AI can sense students levels of stress from phone data and forms

Recently scientists developed a new tool that let them see stress in plants, and while we humans are good at sensing stress in other people it’s still, for the most part, a skill that’s beyond the capability of almost all Artificial Intelligence (AI) systems, even the ones that are being used to monitor students at school in China, although slowly there are AI's emerging that can detect "intent to suicide." But, now that’s changing.

College students lead stressful lives. They’ve got assignments to complete and extracurriculars to attend, not to mention tests to prepare for and job applications to submit. Unfortunately for them, the negative health effects of stress are well-documented, and if left untreated it can cause cardiovascular diseases, affect memory and cognition, and even suppress the immune system.

To help suss out the outsized contributors to social and academic stress researchers at the University of Massachusetts in the US turned to AI which they used to predict stress levels, below median, median, or above median, from questionnaire and smartphone sensor data.

They reported that their model achieved state-of-the-art performance obtaining a 45.1% improvement compared with the baseline on a data set of student sleep patterns, activity, conversation, location, and other related data.

They describe their work in the paper “Personalized Student Stress Prediction with Deep Multitask Network“ which has been published on the preprint server Arxiv.org.

“With the growing popularity of wearable devices, the ability to use physiological data collected from these devices to predict the wearer’s mental state, such as mood and stress, suggests great clinical applications, yet such a task is extremely challenging,” wrote the co-authors. “With the induction of high-quality robust sensors in wearables like Fitbit, Apple Watch, and smartphones, efficient collection of physiological and behavioral data with reasonable accuracy has become affordable.”

The researchers’ AI system, called  Cross-personal Activity LSTM Multitask Auto-encoder Network, or CALM-Net, considers data as time-series that’s taken at successive equally-spaced points in time, and that can identify temporal patterns contained within it.

Additionally, it offers the ability to personalize models and incorporate time-series information, which improves performance as the number of students increases. And it infers and measures things like day of the week, sleep rating, sleep duration, and time to next assignment deadline.

The corpus on which the model was trained, StudentLife, was conducted at Dartmouth and facilitated by an Android app. Each day over a 10 week period it recorded stress data from 48 students via real-time responses to questionnaires, which it then paired with activity data such as walking and running, as well as other data inputs such as sound, phone charge level, and phone lock status.

“The ability of CALM-Net to incorporate granular temporal information and high-level covariates, along with an architecture which is capable of deciphering personalized patterns for each student without overfitting, contributes to its high performance,” they wrote. “[This approach] improves the performance of all evaluated models, showing that stress indicators can generally be better modelled using personalized layers.”