This AI model was created to confront health disparities and educational inequities, amid a rise in mental health issues reported among American college students.
The model is designed to identify a range of mental health issues and increase the quality of students' lives and education.
Yusen Zhai, director of the UAB Community Counseling Clinic, used artificial intelligence to spot patterns in information that schools collect, such as age, biological sex, years in school, race and ethnicity, and majors, that could be indicators of a higher risk of mental health conditions. In a recently published article, Zhai explained potential benefits of using predictive models to help prevent and allow intervention for anxiety and depressive disorders among college students.
“Counselors can help improve mental health, but there is a problem: There are not enough tools or resources for counselors to advise those who are at risk before serious mental health problems occur,” he told AL.com. “There are traditional assessments like the clinical assessments, and self-report questionnaires etc., but they often face challenges such as the student may have stigma or the student may have limited access to those services. So this method usually assesses the risk after the student asks for help. With AI, we can use that to improve the situation by developing the AI model based on data universally collected. This can help us be more proactive rather than reactive.”
Zhai and his team developed machine learning predictive models that do not rely on clinical samples or health-related information, but rather socioeconomic demographics that research has shown to be associated with more severe anxiety and depression. Factors such as gender, race and ethnicity, financial stress, a sense of belonging on campus, disability status, and age are considered in the model’s evaluation.
Students don’t have to fill out additional information; the model simply helps counselors understand which groups of students might benefit from additional resources, like a reminder of how to utilize on-campus resources.
“The model will help us identify which risk factors are the most important and help us rank them from the highest to the lowest so the counselor can focus on the most critical issue first,” he said.
Zhai and his fellow researchers found that biological sex and race intersect in ways that can intensify stress — for example, female students from minority backgrounds may contend with both cultural pressures and racial discrimination, amplifying their risk for anxiety and depression. Similarly, disability status can heighten the impact of financial stress or racial identity, further marginalizing students and increasing their vulnerability to anxiety and depression.
“Human knowledge is essential in the development of this kind of tool and providing services. Empathy and humanity are the two most important imperatives in counseling or any type of professional health services,” Zhai said. “This AI tool is meant to be the assistant rather than replacing the human counselor or health professionals. The purpose for the model is to provide counselors with more data driven insights. So the counselor still uses their own clinical judgment and integrates the data from the AI to make a more ethical and data informed decision.”
The American College Health Association’s Spring 2023 national survey of over 55,000 undergraduate students revealed that approximately 76 percent were experiencing moderate to serious psychological distress. In 2022, BestColleges did a mental health survey of college students and found that 46 percent of respondents said their mental health status was fair or poor. Out of that group of students, only 20 percent have sought assistance through their school. Out of all the students surveyed, 51 percent agreed that their mental health has worsened during their time in college.
Zhai’s AI tool is currently based on college students and universities, but he said the model may also be applied to other groups of people, such as high school students.
“This is our next step,” Zhai said. “There is intensive data collected already by the school district and the high schools, so we are developing the AI model to analyze those data and provide the school psychologists and the teachers more data-driven insight on which students might be at risk and have mental health concerns, or learning issues.”
Breonna Atkinsis a senior at Carver High School in Birmingham and a reporting intern at AL.com through the Birmingham Promise program.
©2025 Advance Local Media LLC. Distributed by Tribune Content Agency, LLC.
