Imaging clinicians want a bigger role in health care, one that allows them a say in patient management. Ideally it would be a role that goes from diagnosis to clinical procedure and continues through follow-up care. They will get this role only if they can demonstrate their involvement adds clinical value, improves patient outcomes and can validate efficiencies that will drive down costs while ensuring maximized patient billing reimbursements. Artificial intelligence (AI) may be the pathway to such a role and it also holds the potential for improved diagnosis.
Perhaps one day intelligent machines utilizing the IBM “Dr. Watson” technology can take the reins during the exam itself to optimize scan protocols on the fly to hone in on pathology. Tapping into streams of imaging data, “Watson” might look for signs of disease and adjust scan parameters to optimize data acquisition, but are smart machines what imaging modalities need? Are they even practical for use in the United States?
Intelligent machines will encounter a major hurdle in the form of the U.S. Food and Drug Administration (FDA). As the first of its kind, these machines will lack the “predicate” devices needed to be regulated under the FDA’s 510(k) system. An example of the enormity of this challenge is illustrated by how difficult it has been for companies making computer-aided detection algorithms. This hurdle alone keeps research groups engaged as they learn and improve their applications.
A “hot topic” at HIMSS 2017 this month in Orlando, Florida, will be the continued exploratory focus of AI, learned machines and their tie into predicative analysis. Several educational track sessions at HIMSS 2017 will be speaking to this topic. In addition, many product vendors will launch their application solutions in the exhibit hall.
Regardless of whether machine- or human-based aids are leveraged, imaging needs such aids. The progression of this advancing imaging timeline in improving patient outcome performance is very important to the future of health care.