The internal–external respiratory motion correlation is unaffected by audiovisual biofeedback

Abstract

This study evaluated if an audiovisual (AV) biofeedback causes variation in the level of external and internal correlation due to its interactive intervention in natural breathing. The internal (diaphragm) and external (abdominal wall) respiratory motion signals of 15 healthy human subjects under AV biofeedback and free breathing (FB) were analyzed and measures of correlation and regularity taken. Regularity metrics (root mean square error and spectral power dispersion metric) were obtained and the correlation between these metrics and the internal and external correlation was investigated. For FB and AV biofeedback assisted breathing the mean correlations found between internal and external respiratory motion were 0.96 ± 0.02 and 0.96 ± 0.03, respectively. This means there is no evidence to suggest (p-value = 0.88) any difference in the correlation between internal and external respiratory motion with the use of AV biofeedback. Our results confirmed the hypothesis that the internal-external correlation with AV biofeedback is the same as for free breathing. Should this correlation be maintained for patients, AV biofeedback can be implemented in the clinic with confidence as regularity improvements using AV biofeedback with an external signal will be reflected in increased internal motion regularity.

Dynamic keyhole: A novel method to improve MR images in the presence of respiratory motion for real-time MRI

“…Alternatively, the external respiratory signals can be easily utilized with only a small penalty in terms of k-line reduction, as there is a high internal-external signal correlation for the dataset used in this study (Pearson's R-value = 0.96). 31 The dynamic keyhole method uses previously acquired full k-space datasets, and in an actual implementation, the library of k-space datasets for the prior peripheral k-space dataset and the central k-space dataset acquisition would be at a different frequency. Even though the library may be acquired at a lower temporal resolution, the dynamic keyhole method should give improved results over the conventional keyhole method because it utilizes multiple prior peripheral datasets matched to various respiratory displacements.…”

“…Overall, for these datasets, the correlation of internal and external signals was found to be very high in a previous study, an average Pearson's R-value of 0.96. 31 The 30 Hz external signal was downsampled to the temporal resolution of 5 Hz for (1) the use of respiratory displacements at the closest acquisition time between respiratory signals and full k-space datasets in Fig. 1(3), and (2) the use of the 60 datasets acquired at 5 Hz temporal resolution during the acquisition of central k-space datasets in Fig.…”

Audiovisual Biofeedback Improves Cine–Magnetic Resonance Imaging Measured Lung Tumor Motion Consistency

“…Audiovisual biofeedback uses an external respiratory signal from a real-time position management (RPM) system (Varian, Palo Alto, CA) to track patient breathing in real-time. Previous AV biofeedback results have demonstrated that the breathing consistency of external and internal surrogates has been improved (10,11,16,17) while maintaining a robust correlation between external and internal breathing motion (20). However, AV biofeedback results on tumor motion have been less conclusive (21,22), with additional further investigation strongly suggesting that patient compliance and performance with AV biofeedback improve with time (23).…”

Audiovisual biofeedback improves the correlation between internal/external surrogate motion and lung tumor motion

“…Lung tumor was automatically contoured using a region growing algorithm and its centroid was used for measuring tumor motion . Diaphragm motion was also automatically measured at the peak of liver dome scout . Thus, respiratory motions of diaphragm, abdomen, and tumor had the same time point and duration. Diaphragm and tumor motion was linearly interpolated from 3.3 to 25 Hz for further correlation calculation, as well as the frequency of abdomen motion (25 Hz).…”

“…Although it has been demonstrated that AV biofeedback improves breathing motion reproducibility, it has been conjectured that this improvement might be offset by lower surrogate to tumor correlation, as the act of biofeedback affects how the patient breathes. A previous study on healthy subjects found that the correlation of respiratory motion between internal and external surrogates (diaphragm and abdomen) was the same with free breathing (FB) and AV biofeedback breathing. However, no patient studies have investigated this correlation and the volunteer study did not consider analysis with respect to motion directions and of the effects of phase shifts on the correlation.…”