WE-G-213CD-07: Enhancing Respiratory Motion Prediction Accuracy Using Audiovisual (AV) Biofeedback.

Area of Science:

• Medical Physics
• Radiotherapy
• Image-Guided Therapy

Background:

• Respiratory motion introduces significant uncertainties in radiotherapy targeting.
• Irregular breathing patterns complicate accurate prediction of tumor movement during treatment.
• Audiovisual (AV) biofeedback offers a method to regularize breathing patterns.

Purpose of the Study:

• To enhance the accuracy of respiratory-motion prediction for radiotherapy.
• To evaluate the efficacy of an AV biofeedback system in conjunction with real-time MRI.
• To investigate the impact of AV biofeedback on breathing pattern complexity and motion prediction.

Main Methods:

• Implementation of an AV biofeedback system integrated with real-time MRI.
• Tracking of abdominal motion using an RPM system to guide patient breathing.
• Utilizing kernel density estimation for real-time motion prediction with optimized parameters (Delay Time/Training Examples).

Main Results:

• A 67% reduction in prediction error was achieved with AV biofeedback compared to free breathing.
• Mean difference between measured and predicted motion decreased from 1.98±2.32mm (free breathing) to 0.65±0.65mm (AV biofeedback).
• Optimal prediction parameters were identified as Delay Time/Training Examples = 500 ms/250.

Conclusions:

• AV biofeedback effectively regularizes breathing patterns, leading to improved respiratory motion prediction accuracy.
• This approach holds promise for enhancing the precision and safety of image-guided radiotherapy.
• The study validates the use of AV biofeedback as a tool to mitigate breathing-related uncertainties in radiation oncology.