• AI Can Now Measure Your Stress Without a Wearable

  • March 13, 2020

    Today, Mobio Interactive published a peer-reviewed scientific paper that demonstrates the company's AI can objectively and accurately measure an individual's stress using nothing more than a 30-second selfie video captured with a mobile camera. Even when tested on video data from people that the AI had never seen, the measure of stress still reached accuracy above 80 percent. The work represents a major step forward in the objective assessment of wellbeing and can be leveraged for countless medical applications. Mobio Interactive is already using these data to personalise the clinically validated resilience app Am Mindfulness to each individual user. The study sampled the data from over ten thousand selfie videos that were completed in tandem with self-assessments of stress primarily before and after practicing meditation. Data from people in over one hundred countries was used, helping to minimize regional bias within the neural network. The results were published today in the journal Biomedical Signal Processing and Control in collaboration with the Vision and Image Processing (VIP) lab at the University of Waterloo.

     

    Read the full study

    ABOUT MI
    Mobio Interactive develops and commercialises clinically validated and objectively quantified software to prevent, measure and treat mental illness.
    https://www.midigitaltherapeutics.com

     

    ABOUT AM MINDFULNESS
    Am Mindfulness is a mobile app that supports a personalized mindfulness practice through a diverse library of guided meditations curated by trained meditation instructors. Unique to Am is the ability to objectively and remotely quantify stress at scale with 86% accuracy in under 30 s through a mobile camera selfie video (no need for a wearable). Am is available in English, Mandarin, Dutch and German.
    https://www.midigitaltherapeutics.com/am-mindfulness.html

     

    ABOUT THE VISION AND IMAGE PROCESSING LAB

    The VIP Lab is dedicated to understanding visual processes and finding solutions for the outstanding problems in visual processing and perception, as well as artificial intelligence, machine learning, and intelligent systems for a wide variety of applications.
    https://uwaterloo.ca/vision-image-processing-lab