Mariani Sara, Migliorini Matteo, Tacchino Giulia, Gentili Claudio, Bertschy Gilles, Werner Sandra, Bianchi Anna M
Politecnico di Milano, Dept. of Biomedical Engineering, Milan, Italy.
Annu Int Conf IEEE Eng Med Biol Soc. 2012;2012:2240-3. doi: 10.1109/EMBC.2012.6346408.
The aim of this study is to identify parameters extracted from the Heart Rate Variability (HRV) signal that correlate to the clinical state in patients affected by bipolar disorder. 25 ECG and activity recordings from 12 patients were obtained by means of a sensorized T-shirt and the clinical state of the subjects was assessed by a psychiatrist. Features in the time and frequency domain were extracted from each signal. HRV features were also used to automatically compute the sleep profile of each subject by means of an Artificial Neural Network, trained on a control group of healthy subjects. From the hypnograms, sleep-specific parameters were computed. All the parameters were compared with those computed on the control group, in order to highlight significant differences in their values during different stages of the pathology. The analysis was performed by grouping the subjects first on the basis of the depression-mania level and then on the basis of the anxiety level.
本研究的目的是识别从心率变异性（HRV）信号中提取的与双相情感障碍患者临床状态相关的参数。通过一件装有传感器的T恤衫获取了12名患者的25份心电图和活动记录，并由一名精神科医生评估了受试者的临床状态。从每个信号中提取了时域和频域特征。HRV特征还通过在一组健康受试者对照组上训练的人工神经网络，用于自动计算每个受试者的睡眠状况。从睡眠图中计算出特定于睡眠的参数。将所有参数与对照组计算出的参数进行比较，以突出在疾病不同阶段其值的显著差异。分析首先根据抑郁-躁狂水平对受试者进行分组，然后根据焦虑水平进行分组。
