A 3D deep learning approach to epicardial fat segmentation in non-contrast and post-contrast cardiac CT images

Deep learning-based workflow for automatic extraction of atria and epicardial adipose tissue on cardiac computed tomography in atrial fibrillation

BACKGROUND

Preoperative estimation of the volume of the left atrium (LA) and epicardial adipose tissue (EAT) on computed tomography (CT) images is associated with an increased risk of atrial fibrillation (AF) recurrence. We aimed to design a deep learning-based workflow to provide reliable automatic segmentation of the atria, pericardium, and EAT for future applications in the management of AF.

METHODS

This study enrolled 157 patients with AF who underwent first-time catheter ablation between January 2015 and December 2017 at Taipei Veterans General Hospital. Three-dimensional (3D) U-Net models of the LA, right atrium (RA), and pericardium were used to develop a pipeline for total, LA-EAT, and RA-EAT automatic segmentation. We defined fat within the pericardium as tissue with attenuation between -190 and -30 HU and quantified the total EAT. Regions between the dilated endocardial boundaries and endocardial walls of the LA or RA within the pericardium were used to detect voxels attributed to fat, thus estimating LA-EAT and RA-EAT.

RESULTS

The LA, RA, and pericardium segmentation models achieved Dice coefficients of 0.960 ± 0.010, 0.945 ± 0.013, and 0.967 ± 0.006, respectively. The 3D segmentation models correlated well with the ground truth for the LA, RA, and pericardium ( r = 0.99 and p < 0.001 for all). The Dice coefficients of our proposed method for EAT, LA-EAT, and RA-EAT were 0.870 ± 0.027, 0.846 ± 0.057, and 0.841 ± 0.071, respectively.

CONCLUSION

Our proposed workflow for automatic LA, RA, and EAT segmentation using 3D U-Nets on CT images is reliable in patients with AF.

Phase Difference Network for Efficient Differentiation of Hepatic Tumors with Multi-Phase CT

Liver cancer has been one of the top causes of cancer-related death. For developing an accurate treatment strategy and raising the survival rate, the differentiation of liver cancers is essential. Multiphase CT recently acts as the primary examination method for clinical diagnosis. Deep learning techniques based on multiphase CT have been proposed to distinguish hepatic cancers. However, due to the recurrent mechanism, RNN-based approaches require expensive calculations whereas CNN-based models fail to explicitly establish temporal correlations among phases. In this paper, we proposed a phase difference network, termed as Phase Difference Network (PDN), to identify two liver cancer, hepatocellular carcinoma and intrahepatic cholangiocarcinoma, from four-phase CT. Specifically, the phase difference was used as interphase temporal information in a differential attention module, which enhanced the feature representation. Additionally, utilizing a multihead self-attention module, a transformer-based classification module was employed to explore the long-term context and capture the temporal relation between phases. Clinical datasets are used in experiments to compare the performance of the proposed strategy versus conventional approaches. The results indicate that the proposed method outperforms the traditional deep learning based methods.

Deep learning-based atherosclerotic coronary plaque segmentation on coronary CT angiography

OBJECTIVES

Volumetric evaluation of coronary artery disease (CAD) allows better prediction of cardiac events. However, CAD segmentation is labor intensive. Our objective was to create an open-source deep learning (DL) model to segment coronary plaques on coronary CT angiography (CCTA).

METHODS

Three hundred eight individuals' 894 CCTA scans with 3035 manually segmented plaques by an expert reader (considered as ground truth) were used to train (186/308, 60%), validate (tune, 61/308, 20%), and test (61/308, 20%) a 3D U-net model. We also evaluated the model on an external test set of 50 individuals with vulnerable plaques acquired at a different site. Furthermore, we applied transfer learning on 77 individuals' data and re-evaluated the model's performance using intra-class correlation coefficient (ICC).

RESULTS

On the test set, DL outperformed the currently used minimum cost approach method to quantify total: ICC: 0.88 [CI: 0.85-0.91] vs. 0.63 [CI: 0.42-0.76], noncalcified: 0.84 [CI: 0.80-0.88] vs. 0.45 [CI: 0.26-0.59], calcified: 0.99 [CI: 0.98-0.99] vs. 0.96 [CI: 0.94-0.97], and low attenuation noncalcified: 0.25 [CI: 0.13-0.37] vs. -0.01 [CI: -0.13 to 0.11] plaque volumes. On the external dataset, substantial improvement was observed in DL model performance after transfer learning, total: 0.62 [CI: 0.01-0.84] vs. 0.94 [CI: 0.87-0.97], noncalcified: 0.54 [CI: -0.04 to 0.80] vs. 0.93 [CI: 0.86-0.96], calcified: 0.91 [CI:0.85-0.95] vs. 0.95 [CI: 0.91-0.97], and low attenuation noncalcified 0.48 [CI: 0.18-0.69] vs. 0.86 [CI: 0.76-0.92].

CONCLUSIONS

Our open-source DL algorithm achieved excellent agreement with expert CAD segmentations. However, transfer learning may be required to achieve accurate segmentations in the case of different plaque characteristics or machinery.

KEY POINTS

• Deep learning 3D U-net model for coronary segmentation achieves comparable results with expert readers' volumetric plaque quantification. • Transfer learning may be needed to achieve similar results for other scanner and plaque characteristics. • The developed deep learning algorithm is open-source and may be implemented in any CT analysis software.