Institute of Life Sciences, Sant'Anna School of Advanced Studies, Pisa, Italy; Department of Clinical Neurosciences, Geneva University Hospital (HUG), Geneva, Switzerland.
Department of Clinical and Experimental Medicine, University of Pisa, Pisa, Italy.
J Affect Disord. 2021 Dec 1;295:323-338. doi: 10.1016/j.jad.2021.08.052. Epub 2021 Aug 27.
Bias-prone psychiatric interviews remain the mainstay of bipolar disorder (BD) assessment. The development of digital phenotyping promises to improve BD management. We present a systematic review of the evidence about the use of portable digital devices for the identification of BD, BD types and BD mood states and for symptom assessment.
We searched Web of Knowledge, Scopus ®, IEEE Xplore, and ACM Digital Library databases (until 5/1/2021) for articles evaluating the use of portable/wearable digital devices, such as smartphone apps, wearable sensors, audio and/or visual recordings, and multimodal tools. The protocol is registered in PROSPERO (CRD42020200086).
We included 62 studies (2325 BD; 724 healthy controls, HC): 27 using smartphone apps, either for recording self-assessments (n = 10) or for passively gathering metadata (n = 7) or both (n = 10); 15 using wearable sensors for physiological parameters; 17 analysing audio and/or video recordings; 3 using multiple technologies. Two thirds of the included studies applied artificial intelligence (AI)-based approaches. They achieved fair to excellent classification performances.
The included studies had small sample sizes and marked heterogeneity. Evidence of overfitting emerged, limiting generalizability. The absence of clear guidelines about reporting classification performances, with no shared standard metrics, makes results hardly interpretable and comparable.
New technologies offer a noteworthy opportunity to BD digital phenotyping with objectivity and high granularity. AI-based models could deliver important support in clinical decision-making. Further research and cooperation between different stakeholders are needed for addressing methodological, ethical and socio-economic considerations.
有偏见倾向的精神病学访谈仍然是双相情感障碍 (BD) 评估的主要方法。数字表型学的发展有望改善 BD 的管理。我们对使用便携式数字设备来识别 BD、BD 类型和 BD 情绪状态以及进行症状评估的证据进行了系统回顾。
我们在 Web of Knowledge、Scopus®、IEEE Xplore 和 ACM Digital Library 数据库中（截至 2021 年 5 月 1 日）搜索了评估使用便携式/可穿戴数字设备（如智能手机应用程序、可穿戴传感器、音频和/或视频记录以及多模态工具）的文章。该方案已在 PROSPERO（CRD42020200086）中注册。
我们纳入了 62 项研究（2325 例 BD；724 例健康对照，HC）：27 项研究使用智能手机应用程序，用于记录自我评估（n=10）或被动收集元数据（n=7）或两者兼而有之（n=10）；15 项研究使用可穿戴传感器测量生理参数；17 项研究分析音频和/或视频记录；3 项研究使用多种技术。纳入研究中有三分之二应用了基于人工智能（AI）的方法。它们实现了公平到优秀的分类性能。
纳入的研究样本量较小且存在明显的异质性。过度拟合的证据表明，可推广性有限。关于报告分类性能缺乏明确的指南，没有共享的标准指标，使得结果难以解释和比较。
新技术为 BD 数字表型学提供了具有客观性和高粒度的重要机会。基于 AI 的模型可以在临床决策中提供重要支持。需要不同利益相关者之间进行进一步的研究和合作，以解决方法学、伦理和社会经济方面的考虑。
