Automated CT and MRI Liver Segmentation and Biometry Using a Generalized Convolutional Neural Network

Abstract

To assess feasibility of training a convolutional neural network (CNN) to automate liver segmentation across different imaging modalities and techniques used in clinical practice and to apply this technique to enable automation of liver biometry. Materials and Methods: A two-dimensional U-Net CNN was trained for liver segmentation in two stages by using 330 abdominal MRI and CT examinations. First, the neural network was trained with unenhanced multiecho spoiled gradient-echo images from 300 MRI examinations to yield multiple signal weightings. Then, transfer learning was used to generalize the CNN with additional images from 30 contrast material-enhanced MRI and CT examinations. Performance of the CNN was assessed by using a distinct multiinstitutional dataset curated from multiple sources (498 subjects). Segmentation accuracy was evaluated by computing Dice scores. These segmentations were used to compute liver volume from CT and T1-weighted MRI examinations and to estimate hepatic proton density fat fraction (PDFF) from multiecho T2*-weighted MRI examinations. Quantitative volumetry and PDFF estimates were compared between automated and manual segmentation by using Pearson correlation and Bland-Altman statistics. Results: Dice scores were 0.94 ± 0.06 for CT (n = 230), 0.95 ± 0.03 (n = 100) for T1-weighted MRI, and 0.92 ± 0.05 for T2*weighted MRI (n = 168). Liver volume measured with manual and automated segmentation agreed closely for CT (95% limits of agreement: −298 mL, 180 mL) and T1-weighted MRI (95% limits of agreement: −358 mL, 180 mL). Hepatic PDFF measured by the two segmentations also agreed closely (95% limits of agreement: −0.62%, 0.80%). Conclusion: By using a transfer-learning strategy, this study has demonstrated the feasibility of a CNN to be generalized to perform liver segmentation across different imaging techniques and modalities. With further refinement and validation, CNNs may have broad applicability for multimodal liver volumetry and hepatic tissue characterization.

Deep Learning Algorithm for Automated Segmentation and Volume Measurement of the Liver and Spleen Using Portal Venous Phase Computed Tomography Images

“…Despite this inherent limitation, we consider the two-dimensional volume measurement to be a practical method since processing three-dimensional volume data requires more time, higher computational capacity, and larger data storage capacity and may potentially increase the need for operators' correction in cases of inaccurate segmentation results in contrast to our approach. Thus, most of the previous studies involving volume measurement of the liver and spleen have utilized two-dimensional CT images (18,(29)(30)(31)(32).…”

“…First, our algorithm was developed and validated using portal venous phase CT images. The application of our algorithm to other CT images may require further training of the algorithm using additional training data through transfer learning (18,26). Second, our algorithm only provides whole liver segmentation.…”

Radiomics and Deep Learning: Hepatic Applications

“…Liver segmentation has direct clinical applications, including liver volume measurement, which is important in pre-operative planning for liver resection (46,47), determination of the radiation dose in liver tumor radioembolization, and measurement of quantitative indices such as the proton density fat fraction (PDFF) from the whole liver (48). Notably, however, liver segmentation is labor-intensive and time-consuming, which limits its usage in clinical practice.…”

“…All of these studies reported high performance values in liver segmentation, with the reported DSS values ranging from 0.92 to 0.95 (50)(51)(52)(53)(54). Recently, Wang et al (48) demonstrated the feasibility of generalized CNN, which can be used for liver segmentation on CT scans and various MRI sequences using the transfer learning technique. They reported DSS values ranging from 0.92 to 0.95 for liver segmentation on CT and MR images.…”

Liver Segmentation from Multimodal Images Using HED-Mask R-CNN

“…Automatic medical image segmentation approaches that are introduced in the last two decades have been the most successful methods for medical image analysis. The feasibility of a CNN to be generalized to perform liver segmentation across various imaging strategies and modalities is used in [9]. Patrick et al [2] presented a method to automatically segment liver and lesions in CT and MRI abdomen images using cascaded fully convolutional neural networks (CFCNs) enabling the segmentation of large-scale medical trials and quantitative image analysis.…”