Generation of annotated multimodal ground truth datasets for abdominal medical image registration

Dominik F Bauer,Frank G Zöllner,Barbara I Waldkirch,Lothar R Schad,Tom Russ,Alena-Kathrin Golla,Christian Tönnes,William P Segars

doi:10.1007/s11548-021-02372-7

Abstract

PurposeSparsity of annotated data is a major limitation in medical image processing tasks such as registration. Registered multimodal image data are essential for the diagnosis of medical conditions and the success of interventional medical procedures. To overcome the shortage of data, we present a method that allows the generation of annotated multimodal 4D datasets.MethodsWe use a CycleGAN network architecture to generate multimodal synthetic data from the 4D extended cardiac–torso (XCAT) phantom and real patient data. Organ masks are provided by the XCAT phantom; therefore, the generated dataset can serve as ground truth for image segmentation and registration. Realistic simulation of respiration and heartbeat is possible within the XCAT framework. To underline the usability as a registration ground truth, a proof of principle registration is performed.ResultsCompared to real patient data, the synthetic data showed good agreement regarding the image voxel intensity distribution and the noise characteristics. The generated T1-weighted magnetic resonance imaging, computed tomography (CT), and cone beam CT images are inherently co-registered. Thus, the synthetic dataset allowed us to optimize registration parameters of a multimodal non-rigid registration, utilizing liver organ masks for evaluation.ConclusionOur proposed framework provides not only annotated but also multimodal synthetic data which can serve as a ground truth for various tasks in medical imaging processing. We demonstrated the applicability of synthetic data for the development of multimodal medical image registration algorithms.

Highlights

BackgroundMultimodal imaging plays an important part in the diagnosis of cancers, such as liver cancer [19]
A high NPS correlation coefficient (NCC) for all modalities indicates that the noise texture was emulated realistically, albeit the NCC is slightly smaller for the synthetic MRI images
The edge preservation ratio (EPR) is similar for all modalities, whereas the edge generation ratio (EGR) is largest for the cone beam CT (CBCT) images

Summary

Introduction

Multimodal imaging plays an important part in the diagnosis of cancers, such as liver cancer [19]. A variety of treatment options are available for hepatocellular carcinoma (HCC), the sixth most common malignancy worldwide and the third. Zöllner and Alena-Kathrin Golla share senior authorship. Leading cause of cancer-related deaths [10]. These include interventional procedures such as transarterial chemoembolizations (TACE) [21] or radioembolization [17]. The treatment planning benefits from using multimodal registration to combine pre- and intrainterventional data. Each imaging modality has strengths and weaknesses. Image registration enables the fusion of complementary information of each modality

Objectives

Methods

Results

Discussion

Conclusion