An AI-powered app developed by Dartmouth University researchers promises to detect signs of depression using facial recognition software on your smartphone – and may be able to alert you before you even recognize it that you are depressed.
The app is called MoodCapture and it uses your phone’s front camera to analyze your emotions. As you use the face unlock feature throughout the day, the app provides “in-the-moment” analysis to “predict your depression,” Andrew Campbell, a computer scientist at Dartmouth who co-created MoodCapture, told the Daily Beast in an email.
“Mental illness has a huge impact on individuals and their families, including mine,” Campbell said. “If we can tackle the challenging issues of privacy and ethics, a future app like MoodCapture has the potential to significantly help individuals with depression. When combined with other AI tools, it could provide personalized interventions to help people function and stay healthy.”
The team plans to present their findings at the Society for Computing Machinery’s CHI conference in May 2024, but they have published an advanced copy of their paper on Arxiv. 177 participants diagnosed with major depressive disorder used MoodCapture over three months in the study.
How AI Can Reveal the World’s Oldest Archaeological Mysteries
The app collected more than 125,000 images with an average of six photos per day. By the end of the study, MoodCapture identified early signs of depression in users with 75 percent accuracy. The team also created a more personalized model that improved accuracy by up to 80 percent.
“The model focuses on multiple facial features, including gestures resulting from muscle activation, eye gaze, head posture, and 2D/3D locations of various facial landmarks such as lips and eyes,” Campbell said. “AI is used to derive meaningful insights from each image, including factors such as lighting conditions, number of people in the image, dominant colors, photo location, and background objects.”
The authors added that the app also has the ability to detect signs of depression before users even realize they have a problem. This could help them get treatment or help before their condition worsens – thereby improving mental health outcomes.
Although the app offers a lot of promise, there are many reasons for concern. For one, using AI technology in such a way could lead to unintended consequences. Emerging technology has a long and storied history of bias and discrimination. In the past, issues with AI facial recognition have led to issues such as Google Photos misidentifying Black users’ faces as animals, or people of color being falsely identified as criminal suspects by justice models criminal.
This can have harmful consequences when it comes to an app to detect depression. For example, it may be able to recognize signs of depression in people of color. Or it could tell someone they are depressed when they are not. Campbell acknowledges the issue of potential bias stating that the model “shows a higher representation of women and white participants, raising concerns about potential biases and limitations to generalizability”.
“Achieving an accuracy level of around 90 percent and dealing with bias before it is widely available is critical to the efficiency and validity of such a method,” he explained, adding later that the “billions of images” by major tech companies. a handle like Google and Apple can help provide the scale to further improve their app.
However, there are still major privacy and security concerns. Although all study participants gave informed consent to have their facial images recorded, third parties collecting personal data remains a major concern for privacy advocates – even if the data collection is being used for noble purposes such as detecting depression and its treatment.
“While there are significant privacy concerns, I believe it is possible to address these issues,” Campbell said. “For example, in the future MoodCapture app, all image processing and training will happen directly on AI processors inside your phone, ensuring that no images leave the device.”
For now, MoodCapture is still a proof of concept. AI is not all chatbots and “woke” image generators. It could also allow you to learn a little more about yourself and your mental health – and get you help when you need it.
“A lot of people don’t have access to mental health professionals, but most people have phones,” Campbell said. “Achieving this goal is like a milestone in the field of mental health.”
Read more at The Daily Beast.
Get the Daily Beast’s biggest scoops and scandals delivered straight to your inbox. Sign up now.
Stay informed and get unlimited access to The Daily Beast’s unmatched reporting. Subscribe now.