Given that depression is one of the most prevalent mental illnesses, developing effective and unobtrusive diagnosis tools is of great importance. Recent work that screens for depression with text messages leverage models relying on lexical category features. Given the colloquial nature of text messages, the performance of these models may be limited by formal lexicons. We thus propose a strategy to automatically construct alternative lexicons that contain more relevant and colloquial terms. Specifically, we generate 36 lexicons from fiction, forum, and news corpuses. These lexicons are then used to extract lexical category features from the text messages. We utilize machine learning models to compare the depression screening capabilities of these lexical category features. Out of our 36 constructed lexicons, 14 achieved statistically significantly higher average F1 scores over the pre-existing formal lexicon and basic bag-of-words approach. In comparison to the pre-existing lexicon, our best performing lexicon increased the average F1 scores by 10%. We thus confirm our hypothesis that less formal lexicons can improve the performance of classification models that screen for depression with text messages. By providing our automatically constructed lexicons, we aid future machine learning research that leverages less formal text.

Depression, a prevalent mental health disorder with severe health and economic consequences, can be costly and difficult to detect. To alleviate this burden, recent research has been exploring the depression screening capabilities of deep learning (DL) models trained on videos of clinical interviews conducted by a virtual agent. Such DL models need to consider the challenges of modality representation, alignment, and fusion as well as small sample sizes. To address them, we propose WavFace, a multimodal deep learning model that inputs audio and temporal facial features. WavFace adds an encoder-transformer layer over pre-trained models to improve the unimodal representation. It also applies an explicit alignment method for both modalities and then uses sequential and spatial self-attention over the alignment. Finally, WavFace fuses the sequential and spatial self-attentions among the two modality embeddings, inspired by how mental health professionals simultaneously observe visual and vocal presentation during clinical interviews. By leveraging sequential and spatial self-attention, WavFace outperforms pre-trained unimodal and multimodal models from the literature. With a single interview question, WaveFace screened for depression with a balanced accuracy of 0.81. This presents a valuable modeling approach for audio-visual mental health screening.

Depression is a common and debilitating mental illness. Given the shortage of mental health professionals, there are delays in depression detection. Interviews conducted by virtual agents could expedite depression screenings. While the interview audio and transcript have received more attention, facial features offer an attractive privacy-preserving screening modality. Thus, we conduct a comprehensive comparative evaluation of the effectiveness of temporal facial features to screen for depression. We extract time series of eye gaze, landmark, and action unit features from video responses to 15 clinical interview questions. We input them into CNN, LSTM, and recurrent convolutional neural network (RCNN) models. An extra attention layer proved critical for CNN and LSTM performance. For a general wellbeing question, eye gaze features screened for depression with an F1 of 0.81. Our study informs the use of temporal facial features in future digital mental illness screening technologies.