Duke University Secures $15M Grant to Expand AI-Driven Mental Health Prediction Model for Adolescents

Duke University's Groundbreaking AI Model for Mental Health Prediction

Duke University researchers have secured a significant $15 million federal grant to expand their artificial intelligence model designed to predict mental illness in adolescents 1

1

2

2

. The Duke Predictive Model of Adolescent Mental Health (Duke-PMA), co-developed by Professor Jonathan Posner, Assistant Professor Matthew Engelhard, and AI Health Fellow Elliot Hill, represents a potential paradigm shift in psychiatric care.

The Duke-PMA: A Proactive Approach to Mental Health

The Duke-PMA is an AI-based tool that assesses various factors related to adolescent mental health, predicting who is most likely to develop a mental illness within a year. Professor Posner emphasizes the model's potential to transform psychiatry from a reactive to a proactive approach 1

1

.

Key features of the Duke-PMA include:

1. 84% accuracy in identifying at-risk adolescents aged 10 to 15
2. Consistent performance across socioeconomic status, race, and sex
3. Utilization of questionnaires instead of expensive imaging or blood tests
4. Identification of key factors driving predictions, enabling targeted interventions

Expanding the Model's Reach

The $15 million grant marks a significant milestone in the project's development. The next phase will involve:

1. Enrolling 2,000 adolescents from rural clinics in North Carolina, Minnesota, and North Dakota
2. Conducting an observational study using Duke-PMA to assess participants and generate predictions
3. Following up with families after a year for detailed psychiatric evaluations

Professor Posner highlights the potential impact on rural settings, where mental health resources are often limited 2

2

.

Balancing Innovation and Ethical Considerations

The use of AI in adolescent mental health raises both excitement and concerns. To address these issues:

1. The Duke-PMA is designed as a supportive tool, not a replacement for clinical judgment
2. Researchers emphasize patient privacy protection in both the study and future applications
3. The model aims to enhance care while preserving essential human presence in clinical decision-making

Collaborative Effort and Future Implications

The success of this research is attributed to the collaboration of experts from various disciplines. As Professor Posner notes, Duke University is uniquely positioned for this type of interdisciplinary work 1

1

.

The Duke-PMA project represents a significant step forward in leveraging AI for mental health care, potentially revolutionizing how we approach adolescent mental health prevention and treatment.