BackgroundBipolar disorder (BD) is linked to circadian rhythm disruptions resulting in aberrant motor activity patterns. We aimed to explore whether motor activity alone, as assessed by longitudinal actigraphy, can be used to classify accurately BD patients and healthy controls (HCs) into their respective groups.MethodsNinety-day actigraphy records from 25 interepisode BD patients (ie, Montgomery–Asberg Depression Rating Scale (MADRS) and Young Mania Rating Scale (YMRS) < 15) and 25 sex- and age-matched HCs were used in order to identify latent actigraphic biomarkers capable of discriminating between BD patients and HCs. Mean values and time variations of a set of standard actigraphy features were analyzed and further validated using the random forest classifier.ResultsUsing all actigraphy features, this method correctly assigned 88% (sensitivity = 85%, specificity = 91%) of BD patients and HCs to their respective group. The classification success may be confounded by differences in employment between BD patients and HCs. When motor activity features resistant to the employment status were used (the strongest feature being time variation of intradaily variability, Cohen’s d = 1.33), 79% of the subjects (sensitivity = 76%, specificity = 81%) were correctly classified.ConclusionA machine-learning actigraphy-based model was capable of distinguishing between interepisode BD patients and HCs solely on the basis of motor activity. The classification remained valid even when features influenced by employment status were omitted. The findings suggest that temporal variability of actigraphic parameters may provide discriminative power for differentiating between BD patients and HCs while being less affected by employment status.

中文翻译：

---

运动活动模式可以区分发作间双相情感障碍患者和健康对照

背景双相情感障碍 (BD) 与导致异常运动模式的昼夜节律紊乱有关。我们旨在探索通过纵向活动记录仪评估的单独运动活动是否可用于将 BD 患者和健康对照 (HC) 准确分类到各自的组中。方法来自 25 位发作间 BD 患者（即 Montgomery-Asberg）的 90 天活动记录仪记录使用抑郁评定量表 (MADRS) 和年轻躁狂评定量表 (YMRS) < 15) 以及 25 个性别和年龄匹配的 HC 来识别能够区分 BD 患者和 HC 的潜在活动记录生物标志物。使用随机森林分类器分析和进一步验证一组标准活动记录特征的平均值和时间变化。结果使用所有活动记录特征，该方法将 88%（敏感性 = 85%，特异性 = 91%）的 BD 患者和 HC 正确分配到各自的组。BD 患者和 HC 之间的就业差异可能会混淆分类的成功。当使用抵抗就业状态的运动活动特征时（最强的特征是日内变异的时间变化，Cohen'sd= 1.33), 79% 的受试者 (敏感性 = 76%, 特异性 = 81%) 被正确分类。结论基于机器学习活动图的模型能够仅根据运动活动区分发作间 BD 患者和 HC。即使省略了受就业状况影响的特征，该分类仍然有效。研究结果表明，活动记录参数的时间变异性可能为区分 BD 患者和 HC 提供区分能力，同时受就业状况的影响较小。