GR US

Liaw and Kakadiaris on Primary Care Artificial Intelligence

The National Herald

Maj. Thomas Beachkofsky, 6th Healthcare Operations Squadron dermatologist, uses a body scanner microscope to take a picture of a spot on his arm at MacDill Air Force Base, FL, Oct. 30, 2019. A new software upgrade allows a complex algorithm to analyze an image captured with a camera and rate the severity of the spot for a dermatologist to review. (Photo: Public domain/ U.S. Air Force photo by Senior Airman Adam R. Shanks)

HOUSTON, TX – Dr. Winston Liaw and Dr. Ioannis A. Kakadiaris, both from the University of Houston (UH), published an article in The Annals of Family Medicine titled Primary Care Artificial Intelligence: A Branch Hiding in Plain Sight.

In the article, Liaw and Kakadiaris write that primary care artificial intelligence “should aim to improve care delivery and health outcomes; using this benchmark, it has yet to make an impact,” AI in Healthcare reported.

The authors cited “the lack of engagement from the primary-care community as a prime reason for the disappointing showing to date,” and “suggest the widespread reticence has real-world consequences,” AI in Healthcare reported.

“Without input from primary care,” they point out, healthcare AI researchers “may fail to grasp the context of primary care data collection, its role within the health system and the forces shaping its evolution,” AI in Healthcare reported.

Liaw and Kakadiaris noted the AI challenges facing primary care, AI in Healthcare reported:

Inefficient data entry- “Without timely data, artificial intelligence systems do not have the information they need to make decisions,” the authors write.

Poorly processed data, “because researchers mistrust the accuracy of the data that does get entered in primary care, the understandable tendency is to ‘omit or modify data according to arbitrary or inappropriate rules, which can lead artificial intelligence systems to learn the wrong lessons,’” AI in Healthcare reported.

Unexplained (“black box”) AI results, Liaw and Kakadiaris write, “For users to trust artificial intelligence systems, they need to understand why decisions are made.”

Magnification of existing biases- “The systematic under- or over-prediction of probabilities for populations emerges for multiple reasons, including biased training data and outcomes influenced by earlier, biased decisions.”  

Siloed data- “This leads to tools that perform worse when used at different institutions. Furthermore, the population on which the tool was trained may shift, causing its performance to suffer over time.”

Privacy concerns- “With the digitization of data, patients are increasingly unable to determine when, how, and to what extent information about them is communicated to others. Breaches and misuse erode trust in artificial intelligence systems and may make individuals reluctant to access care,” the authors write.

AI in Healthcare noted Liaw and Kakadiaris conclusion that “we do not simply need the application of artificial intelligence to primary care, but rather, the development of new methods that are tailored to the breadth, complexity and longitudinality of primary care,” and “generalists are set apart by our overriding interest in people, an interest that is vital to the creation of a bond between physician and patient.”

“To this the authors add that the proliferation of electronic health records (EHRs) and, with it, the rise of AI, threaten this bond by adding ‘more and more layers’ of technology,” AI in Healthcare reported.

In order to prevent this challenge from becoming a serious problem, primary care AI “needs to narrow this divide by facilitating new opportunities for connection” between primary-care researchers and academic AI experts, the authors write, adding that “finding creative solutions to this challenge is necessary if we hope to restore the relationships that sustain us and our patients,” AI in Healthcare reported.

Dr. Winston Liaw, MD, MPH, is chair of the Health Systems and Population Health Sciences department at the University of Houston College of Medicine. His expertise is in using geospatial techniques and community resources to address unmet social needs in primary care settings. In particular, he is an expert on the application of neighborhood deprivation indices to primary care delivery and using geospatial tools to teach population health concepts.

Born in Greece, Dr. Ioannis A. Kakadiaris, PhD, is a Hugh Roy and Lillie Cranz Cullen University Professor of Computer Science, Electrical & Computer Engineering, and Biomedical Engineering at the University of Houston. He joined UH in August 1997 after a postdoctoral fellowship at the University of Pennsylvania. He earned his BSc in Physics at the University of Athens in Greece, his MSc in Computer Science from Northeastern University and his PhD at the University of Pennsylvania. He is also the founder and director of the Computational Biomedicine Lab. His research interests include biometrics, computer vision, and pattern recognition, biomedical image analysis and cardiovascular informatics.