Deep Learning-based Automated Segmentation of Left Ventricular Trabeculations and Myocardium on Cardiac MR Images: A Feasibility Study.

Purpose: To develop and evaluate a complete deep learning pipeline that allows fully automated end-diastolic left ventricle (LV) cardiac MRI segmentation, including trabeculations and automatic quality control of the predicted segmentation.
Materials and Methods: This multicenter retrospective study includes training, validation, and testing datasets of 272, 27, and 150 cardiac MR images, respectively, collected between 2012 and 2018. The reference standard was the manual segmentation of four LV anatomic structures performed on end-diastolic short-axis cine cardiac MRI: LV trabeculations, LV myocardium, LV papillary muscles, and the LV blood cavity. The automatic pipeline was composed of five steps with a DenseNet architecture. Intraobserver agreement, interobserver agreement, and interaction time were recorded. The analysis includes the correlation between the manual and automated segmentation, a reproducibility comparison, and Bland-Altman plots.
Results: The automated method achieved mean Dice coefficients of 0.96 ± 0.01 (standard deviation) for LV blood cavity, 0.89 ± 0.03 for LV myocardium, and 0.62 ± 0.08 for LV trabeculation (mean absolute error, 3.63 g ± 3.4). Automatic quantification of LV end-diastolic volume, LV myocardium mass, LV trabeculation, and trabeculation mass-to-total myocardial mass (TMM) ratio showed a significant correlation with the manual measures ( r = 0.99, 0.99, 0.90, and 0.83, respectively; all P < .01). On a subset of 48 patients, the mean Dice value for LV trabeculation was 0.63 ± 0.10 or higher compared with the human interobserver (0.44 ± 0.09; P < .01) and intraobserver measures (0.58 ± 0.09; P < .01). Automatic quantification of the trabeculation mass-to-TMM ratio had a higher correlation (0.92) compared with the intra- and interobserver measures (0.74 and 0.39, respectively; both P < .01).
Conclusion: Automated deep learning framework can achieve reproducible and quality-controlled segmentation of cardiac trabeculations, outperforming inter- and intraobserver analyses. Supplemental material is available for this article. © RSNA, 2020.
Competing Interests: Disclosures of Conflicts of Interest: : A.B. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: disclosed no relevant relationships. Other relationships: institution has patents: SATT-SUD-EST, date of deposit 12/13/2019, demand number: DSO000042312, record number: DSO2019018721, depositors: AMU/CNRS/AP-HM. J.F. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: disclosed no relevant relationships. Other relationships: institution has patents: SATT-SUD-EST, patent deposit number: DSO2019018721. Z.B. disclosed no relevant relationships. G.H. disclosed no relevant relationships. A.L. disclosed no relevant relationships. M.B. disclosed no relevant relationships. L.B. disclosed no relevant relationships. F.P. disclosed no relevant relationships. J.N.D. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: author is consultant for Microport; author paid for lectures by Bayer Healthcare, Siemens, Shire Takeda, and MSD; author received travel accommodations from GE Healthcare. Other relationships: disclosed no relevant relationships. B.G. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: disclosed no relevant relationships. Other relationships: institution (Aix Marseille University) has issued patents: A software is protected by the Satt Sud Est, deposit number DSO2019018721. A.J. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: disclosed no relevant relationships. Other relationships: institution has issued patents: CNRS/AMU/APHM, date of deposit 12/13/2019, demand number: DSO000042312, record number: DSO2019018721, depositors: AMU/CNRS/AP-HM.
(2021 by the Radiological Society of North America, Inc.)