Machine learning can pinpoint specific activity patterns in the brain that could lead to more targeted therapies.
Coupling machine learning with resting-state functional MRI (rs-fMRI) can identify unique patterns of coordinated activity in the brains of people who have major depressive disorder.
Even when investigators used different protocols, they can still pinpoint these networks that are working together, according to new research. A team of investigators from the Advanced Telecommunications Research Institutes International in Kyoto, Japan published their work in PLOS Biology on Dec. 7.
Diagnosing major depression is not difficult, said the team led by cognitive neuroscience researcher Ayumu Yamashita, but treating it can be. If providers were better able to understand the brain networks that are connected to depression, they might be able to offer better treatments.
To date, machine-learning algorithms have been applied to finding such associations in people with depression, but these studies have had shortcomings. Either they have focused only on specific subtypes of depression or they have not taken the differences in brain imaging protocols that occur between healthcare institutions into account, the team said.
In an effort to side-step these challenges, Yamashita’s team implemented machine learning to evaluate brain network data from 713 people – 149 individuals in the cohort had major depression. To collect the data, they used rs-fMRI, a technique that detects brain activity and creates images that shine a light on coordinated activity – functional connections – between different parts of the brain. For this study, the imaging was performed at different institutions, using different protocols.
Based on the team’s analysis, the machine learning method was able to pinpoint key functional connections in the imaging data that be considered a brain network signature for major depression. In fact, when the researchers applied the new signature to the rs-fMRI data collected from the various institutions from 521 other people, they were able to reach 70-percent accuracy in identifying which individuals had major depressive disorder.
Ultimately, the researcher said, they hope that this newly identified brain network signature can successfully be applied to different imaging protocols. If so, the door opens for it to be a foundation for unearthing additional brain network patterns that are also associated with depression subtypes. It could also reveal other relationships between depression and other disorders, potentially aiding in matching patients to more effective, personalized treatment, as well as inform the development of additional, new therapies.
For more coverage based on industry expert insights and research, subscribe to the Diagnostic Imaging e-Newsletter here.
Stay at the forefront of radiology with the Diagnostic Imaging newsletter, delivering the latest news, clinical insights, and imaging advancements for today’s radiologists.
The Reading Room Podcast: A Closer Look at Remote MRI Safety, Part 2
July 25th 2025In the second of a multi-part podcast episode, Emanuel Kanal, M.D. and Tobias Gilk, MRSO, MRSE, share their perspectives on remote MRI safety protocols for ensuring screening accuracy and adherence to conditional implant guidelines as well as a rapid and effective response to adverse events.
Study Reveals Significant Prevalence of Abnormal PET/MRI and Dual-Energy CT Findings with Long Covid
July 22nd 2025In a prospective study involving nearly 100 patients with Long Covid, 57 percent of patients had PET/MRI abnormalities and 90 percent of the cohort had abnormalities on dual-energy CT scans.
The Reading Room Podcast: Current and Emerging Insights on Abbreviated Breast MRI, Part 2
July 23rd 2025In the second part of a multi-part podcast episode, Stamatia Destounis, MD, Emily Conant, MD and Habib Rahbar, MD, discuss key sequences for abbreviated breast MRI and how it stacks up to other breast cancer screening modalities.
Stroke MRI Study Assesses Impact of Motion Artifacts Upon AI and Radiologist Lesion Detection
July 16th 2025Noting a 7.4 percent incidence of motion artifacts on brain MRI scans for suspected stroke patients, the authors of a new study found that motion artifacts can reduce radiologist and AI accuracy for detecting hemorrhagic lesions.