Korean J Radiol.
2020 Aug;21(8):987-997. 10.3348/kjr.2020.0237.
Deep Learning Algorithm for Automated Segmentationand Volume Measurement of the Liver and Spleen UsingPortal Venous Phase Computed Tomography Images
- Affiliations
-
- 1Department of Radiology and Research Institute of Radiology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea
- 2Departments of Brain and Cognitive Engineering, Korea University, Seoul, Korea
- 3Departments of Artificial Intelligence, Korea University, Seoul, Korea
- 4Department of Radiology, Haeundae Paik Hospital, Inje University College of Medicine, Busan, Korea
- 5Department of Radiology, Hanyang University Medical Center, Hanyang University School of Medicine, Seoul, Korea
- KMID: 2505137
- DOI: http://doi.org/10.3348/kjr.2020.0237
Abstract
Objective
Measurement of the liver and spleen volumes has clinical implications. Although computed tomography (CT)volumetry is considered to be the most reliable noninvasive method for liver and spleen volume measurement, it has limitedapplication in clinical practice due to its time-consuming segmentation process. We aimed to develop and validate a deeplearning algorithm (DLA) for fully automated liver and spleen segmentation using portal venous phase CT images in variousliver conditions.
Materials and Methods
A DLA for liver and spleen segmentation was trained using a development dataset of portal venousCT images from 813 patients. Performance of the DLA was evaluated in two separate test datasets: dataset-1 which included150 CT examinations in patients with various liver conditions (i.e., healthy liver, fatty liver, chronic liver disease, cirrhosis,and post-hepatectomy) and dataset-2 which included 50 pairs of CT examinations performed at ours and other institutions.The performance of the DLA was evaluated using the dice similarity score (DSS) for segmentation and Bland-Altman 95%limits of agreement (LOA) for measurement of the volumetric indices, which was compared with that of ground truth manualsegmentation.
Results
In test dataset-1, the DLA achieved a mean DSS of 0.973 and 0.974 for liver and spleen segmentation, respectively,with no significant difference in DSS across different liver conditions (p = 0.60 and 0.26 for the liver and spleen, respectively).For the measurement of volumetric indices, the Bland-Altman 95% LOA was -0.17 ± 3.07% for liver volume and -0.56 ± 3.78%for spleen volume. In test dataset-2, DLA performance using CT images obtained at outside institutions and our institutionwas comparable for liver (DSS, 0.982 vs. 0.983; p = 0.28) and spleen (DSS, 0.969 vs. 0.968; p = 0.41) segmentation.
Conclusion
The DLA enabled highly accurate segmentation and volume measurement of the liver and spleen using portalvenous phase CT images of patients with various liver conditions.
Keyword
- Full Text Links
-
- Actions
-
Cited
- CITED
-
- Close
- Share
-
- Similar articles
-
- Deep Learning-Based Computed Tomography Image Standardization to Improve Generalizability of Deep Learning-Based Hepatic Segmentation
- Prognostic role of computed tomography analysis using deep learning algorithm in patients with chronic hepatitis B viral infection
- Deep Learning-Based Assessment of Functional Liver Capacity Using Gadoxetic Acid-Enhanced Hepatobiliary Phase MRI
- Spiral CT for the Detection of Metastatic Tumor of the Liver: Relative Value of Arterial, Portal Venous and Delayed Phase Scanning
- Evaluation of Automatic Image Segmentation for 3D Volume Measurement of Liver and Spleen Based on 3D Region-growing Algorithm using Animal Phantom
