Deep learning‐based cardiac cine segmentation: Transfer learning application to 7T ultrahigh‐field MRI. Issue 4 (18th May 2021)

• Record Type: Journal Article
• Authors: Ankenbrand, Markus Johannes; Lohr, David; Schlötelburg, Wiebke; Reiter, Theresa; Wech, Tobias; Schreiber, Laura Maria
• Is Part Of: Magnetic resonance in medicine. Volume 86:Issue 4(2021)
• Page Start: 2179
• Page End: 2191
• Publication Date: 2021-05-18
• Subjects: 7T -- cardiac function -- cardiac magnetic resonance -- deep learning -- neural networks -- segmentation -- transfer learning -- ultrahigh‐field; Nuclear magnetic resonance -- Periodicals; Electron paramagnetic resonance -- Periodicals; 616.07548
• DOI: 10.1002/mrm.28822
• Languages: English
• ISSNs: 0740-3194

Abstract

Abstract : Purpose: Artificial neural networks show promising performance in automatic segmentation of cardiac MRI. However, training requires large amounts of annotated data and generalization to different vendors, field strengths, sequence parameters, and pathologies is limited. Transfer learning addresses this challenge, but specific recommendations regarding type and amount of data required is lacking. In this study, we assess data requirements for transfer learning to experimental cardiac MRI at 7T where the segmentation task can be challenging. In addition, we provide guidelines, tools, and annotated data to enable transfer learning approaches by other researchers and clinicians. Methods: A publicly available segmentation model was used to annotate a publicly available data set. This labeled data set was subsequently used to train a neural network for segmentation of left ventricle and myocardium in cardiac cine MRI. The network is used as starting point for transfer learning to 7T cine data of healthy volunteers (n = 22; 7873 images) by updating the pre‐trained weights. Structured and random data subsets of different sizes were used to systematically assess data requirements for successful transfer learning. Results: Inconsistencies in the publically available data set were corrected, labels created, and a neural network trained. On 7T cardiac cine images the model pre‐trained on public imaging data, acquired at 1.5T and 3T, achieved DICELV = 0.835 and DICEMY = 0.670. Transfer learning using 7T cine data and ImageNet weight initialization improved model performance to DICELV = 0.900 and DICEMY = 0.791. Using only end‐systolic and end‐diastolic images reduced training data by 90%, with no negative impact on segmentation performance (DICELV = 0.908, DICEMY = 0.805). Conclusions: This work demonstrates and quantifies the benefits of transfer learning for cardiac cine image segmentation. We provide practical guidelines for researchers planning transfer learning projects in cardiac MRI and make data, models, and code publicly available.