Depression Identification from Gait Spectrum Features Based on Hilbert-Huang Transform

Depression is a common mental illness, which is extremely harmful to individuals and society. Timely and effective diagnosis is very important for patients’ treatments and depression preventions. In this paper, we treat the trajectory of gait as signal, proposing a new direction to detect the depression with gait frequency features based on Hilbert-Huang transform (HHT). Two groups of participants are recruited in this experiment, including 47 healthy people and 54 depressed patients, respectively. We process the gait data with HHT and build the classification models which verification method is leave-one-out. The best result of our work is 91.09% when the model I is adopted and the classifier is SVM. The corresponding specificity and sensitivity are 87.23% and 94.44% respectively. It verifies that the gait frequency of patients with depression is significantly different from that of healthy people, and the frequency domain features of gait are helpful for the diagnosis of depression.

[1]  Ching Y. Suen,et al.  An EMD-based recognition method for Chinese fonts and styles , 2006, Pattern Recognit. Lett..

[2]  Sam Kwong,et al.  Hilbert-Huang Transform for Analysis of Heart Rate Variability in Cardiac Health , 2011, IEEE/ACM Transactions on Computational Biology and Bioinformatics.

[3]  Nicholas B. Allen,et al.  Multichannel Weighted Speech Classification System for Prediction of Major Depression in Adolescents , 2013, IEEE Transactions on Biomedical Engineering.

[4]  L Sloman,et al.  Gait patterns of depressed patients and normal subjects. , 1982, The American journal of psychiatry.

[5]  V. Rai,et al.  Bearing fault diagnosis using FFT of intrinsic mode functions in Hilbert-Huang transform , 2007 .

[6]  Mohamed Chetouani,et al.  Hilbert-Huang transform based physiological signals analysis for emotion recognition , 2009, 2009 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT).

[7]  Suzanne Kieffer,et al.  Feature extraction and selection for objective gait analysis and fall risk assessment by accelerometry , 2011, Biomedical engineering online.

[8]  Fan Yang,et al.  Designing a framework for assisting depression severity assessment from facial image analysis , 2015, 2015 IEEE International Conference on Signal and Image Processing Applications (ICSIPA).

[9]  Bradley Matthew Battista,et al.  Application of the Empirical Mode Decomposition and Hilbert-Huang Transform to Seismic Reflection Data , 2007 .

[10]  M. Lemke,et al.  Spatiotemporal gait patterns during over ground locomotion in major depression compared with healthy controls. , 2000, Journal of psychiatric research.

[11]  Mohammad H. Mahoor,et al.  Social risk and depression: Evidence from manual and automatic facial expression analysis , 2013, 2013 10th IEEE International Conference and Workshops on Automatic Face and Gesture Recognition (FG).

[12]  Mitchell D. Wilkes,et al.  Evaluation of Voice Acoustics as Predictors of Clinical Depression Scores. , 2017, Journal of voice : official journal of the Voice Foundation.

[13]  Rami J Oweis,et al.  Seizure classification in EEG signals utilizing Hilbert-Huang transform , 2011, Biomedical engineering online.

[14]  Guodong Guo,et al.  Visually Interpretable Representation Learning for Depression Recognition from Facial Images , 2020, IEEE Transactions on Affective Computing.

[15]  Jeffrey M. Hausdorff,et al.  Gait unsteadiness and fall risk in two affective disorders: a preliminary study , 2004, BMC psychiatry.

[16]  Fan Yang,et al.  Facial geometry and speech analysis for depression detection , 2017, 2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[17]  J. Darby,et al.  Speech and voice parameters of depression: a pilot study. , 1984, Journal of communication disorders.

[18]  N. Huang,et al.  The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis , 1998, Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences.