4D MR Imaging Using Internal Respiratory Gating

Martin von Siebenthal, Philippe C. Cattin, U. Gamper, Alan J. Lomax, Gábor Székely

Introduction

Precision in radiation and proton therapy requires accurate tracking of internal organ motion. However, respiratory-induced deformation of organs like the liver is complex, non-rigid, and often poorly correlated with external surrogates like chest belts. Traditional 4D imaging techniques frequently assume periodic, one-dimensional motion, failing to account for irregular breathing patterns and hysteresis—the phenomenon where an organ follows different paths during inhalation and exhalation. This research presents a novel retrospective gating method for 4D MRI that utilizes internal image-based navigators to capture true 3D motion sequences during free breathing.

Methodology: Internal Respiratory Gating

The proposed method employs a slice-stacking approach. Rather than attempting to capture a full 3D volume instantaneously, which would compromise spatial resolution, 2D sagittal slices are acquired over multiple breathing cycles. To synchronize these slices into a coherent 4D sequence, the system uses "navigator slices."

For every data slice ($D_p$) recorded at a specific anatomical position, a navigator slice ($N_p$) is acquired at a fixed reference position. This creates a sequence where data frames are interleaved with navigator frames. The gating algorithm then reconstructs a 3D volume for a specific respiratory state by searching for data slices whose surrounding navigator frames match a reference navigator sequence.

Schematic of the internal gating algorithm. The comparison of temporal neighbors ($N_{i-1}, N_{i+1}$) ensures that the matching process accounts for the direction of motion, effectively capturing hysteretic effects.

A key innovation is the similarity metric used for matching. Instead of simple pixel intensity comparison, the algorithm performs template matching on prominent vascular structures within the liver. By ensuring these internal landmarks are in the same position and moving in the same direction, the method achieves high anatomical fidelity. Furthermore, multiple matching frames for the same state are averaged to improve the signal-to-noise ratio (SNR), resulting in image quality comparable to breath-hold scans.

Experimental Results

The method was validated using six healthy volunteers on a 1.5T MRI system. The liver was chosen as the target organ due to its significant respiratory-induced displacement and complex deformation.

Leave-one-out validation demonstrated high precision, with a mean deviation in frame matching of only 0.3–0.4 mm—well below the 0.9 mm pixel size. The resulting 4D reconstructions achieved a temporal resolution of 5.7 Hz, which is sufficient to capture the dynamics of quiet breathing without assuming perfect periodicity.

Multi-planar reconstructions (MPR) of the liver. The reconstructed 3D volumes show continuous vascular structures and sharp boundaries, demonstrating the effectiveness of the averaging and gating strategy.

Discussion and Conclusion

By gating based on internal anatomy rather than external sensors, this approach eliminates the "uncoupling" errors common in traditional 4D imaging. The inclusion of temporal neighbors in the gating process successfully captures the hysteretic "looping" motion of organs, providing a more realistic model for treatment planning.

While the current 30-minute acquisition time presents challenges regarding baseline drift, the method provides a robust framework for high-resolution 4D MRI. Future work will focus on interleaved acquisition sequences to mitigate drift and the development of motion models for patients unable to tolerate long scan sessions. This internal gating strategy represents a significant step toward truly personalized, motion-aware medical imaging.

Summary updated

Abstract

Respiratory organ motion is a key problem in proton therapy and in many other treatments. This paper presents a novel retrospective gating method for 4D (dynamic 3D) MR imaging during free breathing to capture the full variability of respiratory organ deformation. In contrast to other imaging methods, a constant breathing depth or even strict periodicity are not assumed. 3D images of moving organs can be reconstructed for complete respiratory cycles by retrospective stacking of dynamic 2D images using internal image-based gating. Additional noise reduction by combining multiple images significantly increases the signal-to-noise ratio. The resulting image quality is comparable to breath-hold acquisitions. Although the method was developed for proton therapy planning, the new possibilities to study respiratory motion are valuable to improve other treatments and to assess gating techniques, which rely on stronger assumptions about the breathing pattern.
BibTeX
@inproceedings{siebenthal2005d,
  author = {Martin von Siebenthal and Philippe C. Cattin and U. Gamper and Alan J. Lomax and Gábor Székely},
  title = {{4D MR Imaging Using Internal Respiratory Gating}},
  booktitle = {International Conference on Medical Image Computing and Computer-Assisted Intervention},
  year = {2005},
  doi = {10.1007/11566489\_42},
}
Figure