A deep learning pipeline for automatic analysis of multi-scan cardiovascular magnetic resonance

Abstract

Background Cardiovascular magnetic resonance (CMR) sequences are commonly used to obtain a complete description of the function and structure of the heart, provided that accurate measurements are extracted from images. New methods of extraction of information are being developed, among them, deep neural networks are powerful tools that showed the ability to perform fast and accurate segmentation. Iq1n order to reduce the time spent by reading physicians to process data and minimize intra- and inter-observer variability, we propose a fully automatic multi-scan CMR image analysis pipeline. Methods Sequence specific U-Net 2D models were trained to perform the segmentation of the left ventricle (LV), right ventricle (RV) and aorta in cine short-axis, late gadolinium enhancement (LGE), native T1 map, post-contrast T1, native T2 map and aortic flow sequences depending on the need. The models were trained and tested on a set of data manually segmented by experts using semi-automatic and manual tools. A set of parameters were computed from the resulting segmentations such as the left ventricular and right ventricular ejection fraction (EF), LGE scar percentage, the mean T1, T1 post, T2 values within the myocardium, and aortic flow. The Dice similarity coefficient, Hausdorff distance, mean surface distance, and Pearson correlation coefficient R were used to assess and compare the results of the U-Net based pipeline with intra-observer variability. Additionally, the pipeline was validated on two clinical studies. Results The sequence specific U-Net 2D models trained achieved fast (≤ 0.2 s/image on GPU) and precise segmentation over all the targeted region of interest with high Dice scores (= 0.91 for LV, = 0.92 for RV, = 0.93 for Aorta in average) comparable to intra-observer Dice scores (= 0.86 for LV, = 0.87 for RV, = 0.95 for aorta flow in average). The automatically and manually computed parameters were highly correlated (R = 0.91 in average) showing results superior to the intra-observer variability (R = 0.85 in average) for every sequence presented here. Conclusion The proposed pipeline allows for fast and robust analysis of large CMR studies while guaranteeing reproducibility, hence potentially improving patient’s diagnosis as well as clinical studies outcome.

Assessment of right ventricular size and function from cardiovascular magnetic resonance images using artificial intelligence

“…In a study by Hakim et al, when comparing automated RVEF measurements to a human expert generated reference values, the DL algorithm showed an r-value of 0.76. The major source of error has typically been explained by poor automated segmentation at the basal slice, as was observed in prior studies [ 35 ]. In our prior study [ 15 ], we compared manual RVEF by a clinician to automated RVEF measurements made by three DL methods developed using publicly available CMR datasets and found that all three DL methods performed relatively poorly in a significant proportion of the patients.…”

Diagnostic utility of artificial intelligence for left ventricular scar identification using cardiac magnetic resonance imaging—A systematic review

“…All methods were compared to a minimum of human delineated ground-truth. Four studies compared performance of proposed and pre-defined thresholding methods [13][14][15][16] .…”

“…Metadata can vary between manufacturers and magnet strengths 22 and models trained with specific data may not be generally applicable. Only a small subset of studies included multiple CMR manufactures to mitigate this risk 14,21,[23][24][25][26] . Training data on various field strengths would allow greater generalisability clinically but 1.5T remains the current standard with 3T employed mainly by more experienced imaging centres 5 .…”

Motion correction for native myocardial T1 mapping using self‐supervised deep learning registration with contrast separation

“…Therefore, researchers have introduced deep learning based methods into the cardiac MR image processing, which can provide fast computation once the model has finished the training. For example, researchers have used a convolutional neural network (CNN) for registration and segmentation of cardiac cine or late gadolinium enhancement images [18][19][20][21][22][23][24][25] and for T 1 /T 2 parameter fitting. 26,27 Most of the methods adopt a supervised learning strategy, which requires the ground truth for training.…”