Automated liver segmental volume ratio quantification on non-contrast T1–Vibe Dixon liver MRI using deep learning

Abstract

Purpose: To evaluate the effectiveness of automated liver segmental volume quantification and calculation of the liver segmental volume ratio (LSVR) on a non-contrast T1-vibe Dixon liver MRI sequence using a deep learning segmentation pipeline. Method: A dataset of 200 liver MRI with a non-contrast 3 mm T1-vibe Dixon sequence was manually labeled slice-by-slice by an expert for Couinaud liver segments, while portal and hepatic veins were labeled separately. A convolutional neural network was trained using 170 liver MRI for training and 30 for evaluation. Liver segmental volumes without liver vessels were retrieved and LSVR was calculated as the liver segmental volumes I-III divided by the liver segmental volumes IV-VIII. LSVR was compared with the expert manual LSVR calculation and the LSVR calculated on CT scans in 30 patients with CT and MRI within 6 months. Results: The convolutional neural network classified the Couinaud segments I-VIII with an average Dice score of 0.770 ± 0.03, ranging between 0.726 ± 0.13 (segment IVb) and 0.810 ± 0.09 (segment V). The calculated mean LSVR with liver MRI unseen by the model was 0.32 ± 0.14, as compared with manually quantified LSVR of 0.33 ± 0.15, resulting in a mean absolute error (MAE) of 0.02. A comparable LSVR of 0.35 ± 0.14 with a MAE of 0.04 resulted with the LSRV retrieved from the CT scans. The automated LSVR showed significant correlation with the manual MRI LSVR (Spearman r = 0.97, p < 0.001) and CT LSVR (Spearman r = 0.95, p < 0.001). Conclusions: A convolutional neural network allowed for accurate automated liver segmental volume quantification and calculation of LSVR based on a non-contrast T1-vibe Dixon sequence.