"Pain Differs for Everyone"... Quantifying Chronic Pain Intensity with Brain Imaging [Reading Science]

IBS Analyzes Patient-Specific "Brain Fingerprints" with AI

A Major Step Forward for Precision Diagnostics

The path has been opened to quantify the intensity of chronic pain in individual patients using their own brain imaging. Moving away from the traditional approach of seeking common pain signals, this is the first case where artificial intelligence (AI) was used to analyze patient-specific "brain fingerprints" to predict pain intensity. This is expected to mark a turning point in developing personalized precision diagnostics and treatment strategies.

Woo Chungwon, Associate Director of the Center for Neuroscience Imaging at the Institute for Basic Science (IBS) and Associate Professor at Sungkyunkwan University’s Department of Global Biomedical Engineering, together with Professor Cho Seonggeun of Chungnam National University, led a joint research team that announced their success in precisely predicting the pain intensity perceived by chronic pain patients through brain imaging alone, by analyzing each patient’s unique brain function patterns.

Importance of brain regions as markers of chronic pain based on brain imaging. The color of each region reflects the importance value, representing how much the prediction accuracy decreases when that region is excluded. The top 5 regions with the highest importance are separately indicated, and these core regions varied among participants. Provided by the research team

Chronic pain is a common condition affecting one in five adults, but unlike blood pressure or body temperature, there has been no objective indicator for measurement, limiting diagnosis and treatment. In particular, since chronic pain can occur even without external stimuli, test results often appear "normal," resulting in a vicious cycle where symptom relief is prioritized over fundamental treatment.

The research team conducted repeated functional magnetic resonance imaging (fMRI) scans over several months on patients with fibromyalgia, a condition characterized by widespread, persistent pain throughout the body. fMRI is a device that measures activated brain regions based on changes in blood flow. The team analyzed this data using AI machine learning to derive each patient’s "functional connectome"-a map of brain connectivity.

As a result, the newly developed biomarker was able to predict changes in pain intensity experienced by patients over several months with high accuracy, using only brain imaging data. Notably, pain patterns identified in one patient did not apply to others. This scientifically demonstrated that pain responses are as unique to individuals as fingerprints.

The team adopted a strategy of "collecting enough repeated data from a single individual" rather than relying on group averages. When the number of scans exceeded four to five, prediction accuracy improved significantly, demonstrating the effectiveness of a personalized approach.

Woo Chungwon, Associate Director at IBS, stated, "This study shows that invisible chronic pain can be comparatively objectified and quantified through brain imaging," adding, "We have laid the foundation for precision medicine that will lead to the development of patient-tailored treatments." First author Dr. Lee Jaejung, Postdoctoral Researcher, commented, "The key finding is that the brain connectivity related to pain appears uniquely in each patient," and added, "We aim to develop this into a brain-based precision diagnostic tool applicable in clinical practice."

The results of this study were published online in the international journal Nature Neuroscience on February 26, 2026.

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