Resting-State Electroencephalographic Signatures Predict Treatment Efficacy of Tacs for Refractory Auditory Hallucinations in Schizophrenic Patients

Xiaojuan Wang,Ruxin Hu,Tao Wang,Yuan Chang,Xiaoya Liu,Meijuan Li,Ying Gao,Shuang Liu,Dong Ming
DOI: https://doi.org/10.1109/jbhi.2024.3509438
IF: 7.7
2024-01-01
IEEE Journal of Biomedical and Health Informatics
Abstract:Transcranial alternating current stimulation (tACS) has been reported to treat refractory auditory hallucinations in schizophrenia. Despite diligent efforts, it is imperative to underscore that tACS does not uniformly demonstrate efficacy across all patients as with all treatments currently employed in clinical practice. The study aims to find biomarkers predicting individual responses to tACS, guiding treatment decisions, and preventing healthcare resource wastage. We divided 17 schizophrenic patients with refractory auditory hallucinations into responsive(RE) and non-responsive(NR) groups based on their auditory hallucination symptom reduction rates after one month of tACS treatment. The pre-treatment resting-state electroencephalogram(rsEEG) was recorded and then computed absolute power spectral density (PSD), Hjorth parameters (HPs, Hjorth activity (HA), Hjorth mobility (HM), and Hjorth complexity (HC) included) from different frequency bands to portray the brain oscillations. The results demonstrated that statistically significant differences localized within the high gamma frequency bands of the right brain hemisphere. Immediately, we input the significant dissociable features into popular machine learning algorithms, the Cascade Forward Neural Network achieved the best recognition accuracy of 93.87%. These findings preliminarily imply that high gamma oscillations in the right brain hemisphere may be the main influencing factor leading to different responses to tACS treatment, and incorporating rsEEG signatures could improve personalized decisions for integrating tACS in clinical treatment.
What problem does this paper attempt to address?