"PhysIt" - A Diagnosis and Troubleshooting Tool for Physiotherapists in Training

Area of Science:

• Biomedical Engineering
• Artificial Intelligence in Healthcare
• Clinical Diagnostics

Background:

• Physiotherapy diagnosis is complex due to intricate anatomical relationships and numerous tests.
• Inexperienced trainees often use inefficient, trial-and-error diagnostic methods.
• A need exists for tools to support physiotherapy students in clinical decision-making.

Purpose of the Study:

• To develop an AI-based tool, PhysIt, to assist physiotherapy students in the diagnosis process.
• To model the neuromuscular system for a Model-Based Diagnosis framework.
• To empirically evaluate the effectiveness and usability of PhysIt.

Main Methods:

• Compilation of the neuromuscular system within a Model-Based Diagnosis framework.
• Development of PhysIt, an interactive AI tool for visualization and diagnosis.
• Empirical evaluation including performance analysis on simulated and exam cases, and a user study with clinical students.

Main Results:

• The PhysIt system significantly reduced the number of candidate diagnoses considered by students.
• The system did not exclude the correct diagnosis in simulated and real-world scenarios.
• User study participants found PhysIt to be a helpful tool in their clinical studies.

Conclusions:

• AI-powered tools like PhysIt can effectively support physiotherapy students in diagnostic reasoning.
• Interactive visualization and model-based diagnosis can enhance learning and efficiency in clinical training.
• PhysIt demonstrates potential to improve the quality and speed of physiotherapy diagnostics for trainees.