Digital health applications using Artificial Intelligence (AI) are a promising opportunity to address the widening gap between available resources and mental health needs globally. Increasingly, passively acquired data from wearables are augmented with carefully selected active data from depressed individuals to develop Machine Learning (ML) models of depression based on mood scores. However, most ML models are black box in nature, and hence the outputs are not explainable. Depression is also multimodal, and the reasons for depression may vary significantly between individuals. Explainable and personalised models will thus be beneficial to clinicians to determine the main features that lead to a decline in the mood state of a depressed individual, thus enabling suitable personalised therapy. This is currently lacking. Therefore, this study presents a methodology for developing personalised and accurate Deep Learning (DL)-based predictive mood models for depression, along with novel methods for identifying the key facets that lead to the exacerbation of depressive symptoms. We illustrate our approach by using an existing multimodal dataset containing longitudinal Ecological Momentary Assessments of depression, lifestyle data from wearables and neurocognitive assessments for 14 mild to moderately depressed participants over one month. We develop classification- and regression-based DL models to predict participants’ current mood scores—a discrete score given to a participant based on the severity of their depressive symptoms. The models are trained inside eight different evolutionary-algorithm-based optimisation schemes that optimise the model parameters for a maximum predictive performance. A five-fold cross-validation scheme is used to verify the DL model’s predictive performance against 10 classical ML-based models, with a model error as low as 6% for some participants. We use the best model from the optimisation process to extract indicators, using SHAP, ALE and Anchors from explainable AI literature to explain why certain predictions are made and how they affect mood. These feature insights can assist health professionals in incorporating personalised interventions into a depressed individual’s treatment regimen.

Depression and anxiety are the leading causes of health loss globally, and the Covid-19 pandemic has significantly exacerbated the effect of these disorders. There is a widening gap between available resources and mental health needs globally. Digital health applications using artificial Intelligence (AI) are a promising opportunity to address this widening gap. Increasingly, passively acquired data from wearables is augmented with carefully selected active data from the participants to develop machine learning (ML) models of depression. However, these ML models are black-box in nature, and hence the outputs are not explainable. Depression is also multi-modal, and the reasons for depression may vary significantly between individuals. Explainable and personalised models will thus be beneficial to clinicians for determining the main features that lead to a decline in the mood state of a patient, thus enabling suitable personalised therapy. This is currently lacking. Therefore, this study presents the first methodology for developing personalised and accurate deep learning (DL)-based models for depression, along with novel methods for identifying the key facets that lead to the exacerbation of depressive symptoms. We illustrate our approach using an existing multi-modal dataset containing longitudinal ecological momentary assessments of depression, lifestyle data from wearables, and neurocognitive assessments for 14 mild to moderately depressed participants over one month. We train classification- and regression-based DL models to predict participants’ mood scores - a discrete score given to a participant based on the severity of their depressive symptoms. The models are trained inside eight different evolutionaryalgorithm-based optimisation schemes that optimise the model parameters for maximum predictive performance. A 5-fold cross-validation scheme is used to verify the model performance, with model error as low as 6% for some participants. We use the best model from the optimisation process to extract indicators, using SHAP, ALE and Anchors from Explainable AI literature to explain why certain predictions are made and how they affect mood. These feature insights can assist health professionals in incorporating personalised interventions into the patient’s treatment regimen.