Background: Armed Conflicts and forced displacement in Sudan have resulted in a high burden of psychiatric problems, including Stress Disorders, Post-Traumatic (PTSD Sadness, and Anxiety. Existing diagnostic approaches are resource-intensive and poorly adapted to crises.
Aim of the study: This study aimed to develop and validate an Artificial Intelligence (AI) driven system for the early identification of psychological risk among conflict-affected individuals and guide intervention efforts.
Methods: Using structured survey data from 349 Humans, we developed a psychological Risk Assessment framework. We trained and compared four machine learning models to predict high-risk status. The best-performing model was operationalized through an interactive, bilingual (Arabic/English) dashboard designed for humanitarian workers.
Results: Machine learning models were trained on structured survey data from 349 Humans. Gradient Boosting achieved the highest performance (Accuracy = 88.1%, F1 = 0.875), followed by Artificial Neural Networks, Computer (87.3%, 0.867) and Support Vector Machine (85.9%, 0.852). A bilingual Arabic–English interactive dashboard was developed to visualize Risk Factors, track predicted probabilities, and provide targeted Mental Health recommendations.
Conclusion: The system provides humanitarian organizations with a scalable, evidence-based instrument for early identification, enhanced Resource Allocation, and swift decision-making in Mental Health assistance.