Dataset for Systematic Review: Computational Enhancement of Behavioral Activation in Depression: A Systematic Review of AI Methods and Neurobiological Markers

This dataset supports the systematic review titled "Computational Enhancement of Behavioral Activation in Depression: A Systematic Review of AI Methods and Neurobiological Markers" (Tallón Fuentes et al., 2026). It contains metadata and supporting materials for the 65 studies included after screening (updated search to February 1, 2026), extracted from PubMed, Scopus, ACM Digital Library, and Web of Science. The review synthesizes evidence on how artificial intelligence (AI) and computational methods can support Behavioral Activation (BA) interventions for depression, integrating neurobiological mechanisms (e.g., prefrontal/hippocampal alterations, neuroinflammation, BDNF, HPA axis), frailty as a moderator, digital phenotyping (wearables, activity/passivity indices), and multimodal detection/monitoring approaches. Files included in this dataset:- dataset.csv: Main metadata table for the 65 included studies, with columns for Authors, Year, Title, Journal, Volume/Pages, DOI, Full Citation, Key Findings, Risk of Bias assessment, and Research Question association.- S1_File.docx: Retrospective review protocol (including refined scope, research questions, methodology, limitations, and ethical considerations).- Online_Resource_1.pdf: Full search terms, Boolean strings, variants, and detailed search strategy per database.- S2_File.docx: Completed PRISMA 2020 checklist for systematic reviews and meta-analyses (updated to reflect the final flow and GRADE certainty ratings).. The data were curated following PRISMA 2020 guidelines, with dual independent screening, risk of bias assessment (RoB 2, NOS, PROBAST, AMSTAR-2), and narrative synthesis due to high heterogeneity. GRADE certainty ratings (Low to Moderate overall) are included in the supporting materials. This dataset enables full reproducibility of the review's findings on AI-supported BA, multimodal neuroimaging, digital phenotyping, and frailty moderation in depression care. License: CC BY 4.0 (Creative Commons Attribution 4.0 International). Users are free to share, copy, redistribute, adapt, remix, and build upon the material for any purpose, even commercially, provided appropriate credit is given to the authors (Tallón Fuentes et al.), a link to the license is provided, and any changes are indicated.