Algorithm improves electroencephalography’s sensitivity to depression-related brain activity

The inexpensive and readily accessible electroencephalography (EEG) method allows doctors to characterize the brain’s electrical activity, enabling them to diagnose sleep disorders and seizures. However, EEG has limited spatial resolution, which makes it less effective for characterizing conditions like depression, which require understanding the brain’s physical neural network. Ding et al. developed an algorithm that improves EEG’s sensitivity to depression-related brain behavior.

The team’s method, the Inner Composition Alignment algorithm, results in higher quantitative differences in brain activity — corresponding to improved sensitivity — than traditional algorithms in several key measures. For example, in one index that measures the brain’s ability to dynamically reconfigure itself to meet task demands, the authors found that depression patients deviated from a critical threshold significantly more than a control group — 82.6% compared to 24.3%. Finally, the authors found that certain brainwave frequencies — grouped in ranges known as alpha, beta gamma, and theta bands — were different in depression patients compared to the control group.

“In theta/alpha bands, the integration state duration is significantly reduced, and in beta/gamma bands, the segregation state is enhanced,” said author Jiahao Ding. “Meanwhile, the state transition frequency shows abnormal elevation in the alpha band but significant reduction in the beta and gamma bands.”

The EEG source data came from a database hosted by the University of Arizona. Participants were divided into a control group and high-depressive-symptom group based on their Beck Depression Index scores, a widely used psychometric test for evaluating depression.

Future work includes validating the study through bigger sample sizes of over 200 patients and exploring the evolution of brain network properties over time.

Source: “Nested spectral analysis of brain functional networks in depression based on inner composition alignment,” by Jiahao Ding, Xiaoyu Xie, Bincheng Fu, Jianchun Hua, and Jun Wang, AIP Advances (2025). The article can be accessed at https://doi.org/10.1063/5.0266528 .