Bayesian Networks for the robust and unbiased prediction of depression and its symptoms utilizing speech and multimodal data

TY - CHAP

T1 - Bayesian Networks for the robust and unbiased prediction of depression and its symptoms utilizing speech and multimodal data

AU - Fara, Salvatore

AU - Hickey, Orlaith

AU - Georgescu, Alexandra

AU - Goria, Stefano

AU - Molimpakis, Emilia

AU - Cummins, Nicholas

N1 - Publisher Copyright: © 2023 International Speech Communication Association. All rights reserved.

PY - 2023/8/20

Y1 - 2023/8/20

N2 - Predicting the presence of major depressive disorder (MDD) using speech is highly non-trivial. The heterogeneous clinical profile of MDD means that any given speech pattern may be associated with a unique combination of depressive symptoms. Conventional discriminative machine learning models may lack the complexity to robustly model this heterogeneity. Bayesian networks, however, are well-suited to such a scenario. They provide further advantages over standard discriminative modeling by offering the possibility to (i) fuse with other data streams; (ii) incorporate expert opinion into the models; (iii) generate explainable model predictions, inform about the uncertainty of predictions, and (iv) handle missing data. In this study, we apply a Bayesian framework to capture the relationships between depression, depression symptoms, and features derived from speech, facial expression and cognitive game data. Presented results also highlight our model is not subject to demographic biases.

AB - Predicting the presence of major depressive disorder (MDD) using speech is highly non-trivial. The heterogeneous clinical profile of MDD means that any given speech pattern may be associated with a unique combination of depressive symptoms. Conventional discriminative machine learning models may lack the complexity to robustly model this heterogeneity. Bayesian networks, however, are well-suited to such a scenario. They provide further advantages over standard discriminative modeling by offering the possibility to (i) fuse with other data streams; (ii) incorporate expert opinion into the models; (iii) generate explainable model predictions, inform about the uncertainty of predictions, and (iv) handle missing data. In this study, we apply a Bayesian framework to capture the relationships between depression, depression symptoms, and features derived from speech, facial expression and cognitive game data. Presented results also highlight our model is not subject to demographic biases.

UR - http://www.scopus.com/inward/record.url?scp=85171578170&partnerID=8YFLogxK

U2 - 10.21437/Interspeech.2023-1709

DO - 10.21437/Interspeech.2023-1709

M3 - Conference paper

VL - 2023-August

T3 - Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH

SP - 1728

EP - 1732

BT - Proc. INTERSPEECH 2023

PB - ISCA-INST SPEECH COMMUNICATION ASSOC

ER -