Subpopulation-based correspondence modelling for improved respiratory motion estimation in the presence of inter-fraction motion variations

Abstract

Correspondence modelling between low-dimensional breathing signals and internal organ motion is a prerequisite for application of advanced techniques in radiotherapy of moving targets. Patient-specific correspondence models can, for example, be built prior to treatment based on a planning 4D CT and simultaneously acquired breathing signals. Reliability of pre-treatment-built models depends, however, on the degree of patient-specific inter-fraction motion variations. This study investigates whether motion estimation accuracy in the presence of inter-fraction motion variations can be improved using correspondence models that incorporate motion information from different patients. The underlying assumption is that inter-patient motion variations resemble patient-specific inter-fraction motion variations for subpopulations of patients with similar breathing characteristics. The hypothesis is tested by integrating a sparse manifold clustering approach into a regression-based correspondence modelling framework that allows for automated identification of patient subpopulations. The evaluation is based on a total of 73 lung 4D CT data sets, including two cohorts of patients with repeat 4D CT scans (cohort 1: 14 patients; cohort 2: ten patients). The results are consistent for both cohorts: The subpopulation-based modelling approach outperforms general population modelling (models built on all data sets available) as well as pre-treatment-built models trained on only the patient-specific motion information. The results thereby support the hypothesis and illustrate the potential of subpopulation-based correspondence modelling.

Continuous generation of volumetric images during stereotactic body radiation therapy using periodic kV imaging and an external respiratory surrogate

“…Incorporating models developed from pre-treatment 4DCBCT [25] rather than a planning 4DCT may also improve the accuracy of the method by compensating for day-to-day changes in patient respiratory motion that may differ from the planning 4DCT model during treatment. Compensating for the changes in internal patient anatomy motion the occurs between the time of planning and the time of treatment could also be done by using subpopulation-based correspondence modeling [96]. Moreover, some techniques have been proposed to model respiratory motion without 4DCT images obtained prior to treatment, which could further reduce the dose delivered to the patient in a clinical context [97,98].…”

A 4D global respiratory motion model of the thorax based on CT images: A proof of concept

“…The clinical data used in this study were acquired on a GE Lightspeed multislice CT (MSCT) and Discovery combined PET/CT scanners with a cine CT scan capability. These datasets are acquired in a similar fashion as those used in Fayad et al . During data acquisition, the cine 4D CT utilizes the data acquisition modes of 8 × 2.5 mm (2 cm) to produce images of 2.5 mm slice thickness .…”

“…In order to resolve this issue, many population models have been previously developed . These models have shown that motion estimation accuracy in the presence of interfraction motion variations can be improved using correspondence models that incorporate motion information from different patients.…”

“…8 In order to resolve this issue, many population models have been previously developed. [9][10][11] These models have shown that motion estimation accuracy in the presence of interfraction motion variations can be improved using correspondence models that incorporate motion information from different patients. However, despite this demonstrated improvement of the interfraction motion variation, a potential drawback is that these models require a 4D CT acquisition for each patient and most of them are still based on 1D or 2D respiratory signals.…”

PET respiratory motion correction: quo vadis?

“…In CT, many population models have been previously developed (McQuaid et al 2009, Geimer et al 2017, Fayad, Buerger, Tsoumpas, Cheze-Le-Rest & Visvikis 2012, Wilms et al 2017). These models have shown that motion estimation accuracy in the presence of inter-fraction motion variations can be improved using correspondence models that incorporate motion information from different patients.…”