CAS Key Laboratory of Behavioral Science, Institute of Psychology, Beijing 100101, China; Department of Psychology, University of Chinese Academy of Sciences, Beijing 100049, China.
School of Optometry, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong.
J Affect Disord. 2021 Jun 1;288:161-166. doi: 10.1016/j.jad.2021.04.004. Epub 2021 Apr 20.
Machine-learning methods using acoustic features in the diagnosis of major depressive disorder (MDD) have insufficient evidence from large-scale samples and clinical trials. This study aimed to evaluate the effectiveness of the promising i-vector method on a large sample of women with recurrent MDD diagnosed clinically, examine its robustness, and provide an explicit acoustic explanation of the i-vectors.
We collected utterances edited from clinical interview speech records of 785 depressed and 1,023 healthy individuals. Then, we extracted Mel-frequency cepstral coefficient (MFCC) features and MFCC i-vectors from their utterances. To examine the effectiveness of i-vectors, we compared the performance of binary logistic regression between MFCC i-vectors and MFCC features and tested its robustness on different utterance durations. We also determined the correlation between MFCC features and MFCC i-vectors to analyze the acoustic meaning of i-vectors.
The i-vectors improved 7% and 14% of area under the curve (AUC) for MFCC features using different utterances. When the duration is > 40 s, the classification results are stabilized. The i-vectors are consistently correlated to the maximum, minimum, and deviations of MFCC features (either positively or negatively).
This study included only women.
The i-vectors can improve 14% of the AUC on a large-scale clinical sample. This system is robust to utterance duration > 40 s. This study provides a foundation for exploring the clinical application of voice features in the diagnosis of MDD.
使用声学特征诊断重度抑郁症（MDD）的机器学习方法缺乏来自大样本和临床试验的充分证据。本研究旨在评估 i-vector 方法在经过临床诊断的复发性 MDD 女性大样本中的有效性，检验其稳健性，并提供 i-vector 的明确声学解释。
我们收集了 785 名抑郁患者和 1023 名健康个体的临床访谈语音记录中编辑过的话语。然后，我们从他们的话语中提取梅尔频率倒谱系数（MFCC）特征和 MFCC i-vectors。为了检验 i-vectors 的有效性，我们比较了 MFCC i-vectors 和 MFCC 特征的二元逻辑回归性能，并测试了在不同话语时长下的稳健性。我们还确定了 MFCC 特征和 MFCC i-vectors 之间的相关性，以分析 i-vectors 的声学意义。
对于不同时长的话语，i-vectors 分别提高了 MFCC 特征 AUC 的 7%和 14%。当时长>40s 时，分类结果趋于稳定。i-vectors与 MFCC 特征的最大值、最小值和偏差（无论是正相关还是负相关）始终相关。
本研究仅包括女性。
i-vectors 可以提高大型临床样本中 14%的 AUC。该系统对时长>40s 的话语具有稳健性。本研究为探索语音特征在 MDD 诊断中的临床应用奠定了基础。
