We are excited to announce the MAISON-LLF Data Challenge as part of the ARIAL Workshop. The MAISON-LLF dataset (Version 4) is publicly available for download (https://doi.org/10.5281/zenodo.17943110) and was collected from 18 older adults living alone in the community following a lower limb fracture.
The data were collected using MAISON, a Multimodal AI-based Sensor platform for Older Individuals that included a smartwatch, smartphone, sleep mat, and indoor motion sensor to capture movement, activity, sleep, and physiological data. Each participant contributed data for 8 weeks, beginning in the first week after discharge. This resulted in a total of 1,008 days of continuous multimodal sensor data, complemented by biweekly clinical questionnaire data. The de-identified dataset contains 84 features/columns, including:
4 identifier and timestamp variables, such as participant, timestamp, and clinical-timestamp
46 numeric features
35 clinical scores
Details of the study protocol and dataset features are provided in the accompanying dataset paper.
This novel dataset supports a wide range of research directions, including classification, regression, clustering, temporal prediction, explainability, visualization, large language model applications, and hypothesis generation.
We invite researchers to submit novel papers to the workshop based on the MAISON-LLF dataset, excluding work already reported in the original dataset paper. We particularly encourage submissions that advance methodological development, data processing pipelines, and clinically relevant applications.
Submitted papers, whether short or long, will be peer reviewed as regular workshop submissions and included in the workshop proceedings. Since the dataset is already publicly available, authors are also encouraged to release their code to support reproducibility and further research in this area.
All inquiries related to MAISON-LLF Data Challenge submissions may be directed to Dr. Ali Abedi.