Clinical Implication of Maumgyeol Basic Biotypes-Electroencephalography- and Photoplethysmogram-Based Bwave State Inventory

Abstract

Objective: The development of individual subtypes based on biomarkers offers a cost-effective and timely avenue to comprehending individual differences pertaining to mental health, independent from individuals' subjective insights. Incorporating 2-channel electroencephalography (EEG) and photoplethysmogram (PPG), we sought to establish a subtype classification system with clinical relevance.

Methods: One hundred healthy participants and 99 patients with psychiatric disorders were recruited. Classification thresholds were determined using the EEG and PPG data from 2,278 individuals without mental disorders, serving to classify subtypes in our sample of 199 participants. Multivariate analysis of variance was applied to examine psychological distinctions among these subtypes. K-means clustering was employed to verify the classification system.

Results: The distribution of subtypes differed between healthy participants and those with psychiatric disorders. Cognitive abilities were contingent upon brain subtypes, while mind subtypes exhibited significant differences in symptom severity, overall health, and cognitive stress. K-means clustering revealed that the results of our theory-based classification and data-driven classification are comparable. The synergistic assessment of both brain and mind subtypes was also explored.

Conclusion: Our subtype classification system offers a concise means to access individuals' mental health. The utilization of EEG and PPG signals for subtype classification offers potential for the future of digital mental healthcare.

Keywords: Biotype; Digital healthcare; Electroencephalography; Mental health; Photoplethysmogram.

Figure 1.

Proportion of 6 types of EBSI from Maumgyeol brain. EBSI brain subtypes comprised stable, fog, tense, mindful, talented, and fog-tense categories. Among individuals with psychiatric disorders, the fog and tense subtypes were more frequently observed, whereas healthy participants commonly exhibited the stable and talented subtypes. A: All participants (N=199). B: Healthy controls (N=100). C: Patients with psychiatric disorders (N=99). EBSI, electroencephalography- and photoplethysmogram-based Bwave State Inventory.

Figure 2.

Proportion of 6 types of EBSI from Maumgyeol mind. EBSI mind subtypes included stable, weak, weak-depressive, weak-anxious, depressive, and anxious types. Patients with psychiatric disorders exhibited a greater prevalence of weak, weak-depressive, weak-anxious, and anxious subtypes, while the stable and depressive subtypes were more commonly found in healthy participants. A: All participants (N=199). B: Healthy controls (N=100). C: Patients with psychiatric disorders (N=99). EBSI, electroencephalographyand photoplethysmogram-based Bwave State Inventory.

Figure 3.

Results of post-hoc analyses of EBSI subtypes. Error bars represent 1 standard error of the mean. DSST score significantly differed depending on the brain subtypes, while GHQ-12, MHS-D, CGI-S, and CSRS score showed significant difference depending on the mind subtypes. *p<0.05; **p<0.01. DSST, Digit-Symbol Substitution Test; GHQ-12, Korean Version of the General Health Questionnaire; MHS-D, Mental Health Screening Tool for Depressive Disorders; CGI-S, Clinical Global Impression-Severity scale; CSRS, cognitive stress response scale; EBSI, electroencephalography- and photoplethysmogram-based Bwave State Inventory.