Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This study introduces the Neurai-VN benchmark, a standardized framework for evaluating machine learning models in the context of multimodal digital phenotyping for mental health classification. By utilizing a high-resolution dataset collected from 100 Vietnamese adults, the authors define four clinically relevant binary classification tasks and assess various baseline models, achieving mean F1 scores of 0.71 for Healthy Control vs. Depression. These results establish reproducible baselines that can facilitate future research in mental health monitoring through digital phenotyping.
Standardized benchmarks reveal that machine learning models can achieve up to 71% accuracy in distinguishing between healthy individuals and those with depression using multimodal digital data.
Digital phenotyping (DP) using smartphones and wearable devices has shown considerable potential for mental health monitoring. However, progress remains difficult to evaluate due to heterogeneous datasets, inconsistent preprocessing pipelines. In this study, we present a reproducible benchmark built upon the Neurai-VN dataset, a high-resolution, multimodal dataset comprising passive sensing and active assessment from wearable and smartphone devices, collected from 100 Vietnamese adults over two weeks. The benchmark defines four clinically relevant binary classification tasks evaluated using standardized subject-wise cross-validation. Representative linear, tree-based, and neural baseline models are evaluated across predefined feature configurations. Mean subject-level F1 scores across five cross-validation folds reached 0.71 for Healthy Control vs. Depression and Healthy Control vs. Clinical, while Healthy Control vs. Anxiety and Depression vs. Anxiety achieved 0.69 and 0.56, respectively. These benchmark results provide reproducible baselines for future research on multimodal DP for mental health classification tasks.