OBJECTIVE This study aimed to identify an Electroencephalography (EEG) complexity biomarker that could predict treatment resistance in Obsessive compulsive disorder (OCD) patients. Additionally, the statistical differences between EEG complexity values in treatment-resistant and treatment-responsive patients were determined. Moreover, the existence of correlations between EEG complexity and Yale-Brown Obsessive Compulsive Scale (YBOCS) score were evaluated. METHODS EEG data for 29 treatment-resistant and 28 treatment-responsive OCD patients were retrospectively evaluated. Approximate entropy (ApEn) method was used to extract the EEG complexity from both whole EEG data and filtered EEG data, according to 4 common frequency bands, namely delta, theta, alpha, and beta. The random forests method was used to classify ApEn complexity. RESULTS ApEn complexity extracted from beta band EEG segments discriminated treatment-responsive and treatment-resistant OCD patients with an accuracy of 89.66% (sensitivity: 89.44%; specificity: 90.64%). Beta band EEG complexity was lower in the treatment-resistant patients and the severity of OCD, as measured by YBOCS score, was inversely correlated with complexity values. CONCLUSIONS The results indicate that, EEG complexity could be considered a biomarker for predicting treatment response in OCD patients. SIGNIFICANCE The prediction of treatment response in OCD patients might help clinicians devise and administer individualized treatment plans.

中文翻译：

---

基于脑电图复杂性作为生物标志物的强迫症患者治疗抵抗预测

目的 本研究旨在确定一种脑电图 (EEG) 复杂性生物标志物，该生物标志物可以预测强迫症 (OCD) 患者的治疗抵抗。此外，还确定了治疗抵抗和治疗反应患者的 EEG 复杂性值之间的统计差异。此外，还评估了 EEG 复杂性与耶鲁-布朗强迫症量表 (YBOCS) 评分之间是否存在相关性。方法回顾性评估了 29 名治疗抵抗和 28 名治疗有效的强迫症患者的脑电图数据。根据 delta、theta、alpha 和 beta 4 个常见频段，使用近似熵 (ApEn) 方法从整个 EEG 数据和滤波后的 EEG 数据中提取 EEG 复杂性。随机森林方法用于对 ApEn 复杂性进行分类。结果 从 β 波段脑电图片段中提取的 ApEn 复杂性以 89.66% 的准确度（敏感性：89.44%；特异性：90.64%）区分治疗反应性和治疗抵抗性强迫症患者。Beta 波段 EEG 复杂性在治疗抵抗的患者中较低，并且根据 YBOCS 评分衡量的 OCD 严重程度与复杂性值呈负相关。结论 结果表明，脑电图复杂性可被视为预测强迫症患者治疗反应的生物标志物。意义 预测强迫症患者的治疗反应可能有助于临床医生设计和实施个体化治疗计划。Beta 波段 EEG 复杂性在治疗抵抗的患者中较低，并且根据 YBOCS 评分衡量的 OCD 严重程度与复杂性值呈负相关。结论 结果表明，脑电图复杂性可被视为预测强迫症患者治疗反应的生物标志物。意义 预测强迫症患者的治疗反应可能有助于临床医生设计和实施个体化治疗计划。Beta 波段 EEG 复杂性在治疗抵抗的患者中较低，并且根据 YBOCS 评分衡量的 OCD 严重程度与复杂性值呈负相关。结论 结果表明，脑电图复杂性可被视为预测强迫症患者治疗反应的生物标志物。意义 预测强迫症患者的治疗反应可能有助于临床医生设计和实施个体化治疗计划。