Respiratory Motion Modelling Using cGANs

Abstract

Respiratory motion models in radiotherapy are considered as one possible approach for tracking mobile tumours in the thorax and abdomen with the goal to ensure target coverage and dose conformation. We present a patient-specific motion modelling approach which combines navigator-based 4D MRI with recent developments in deformable image registration and deep neural networks. The proposed regression model based on conditional generative adversarial nets (cGANs) is trained to learn the relation between temporally related US and MR navigator images. Prior to treatment, simultaneous ultrasound (US) and 4D MRI data is acquired. During dose delivery, online US imaging is used as surrogate to predict complete 3D MR volumes of different respiration states ahead of time. Experimental validations on three volunteer lung datasets demonstrate the potential of the proposed model both in terms of qualitative and quantitative results, and computational time required.

Inter-fractional Respiratory Motion Modelling from Abdominal Ultrasound: A Feasibility Study

“…The model predicts dense motion information 133 ms into the future which allows for system latency compensation. The obtained results are similar in terms of accuracy to those presented in previous studies [2,9]. However, we additionally present preliminary findings for inter-fractional motion modelling which involves a repositioning of the US probe.…”

“…Given the p latest surrogates {s t−i } p−1 i=0 , the signal s t+n at time t + n is approximated by applying the AR model n times. Finally, the motion estimate y t+n is computed given equation (2) and warped in order to match the actual patient position.…”

“…It requires precise co-registration of US and MR images in order to establish correspondence between tracked liver points. Indirect lung tumour tracking strategies based on 2D abdominal US have only been proposed recently [2,7]. Mostafaei et al [7] combine US imaging and cone-beam CT (CBCT) in order to reduce the CBCT imaging frequency and therefore the imaging dose to the patient.…”

“…However, the tumour motion is estimated in superior-inferior (SI) direction only. In [2] dense motion information was predicted based on an adversarial neural network. Although promising, it is not clear how this approach performs if the US imaging plane is shifted.…”

Deep Learning in the Biomedical Applications: Recent and Future Status

“…To model the respiratory motion for tracking mobile tumors in the thorax and abdomen in a radiotherapy, Giger et al [320] developed a conditional generative adversarial network trained to learn the relation between temporally related ultrasound and 4D MRI navigator images.…”