Analysis of wearable time series data in endocrine and metabolic research.
Analysis of wearable time series data in endocrine and metabolic research.
复制标题
DOI：
10.1016/j.coemr.2022.100380
复制
发表时间：
2022-08
影响因子：
--
通讯作者：
中科院分区：
文献类型：
--
作者：
体内的许多激素以不同的频率和振幅振荡，创造了一个对维持健康至关重要的动态环境。在人类中，这些节律的破坏与发病率和死亡率的增加密切相关。虽然数学模型可以帮助我们理解节律失调，但将这种见解转化为个性化医疗技术需要解决额外的挑战。在这里，我们讨论如何结合微创，高频生物采样技术与可穿戴设备可以帮助激素替代品的发展。我们回顾定制的算法，可以帮助分析多维，嘈杂，时间序列数据，并确定可穿戴信号，可能构成内分泌节律的临床代理。这些技术可以支持计算生物标志物的开发，以支持内分泌和代谢疾病的诊断和管理。疾病期间内分泌和代谢节律经常失调。量化内分泌节律变异性是个性化健康干预的关键。多模式，高频率采样可以帮助解开内分泌系统的功能。多维可穿戴数据集可以构成内分泌节律的替代物。数字表型现在可以通过节奏分析算法广泛使用。
Many hormones in the body oscillate with different frequencies and amplitudes, creating a dynamic environment that is essential to maintain health. In humans, disruptions to these rhythms are strongly associated with increased morbidity and mortality. While mathematical models can help us understand rhythm misalignment, translating this insight into personalised healthcare technologies requires solving additional challenges. Here, we discuss how combining minimally invasive, high-frequency biosampling technologies with wearable devices can assist the development of hormonal surrogates. We review bespoke algorithms that can help analyse multidimensional, noisy, time series data and identify wearable signals that could constitute clinical proxies of endocrine rhythms. These techniques can support the development of computational biomarkers to support the diagnosis and management of endocrine and metabolic conditions. Endocrine and metabolic rhythms often become misaligned during disease. Quantifying endocrine rhythm variability is key to personalise health interventions. Multimodal, high-frequency sampling can help disentangle endocrine systems function. Multi-dimensional wearable datasets can constitute surrogates of endocrine rhythms. Digital phenotyping is now widely accessible through rhythmic analysis algorithms.