(Sunday, 7/29/2018) 3:00 PM - 6:00 PM
Room: Exhibit Hall
Purpose: To evaluate the performance of respiration prediction model of a commercial respiration visual coaching device (VCD).
Methods: The commercial respiration VCD (Varian Medical Systems, Palo Alto, CA) enables patient coaching for consistent and reproducible respiration during a 4D CT scan or respiration gated radiotherapy. The device displays the respiration signal with a band of +/- 12.5% tolerance range on a video monitor to guide the patient to breath within the tolerance range. 40 visual coaching sessions tested with respiration signals (10 cycles/minute) simulated with a motion phantom were recorded with both 10% and 20% Breath Predictive Filters. The difference between real-time tracking and VCD predicted respirations can make the periodic respiration signal fall out of the tolerance. These recorded coaching sessions were analyzed to evaluate the performance of the prediction model of the VCD in terms of the time till failure (TTF), i.e., the time from the beginning of visual coaching till the respiration falling out of the tolerance.
Results: The average TTFs are 171s and 228s for the 10% and 20% Breath Predictive Filter, respectively. The lowest TTF recorded was 35s, corresponding to less than 6 respiration cycles. If respiration tracking was started at the same time as the visual coaching session, the respiration pattern predicted by the VCD tended to be more accurate, yielding a much higher average TTF of 389s, compared to 127s if real-time tracking and visual coaching were not synchronized.
Conclusion: The uncertainty of the VCD respiration prediction model can result in relatively large errors in the predicted respiration pattern. The VCD prediction model is more accurate when the coaching session is synchronized with respiration tracking.
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email
