ABSTRACT

Objective: This study aimed to develop an AI-based chatbot system, the Sensory Profile 2 Summary and Interpretation (SPSI), to improve the interpretability, objectivity, and visualization of the Sensory Profile 2 (SP-2), a widely used sensory integration assessment tool in occupational therapy.Methods: The structure and scoring rules of the SP-2 were embedded into a chatbot framework using a GPT-based language model. The system provides automated scoring, quadrant-based interpretation, personalized strategy suggestions, and visual summaries (e.g., pie charts and keyword-based graphs) through natural language interaction. OpenAPI integration enabled seamless data input and output across platforms.Results: The SPSI reduced evaluation time and interpretation burden by automating the scoring and reporting processes. It improved assessment clarity and accessibility for nonspecialist users, including caregivers, by delivering results in intuitive visual and textual formats. The chatbot consistently provided standardized interpretations based on SP-2 guidelines, minimizing evaluator bias and enhancing result communication.Conclusion: The SPSI chatbot system demonstrates the potential of AI-driven tools to facilitate user-centered, efficient, and comprehensible sensory assessments. Future work should include clinical validation, incorporation of explainable AI for decision transparency, and implementation of robust data protection measures for practical deployment.