HITInfrastructure

Latest News

AMA: Clinical Decision Support One of Many Healthcare AI Uses

Clinical decision support is one of many potential healthcare AI uses, observed a recent American Medical Association report.

artificial intelligence

Source: Thinkstock

By Fred Donovan

- Clinical decision support is one of many potential healthcare AI uses, observed a recent American Medical Association (AMA) report.

Other healthcare artificial intelligence uses include patient monitoring and coaching, automated devices to assist surgery or patient care, medical imaging, precision medicine, predictive analytics, and management of healthcare systems.

AMA said it prefers to use the term augmented intelligence (AI), which combines artificial intelligence methods and techniques and their impact on human decision-making capabilities.

“Software algorithms developed using these evolving methods and techniques, coupled with proliferating sources of data (datasets) pertinent to health and medicine, offer the promise of new and more powerful ways to augment human intelligence and expertise in health care,” according to the report.

AMA noted that AI could improve a physician’s ability to diagnose disease and gauge prognosis.

“In health care, the next three to five years will be marked by efforts to scale AI options involving patient-centered wearables that support clinical care, improved tools for diagnosis and physician training, and health system initiatives to improve patient care and clinical decision support,” the report predicted.

AMA detailed several early examples of AI in healthcare.

AI could improve the performance of wearable monitoring devices. In one study, an application enhanced by AI was developed to work with the Apple Watch to monitor patients with hypertension and sleep apnea. The application and its AI system, DeepHeart, was able to detect hypertension with 82 percent accuracy and sleep apnea with 90 percent accuracy.  

Machine learning algorithms are being used to boost clinical decision making. For example, the nonprofit Human Diagnosis Project (Human Dx) enables physicians to ask for assistance on problematic medical cases from an online physicians' community. Machine learning combines the responses to create a synthesized collective assessment for each case.

“This collective insight is designed to augment clinical decision making with machine intelligence, providing useful information to physicians and patients who may not otherwise have access to specialist expertise,” the report related.

In addition, the Human Dx is being used as a platform for medical education through the Global Morning Report teaching cases. Residents from more than 40 percent of internal medicine residency programs in the United States have access to these cases. Human Dx checks the quality of the response by comparing them with how physicians would solve the training cases in order to calculate a quantitative measure of reasoning, known as the Clinical Quotient.

AI is also being combined with health system data to improve care. For example, the University of Pittsburgh Medical Center (UPMC) has developed an algorithm that identifies hospital patients who are at high risk of re-hospitalization within 30 days of discharge. In a pilot project using one patient unit, the algorithm was able to cut re-hospitalization in half.

UPMC has also used machine learning to reduce false positive for lung cancer testing. UPMC said that this is the first time machine learning has been used to sorting out benign from cancerous nodules in lung cancer screening.

“We were able to rule out cancer in about a third of patients, so they wouldn’t need biopsies, they wouldn’t need PET scans or a short-interval CT scan. They just need to come back in a year,” said David Wilson, co-director of the UPMC Hillman Lung Cancer Center and an associate professor of medicine, cardiothoracic surgery, and clinical and transformational science at the University of Pittsburgh.

The AMA report noted that “healthcare AI can offer a transformative set of tools to help patients, physicians, and the nation face these looming challenges.”

“Given the number of stakeholders and policymakers involved in the evolution of AI in healthcare, it is important that our AMA not only adopt a base level of policy to guide our engagement, but equally continue to refine our policy as an organization to ensure that the perspective of physicians in various practice settings informs and influences the dialogue as this technology develops,” it concluded.

X

Sign up for our free newsletter covering the latest IT technology for Hospitals:

Our privacy policy


no, thanks

Continue to site...