Detection of Minor and Major Depression through Voice as a Biomarker Using Machine Learning

Abstract

Both minor and major depression have high prevalence and are important causes of social burden worldwide; however, there is still no objective indicator to detect minor depression. This study aimed to examine if voice could be used as a biomarker to detect minor and major depression. Ninety-three subjects were classified into three groups: the not depressed group (n = 33), the minor depressive episode group (n = 26), and the major depressive episode group (n = 34), based on current depressive status as a dimension. Twenty-one voice features were extracted from semi-structured interview recordings. A three-group comparison was performed through analysis of variance. Seven voice indicators showed differences between the three groups, even after adjusting for age, BMI, and drugs taken for non-psychiatric disorders. Among the machine learning methods, the best performance was obtained using the multi-layer processing method, and an AUC of 65.9%, sensitivity of 65.6%, and specificity of 66.2% were shown. This study further revealed voice differences in depressive episodes and confirmed that not depressed groups and participants with minor and major depression could be accurately distinguished through machine learning. Although this study is limited by a small sample size, it is the first study on voice change in minor depression and suggests the possibility of detecting minor depression through voice.

Keywords: dimensional approach; machine learning; major depressive episode; minor depressive episode; voice.

Figure 1

Difference of voice features by depressive episode by Benjamini–Hochberg test: (a) Spectral_centroid between three groups; (b) spectral_rolloff between three groups; (c) sq_mean_pitch between three groups; (d) stdev_pitch between three groups; (e) mean_magnitude between three groups; (f) zero-crossing-rate between three groups; (g) voice portion between three groups. * p value < 0.05, ** p value < 0.01, *** p value < 0.001. Abbreviations: ND—not depressed, mDE—minor depressive episode, MDE—major depressive episode, sqrt—square root, sq—squared, stdev—standard deviation.

Figure 1

Difference of voice features by depressive episode by Benjamini–Hochberg test: (a) Spectral_centroid between three groups; (b) spectral_rolloff between three groups; (c) sq_mean_pitch between three groups; (d) stdev_pitch between three groups; (e) mean_magnitude between three groups; (f) zero-crossing-rate between three groups; (g) voice portion between three groups. * p value < 0.05, ** p value < 0.01, *** p value < 0.001. Abbreviations: ND—not depressed, mDE—minor depressive episode, MDE—major depressive episode, sqrt—square root, sq—squared, stdev—standard deviation.

Figure 2

AUC curve predicting minor and major episodes using MLP: (a) AUC for minor episode, 7:3 training; (b) AUC for major episode, 7:3 training; (c) AUC for minor episode, 8:2 training; (d) AUC for major episode, 8:2 training; We only have the averaged result (for all the episodes) in Table 4, while this figure incorporates the result for each major and minor episode. Abbreviations: MLP—multi-layer perceptron, AUC—area under curve.