Abstract
Depression is a common psychiatric disease. At present, psychometric scales are the main methods for detecting depression in patients and evaluating the clinical treatment effect of depression. However, the accuracy of the scales is influenced by the subjective factors of patients and doctors. This paper explored the construction of a depression detection model based on task-state heart rate variability (HRV) parameters and the effect of therapy on the related HRV parameters. The candidate HRV parameters were first extracted from the task-state electrocardiogram (ECG) collected before treatment and at three observation points during treatment. Then, a statistical t-test was used to screen those characteristic HRV parameters with a significant difference between the depressed and normal groups before treatment. The characteristic HRV parameters and a support-vector machine (SVM) were combined to construct the detection model. Finally, a score model was designed to reveal dynamic changes in the HRV parameters during the treatment process. This paper constructed an automatic, simple, and efficient depression detection model: peakHF+SVM. Detection accuracy reached 89.66%, and this model had comprehensive advantages compared with other related methods. During the entire treatment process, the change in the scores and the time to achieve the maximal scores were different among patients. The type and number of HRV parameters related to the maximal score of each patient also were different. The depression detection model has good application prospects in the objective, quantitative, and automatic detection of depression. The same curative method produced different effects on each patient with depression. The proposed score model may be helpful for the quantitative assessment of the therapeutic effect.
Highlights
Depression is a common psychiatric disease characterized by high morbidity, recurrent attacks and a high mortality and disability rate
We quantitatively evaluated the difference between the task-state heart rate variability (HRV) of a depressed patient and a healthy participant on two levels: difference in absolute values and correlation difference
Previous studies on the automatic detection of depression were mainly based on EEG, fNIRS and fMRI imaging features and rest-state HRV parameters
Summary
Depression is a common psychiatric disease characterized by high morbidity, recurrent attacks and a high mortality and disability rate. It seriously affects the work and life of patients and places a great burden on society. Psychometric scales and behavioral observations are used to identify patients with depression. Statistical studies have confirmed that the accuracy rate of clinicians diagnosing depression by psychometric scales is only 47.3% [1]. Some researchers have extracted the characteristics from EEG signals [2], nearinfrared spectral signals [3] and magnetic resonance images [4] to automatically detect and identify depression.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
