Key result
Personalized multi-task learning model detects stress with ~73% accuracy, outperforming a general model.
Why the study?
Does a personalized multi-task learning neural network model using wearable sensors improve real-time stress detection compared to general or random models in people with and without intellectual disabilities?
Observational (n=92)
Does a personalized multi-task learning neural network model using wearable sensors improve real-time stress detection compared to general or random models in people with and without intellectual disabilities?
A personalized machine learning model using wearable physiological sensors can accurately detect stress in real-time for people with and without intellectual disabilities.
Should not yet change practice for stress monitoring; leaves open the value of personalized multi-task learning models.
< .001. The findings suggest that the personalized model can detect stress in people with and without ID. A larger-scale study is required to validate the system for people with ID.
No takes yet. Share an insight, caveat, or question.
Vries et al. (2023) conducted an observational in Intellectual disability and healthy controls (n=92). Personalized multi-task learning neural network model vs. General model and random chance model was evaluated on Balanced accuracy of stress detection (p=<0.001). A personalized multi-task learning neural network model detected stress with an average balanced accuracy of 0.73, significantly outperforming a general model (0.60) across all participant groups.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: