Aiberry is creating a multi-modal artificial intelligence (AI) platform that analyzes facial, audio and text features to screen for mental illness. This multicenter study will be used to collect data to validate the platform's ability to detect depression and anxiety in a diverse patient population.
Aiberry is developing a proprietary AI platform that leverages machine learning to screen for mental illness using a multi-modal approach that fuses rich visual, audio, text, gestures, and eye gaze information. Our objective in this study is to validate algorithms that predict how individuals will respond to self-report questionnaires used to screen for major depressive disorder (MDD) and general anxiety disorder (GAD). Prior to enrollment, participants will participate in a baseline screening questionnaire to collect demographic information, health history, and current depression severity. If participants meet eligibility requirements and demographic recruitment targets, they will be invited to a single virtual study visit in which they will complete a 10-15 minute recorded interview with a study staff member along with three brief self-report questionnaires: 1) the Quick Inventory of Depression Symptoms Self Report (QIDS-SR-16), 2) the General Anxiety Disorder (GAD-7), and 3) the mini-version of the Mood and Anxiety Questionnaire (mini-MASQ), which will be used as a validity check that participant responses are consistent between this and the previous two questionnaires. We will evaluate how well the AI technology is able to predict self-reported symptoms of depression and anxiety by analyzing facial, audio, and text features from interview videos.
Study Type
OBSERVATIONAL
Enrollment
800
University of Arizona
Tucson, Arizona, United States
Georgetown University
Washington D.C., District of Columbia, United States
Hapworth Research Inc.
New York, New York, United States
Prediction of depression score from a self-report instrument
Error when predicting depression severity (none, mild, moderate, severe) based on QIDS-SR-16.
Time frame: Single assessment
Classification of moderate-severe depression
Accuracy, sensitivity, specificity, and positive and negative predictive value when classifying whether participants meet established cut-off criterion for suspected MDD (QIDS \> 10)
Time frame: Single assessment
Classification of moderate-severe anxiety
Accuracy, sensitivity, specificity, and positive and negative predictive value when classifying whether participants meet established cut-off criterion for suspected GAD (GAD-7 \> 9)
Time frame: Single assessment
Prediction of anxiety score from a self-report instrument
Error when predicting anxiety severity (none, mild, moderate, severe) based on GAD-7.
Time frame: Single assessment
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.