our work has the potential to touch many people’s lives.
Since IBM Watson Health was announced four months ago, we have acquired two companies and forged alliances aimed in part at providing people using Watson with access to a tremendous amount and variety of valuable data. For instance, Explorys, which we acquired in April, has a database containing the de-identified records of millions of patients. Through our alliance with Apple, users of Watson will be able to tap into data collected via Apple’s HealthKit and ResearchKit applications for gathering and sharing medical, health and lifestyle information–with the owners’ permission, of course.
These sources of health and medical information along with data streams coming from a host of new mobile fitness and health monitoring devices, start to provide medical professionals and others with a 360-degree view of the health of the people on the planet.
The implications are huge. With this wealth of different kinds of medical, behavioral and social data, we can create detailed computer models that aid researchers and medical professionals in their effort to understand the causes of disease, so they can diagnose them accurately, to help predict their courses, and to help clinicians develop personalized treatment plans. The aim: better outcomes for individuals and more cost-efficient use of healthcare dollars.
To achieve these goals, it’s essential for individuals, organizations and professionals across the healthcare continuum to share data and insights. That’s why we are building Watson Health Cloud. This data-sharing and application-development platform allows IBM clients and partners to safely and securely pool de-identified data from genomic and electronic medical records, clinical trials, and health-monitoring devices, and combine it with socio-economic and demographic information.
Watson Health Cloud promotes open innovation. IBM, business partners and healthcare organizations will be able to build products and services that combine access to a wide variety of data with Watson’s cognitive analytics capabilities.
This is the wave of the future. We’re beginning to tap into the collective memory of the healthcare system and mine it for insights. Digital images–combined with the ability to interpret them using computers–will be an essential piece of this new world-transforming capability. Ultimately, every other domain, profession and industry will follow the same path.
To learn more about the new era of computing, read Smart Machines: IBM’s Watson and the Era of Cognitive Computing.