Automated T1 and T2 mapping segmentation on cardiovascular magnetic resonance imaging using deep learning

András Kalapos; Liliána Szabó; Zsófia Dohy; Máté Kiss; Béla Merkely; Bálint Gyires-Tóth; Hajnalka Vágó

doi:10.3389/fcvm.2023.1147581

Automated T1 and T2 mapping segmentation on cardiovascular magnetic resonance imaging using deep learning

Front Cardiovasc Med. 2023 Jul 7:10:1147581. doi: 10.3389/fcvm.2023.1147581. eCollection 2023.

Authors

András Kalapos¹, Liliána Szabó², Zsófia Dohy², Máté Kiss³, Béla Merkely^{2

4}, Bálint Gyires-Tóth¹, Hajnalka Vágó^{2

4}

Affiliations

¹ Department of Telecommunications and Media Informatics, Faculty of Electrical Engineering and Informatics, Budapest University of Technology and Economics, Budapest, Hungary.
² Semmelweis University, Heart and Vascular Centre, Budapest, Hungary.
³ Siemens Healthcare, Budapest, Hungary.
⁴ Department of Sports Medicine, Semmelweis University, Budapest, Hungary.

Abstract

Introduction: Structural and functional heart abnormalities can be examined non-invasively with cardiac magnetic resonance imaging (CMR). Thanks to the development of MR devices, diagnostic scans can capture more and more relevant information about possible heart diseases. T1 and T2 mapping are such novel technology, providing tissue specific information even without the administration of contrast material. Artificial intelligence solutions based on deep learning have demonstrated state-of-the-art results in many application areas, including medical imaging. More specifically, automated tools applied at cine sequences have revolutionized volumetric CMR reporting in the past five years. Applying deep learning models to T1 and T2 mapping images can similarly improve the efficiency of post-processing pipelines and consequently facilitate diagnostic processes.

Methods: In this paper, we introduce a deep learning model for myocardium segmentation trained on over 7,000 raw CMR images from 262 subjects of heterogeneous disease etiology. The data were labeled by three experts. As part of the evaluation, Dice score and Hausdorff distance among experts is calculated, and the expert consensus is compared with the model's predictions.

Results: Our deep learning method achieves 86% mean Dice score, while contours provided by three experts on the same data show 90% mean Dice score. The method's accuracy is consistent across epicardial and endocardial contours, and on basal, midventricular slices, with only 5% lower results on apical slices, which are often challenging even for experts.

Conclusions: We trained and evaluated a deep learning based segmentation model on 262 heterogeneous CMR cases. Applying deep neural networks to T1 and T2 mapping could similarly improve diagnostic practices. Using the fine details of T1 and T2 mapping images and high-quality labels, the objective of this research is to approach human segmentation accuracy with deep learning.

Keywords: MRI; T1 and T2 mapping; artificial intelligence; cardiac segmentation; deep learning.

Grants and funding

The work reported in this paper, carried out as a collaboration between SE and BME, has been partly supported by the European Union project RRF-2.3.1-21-2022-00004 within the framework of the Artificial Intelligence National Laboratory. Project no. TKP2021-NKTA- 46 has been implemented with the support provided by the Ministry of Innovation and Technology of Hungary from the National Research, Development and Innovation Fund, financed under the TKP2021-NKTA funding scheme. The research reported in this paper is part of project no. BME-NVA-02, implemented with the support provided by the Ministry of Innovation and Technology of Hungary, financed under the TKP2021 funding scheme.