Concentration of Imaging and Numerical Data:
Urology Shows High Compatibility with AI


Legal Liability Vacuum and Alert Fatigue:
The Need for "Silent Intelligence"


"The Final Decision Lies in the Physician-Patient Relati

"The true role of medical AI is not to replace specialists, but to assist in clinical decision-making. Its core value lies in eliminating 'variability'—which arises from subjective interpretation and physician fatigue—thereby ensuring equal quality in medical care."


Won Junhee, Clinical Professor of Urology at Weill Cornell Medicine, spoke at a media interview for "Medical Korea 2026" hosted by the Korea Health Industry Development Institute on the 20th at COEX in Seoul. Korea Health Industry Development Institute

Won Junhee, Clinical Professor of Urology at Weill Cornell Medicine, spoke at a media interview for "Medical Korea 2026" hosted by the Korea Health Industry Development Institute on the 20th at COEX in Seoul. Korea Health Industry Development Institute

View original image

Won Junhee, Clinical Assistant Professor of Urology at Weill Cornell Medical College and former President of the Korean American Medical Association, stated this during a media interview for the global healthcare & medical tourism conference "Medical Korea 2026," hosted by the Korea Health Industry Development Institute at COEX in Seoul on March 20.


The concept Professor Won presents as the key value of AI in medicine is "diagnostic democratization." This means that even junior doctors with less experience can achieve diagnostic performance comparable to that of fellowship-trained experts when assisted by AI. AI reduces the variability in diagnostic quality that previously depended on an individual physician's level of experience or condition.


In urology, which handles various imaging and numerical data such as MRI, CT, and biomarkers, this effect is especially pronounced. For example, in prostate cancer, AI-based MRI readings can predict the likelihood of cancer in advance, thereby reducing unnecessary biopsies. In cases of benign prostatic hyperplasia, AI integrates and analyzes multiple variables such as urinary flow and symptoms to predict prognosis. He explained, "By applying machine learning to data from 5,000 patients, it becomes possible to make specific clinical judgments, such as determining there is a 72% probability that surgery will be needed within two years."


Beyond diagnosis, the adoption of AI is progressing in the operating room as well. In robotic surgery systems like da Vinci, AI serves as a sensor, issuing warnings when surgical instruments dangerously approach major blood vessels or tumor boundaries. By analyzing subtle patterns in surgical movements, AI can also predict the risk of postoperative complications such as incontinence in advance.


However, Professor Won also pointed out challenges that hinder the establishment of AI in medicine. The first is the "liability vacuum." For instance, if AI recommends that a bladder cancer patient be monitored every nine months instead of every three months, and the cancer recurs, it is unclear whether responsibility lies with the physician, the hospital, or the software company. Alert fatigue is another variable. Professor Won noted, "If notifications become too frequent, physicians may lose focus and eventually turn off the AI altogether." Ultimately, his conclusion is that the ideal medical AI should operate unobtrusively in the background, intervening only at critical moments as a form of "silent intelligence," without disrupting the clinical workflow.



No matter how sophisticated AI becomes, Professor Won maintains that the final decision must rest with physicians and patients. While medicine is a precise science, it is also an area shaped by extensive clinical experience and judgment, and accuracy alone cannot explain or resolve everything. He emphasized, "The final decision must arise from the relationship between the patient and the physician," adding, "Above all, a close relationship and open communication are the most important factors."


This content was produced with the assistance of AI translation services.

© The Asia Business Daily. All rights reserved. Unauthorized AI training and use prohibited.

Today’s Briefing