About Us
Let Your Happiness Journey Begin
I developed a prototype depression detection program that uses two primary inputs: facial micro-expression analysis and a brief mood questionnaire grounded in DSM-5 criteria. The technology is inspired by cutting-edge research. For example, Dartmouth College’s MoodCapture app recently demonstrated that AI can spot early depressive signs from smartphone images with about 75% accuracy. Our software operates on similar principles, utilizing open-source machine learning libraries and a training dataset of facial images labeled for emotion. The prototype has undergone initial testing with anonymized sample photos, showing promising results in correctly flagging high-distress expressions. Going forward, the plan is to refine the accuracy (possibly by incorporating voice tone analysis or social media sentiment data) and to integrate the software into a confidential school pilot program.
Why It Matters
Teen depression is an urgent, yet often hidden, problem. According to the CDC, 40% of U.S. high school students in 2023 reported persistent feelings of sadness or hopelessness. However, those struggling frequently mask their pain – it’s notoriously difficult to detect when someone is depressed if they don’t speak up about their feelings. In our own school, many students grapple with stress, isolation, or low mood without ever approaching an adult. Early identification can make a life-saving difference: the sooner a student in crisis is noticed, the sooner we can connect them with counseling or other help. This project addresses a real-world need by acting as a quiet sentinel for mental health. Rather than replacing human counselors, the software serves as a safety net, flagging potential signs of depression that might otherwise be overlooked. In an age where teens are comfortable with technology, using an AI tool to assist in mental health screening could reduce stigma and make it easier for students to accept help. Ultimately, this is about ensuring no student’s despair goes unnoticed.
How It Works
When a student voluntarily uses the tool (for instance, via a school-provided wellness app or at a kiosk in the counseling office), the front camera captures a short video of their face. The AI analyzes this in real-time for micro-expressions – those tiny, involuntary facial movements (such as tensed muscles around the eyes or subtle frowns) that can indicate underlying emotions. Concurrently, the student user answers a few quick yes/no questions about key symptoms (drawn from DSM-5 depression symptoms, like changes in sleep, appetite, or interest in activities). The software’s algorithm combines these data points to generate a risk score for depression. Importantly, this tool has been developed with privacy in mind: no images or personal data are stored without consent – it simply provides the student (and optionally, a counselor) with an assessment.
Our Clients
