▲(From Left) Professor Hangsik Shin and Researcher Gayeon Ryu of the Department of Convergence Medicine and Professors Byung-Moon Choi and Jae Moon Choi of the Department of Anesthesiology and Pain Medicine at Asan Medical Center

Artificial intelligence is expected to help objectively measure pain levels during and after surgery, enabling customized pain management for patients.

A team composed of Professor Hangsik Shin and Researcher Gayeon Ryu of the Department of Convergence Medicine, and Professors Byung-Moon Choi and Jae Moon Choi of the Department of Anesthesiology and Pain Medicine at Asan Medical Center has developed a new pain assessment methodology using machine learning algorithms aimed at analyzing subjective patient-reported pain levels based on objective indicators.

The essence of the study is that when pain occurs, the autonomic nervous system undergoes subtle changes, such as an increase in heart rate and contraction of peripheral blood vessels. When identified using photoplethysmography (PPG), which detects changes in blood volume in the microvascular bed, the extent of pain can be determined quickly.

▲ The medical staff of the Department of Anesthesiology and Pain Medicine at Asan Medical Center is monitoring the patient's vital signs during surgery.

The research team studied 242 patients who underwent various surgical procedures by measuring the changes in pain levels obtained from their blood pressure, heart rate, and PPG signals. After selecting 6 key features, including PPG area changes and pulse interval variability, that significantly contributed to pain prediction, they applied them to a machine-learning model to confirm the degree of pain occurrence. As a result, the accuracy of intraoperative pain assessment remained at 83%, consistent with the accuracy of the existing pain assessment model. Additionally, the accuracy of postoperative pain prediction was 93%, significantly improved from 58%.

The study findings were recently published in the international academic journal ‘npj Digital Medicine.’