RESUMO
A significant level of stigma and inequality exists in mental healthcare, especially in under-served populations. Inequalities are reflected in the data collected for scientific purposes. When not properly accounted for, machine learning (ML) models learned from data can reinforce these structural inequalities or biases. Here, we present a systematic study of bias in ML models designed to predict depression in four different case studies covering different countries and populations. We find that standard ML approaches regularly present biased behaviors. We also show that mitigation techniques, both standard and our own post-hoc method, can be effective in reducing the level of unfair bias. There is no one best ML model for depression prediction that provides equality of outcomes. This emphasizes the importance of analyzing fairness during model selection and transparent reporting about the impact of debiasing interventions. Finally, we also identify positive habits and open challenges that practitioners could follow to enhance fairness in their models.
Assuntos
Depressão , Hábitos , Humanos , Depressão/diagnóstico , Viés , Instalações de Saúde , Aprendizado de MáquinaRESUMO
The current epidemics of cardiovascular and metabolic noncommunicable diseases have emerged alongside dramatic modifications in lifestyle and living environments. These correspond to changes in our "modern" postwar societies globally characterized by rural-to-urban migration, modernization of agricultural practices, and transportation, climate change, and aging. Evidence suggests that these changes are related to each other, although the social and biological mechanisms as well as their interactions have yet to be uncovered. LongITools, as one of the 9 projects included in the European Human Exposome Network, will tackle this environmental health equation linking multidimensional environmental exposures to the occurrence of cardiovascular and metabolic noncommunicable diseases.
