Neurai-VN, A Multimodal Digital Phenotyping Dataset for Depression and Anxiety Assessment in Free-Living Conditions

ABSTRACT: Despite affecting hundreds of millions of people globally, depression and anxiety remain understudied through passive sensing in low- and middle-income countries (LMICs), where severe clinical workforce shortages heighten the need for scalable monitoring. To address this gap, this study presents Neurai-VN, a high-resolution, multimodal dataset comprising passive sensing from wearable and smartphone devices, collected from 100 Vietnamese adults (aged 18–50) over two weeks. Participants were clinically screened and categorized into four mutually exclusive groups: depression, anxiety (generalized or social anxiety disorder), healthy controls, and other psychiatric conditions. The dataset contains (1) continuous wearable physiological signals and smartphone-derived behavioral data collected in real-world settings; (2) clinical labels, including DSM-5 diagnoses, symptom severity ratings, and validated self-report measures (PHQ-9, GAD-7, and daily mood assessments); and (3) standardized day-level features comprising 1,730 participant records across 14 sensing modalities, together with 2,096 validated self-report entries. Beyond its primary use, the dataset is anticipated to facilitate the discovery of mental health biomarkers and enable earlier detection of depression and anxiety, particularly in low-resource settings. Latest Update: May, 15, 2026 Version  Released Date Description Version 1.0.0 May, 15, 2026 Added P0XXX zip file, indicates individual data file. Change metadata This dataset is currently under preparation for submission to Nature Scientific Data.The dataset is publicly released via Zenodo to facilitate open scientific use and reproducibility. Users of these data are requested to cite the Zenodo record (DOI) in any resulting publications, presentations, software, or derivative works.A peer-reviewed publication associated with this dataset will be linked upon publication. For questions concerning the dataset, please contact: 24cuong.pq@vinuni.edu.vn