Digital Mental Health Research Wins Distinguished Paper Award

Co-authors Andrew Campbell, professor of computer science, HealthX Lab graduate students Subigya Nepal and Weichen Wang, and Jeremy Huckins, Department of Psychological and Brain Sciences won a Distinguished Paper Award at the 2023 ACM UbiComp Conference for their paper titled "GLOBEM: Cross-Dataset Generalization of Longitudinal Human Behavior Modeling." Eight out of 210 papers received the Distinguished Paper Award, presented at UbiComp and published in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies (Volume 6).

The GLOBEM paper, inspired and led by Dr. Xuhai Xu (MIT), marked the first attempt to evaluate the cross-dataset generalizability of longitudinal behavior models for predicting depression. The research reported in the paper used the StudentLife dataset and its depression model, which includes phone and wearable data from 83 Dartmouth undergraduates across two 10-week terms during the winter and spring of 2016. StudentLife algorithms predicted the dynamics of week-by-week depression changes in student mental health.

Prior to the GLOBEM paper, most studies built and evaluated machine learning models using data collected from a single population. GLOBEM has the potential to assist researchers in utilizing, developing, and evaluating various open source longitudinal behavior modeling methods presented in this cross-dataset breakthrough paper.