Click here to


Are you sure ?

Yes, do it No, cancel

Reproducibility and Correction of the Respiratory Phase Shift for Enhanced Internal-External Motion Correlation

A Milewski*, D Olek, W Martin, A Rimner, G Li, Memorial Sloan Kettering Cancer Center, New York, NY


(Sunday, 7/12/2020)   [Eastern Time (GMT-4)]

Room: AAPM ePoster Library

Purpose: To determine whether the patient-specific respiratory phase shift is reproducible over time and whether phase-shift correction effectively enhances internal-external correlation.

Methods: Under an IRB-approved protocol, ten healthy volunteers received coronal and sagittal respiratory-correlated four-dimensional magnetic resonance imaging (RC-4DMRI) scans 10 minutes apart. Each 4DMRI scan lasted 5-16 minutes. Internal-navigator and external-bellows waveforms were simultaneously acquired during each scan: The navigator interrogated the lung-diaphragm interface and the bellows was placed 5cm inferior to the sternum. We designed an algorithm that maximizes the waveforms’ cross-correlation to calculate and correct the phase shift between the two waveforms within a series of automatically-selected ten-second time window. We compared the phase shifts and correlation enhancement effected by phase-shift correction in the coronal and sagittal scans. Statistical significance was evaluated by the Mann-Whitney U test (p<0.05).

Results: Averaged across all volunteers, the mean phase shift is 39o±12o during the coronal scan and 41o±15o during the sagittal scan. The distributions of mean phase shifts for the coronal and sagittal scans are indistinguishable (p=0.79). The coronal scan’s mean phase shift is within 14% of the sagittal scan’s mean for 8/10 volunteers. The phase-shift distributions for the coronal and sagittal scans are indistinguishable in 6/10 volunteers. Phase-shift correction increases the mean correlation from 0.57±0.23 to 0.83±0.08 for the coronal scan and from 0.50±0.27 to 0.79±0.09 for the sagittal scan. The distributions of mean enhanced correlations for the coronal and sagittal scans are indistinguishable (p=0.27). The coronal scan’s mean enhanced correlation is within 16% of the sagittal scan’s mean for 9/10 volunteers (<6% for 7/10, <28% for 10/10).

Conclusion: This study suggests that the phase shift between internal and external respiratory motions is stable over 15 minutes in most individuals. Correcting the phase shift enhances the correlation between these motions and is potentially useful for respiratory-gated radiotherapy.

Funding Support, Disclosures, and Conflict of Interest: The MSKCC and Philips Healthcare have a master research agreement.


Image-guided Therapy, MRI, Target Localization


TH- RT Interfraction Motion Management: external markers-based

Contact Email