Xikai Tang1,2, Esmaeel Jafargholi Rangraz3,4, Walter Coudyzer5, Jeroen Bertels4,6, David Robben4,6, Georg Schramm3,4, Wies Deckers7, Geert Maleux5,8, Kristof Baete3,7, Chris Verslype9, Mark J Gooding10, Christophe M Deroose3,7, Johan Nuyts3,4. 1. Nuclear Medicine and Molecular Imaging, KU Leuven, Leuven, Belgium. xikai.tang@kuleuven.be. 2. Medical Imaging Research Center (MIRC), KU Leuven, UZ Herestraat 49 - box 7003, 3000, Leuven, Belgium. xikai.tang@kuleuven.be. 3. Nuclear Medicine and Molecular Imaging, KU Leuven, Leuven, Belgium. 4. Medical Imaging Research Center (MIRC), KU Leuven, UZ Herestraat 49 - box 7003, 3000, Leuven, Belgium. 5. Radiology, University Hospitals Leuven, Leuven, Belgium. 6. Medical Image Computing (ESAT/PSI), KU Leuven, Leuven, Belgium. 7. Nuclear Medicine, University Hospitals Leuven, Leuven, Belgium. 8. Radiology, KU Leuven, Leuven, Belgium. 9. Digestive Oncology, University Hospitals Leuven, Leuven, Belgium. 10. Mirada Medical Ltd, Oxford, UK.
Abstract
PURPOSE: In selective internal radiation therapy (SIRT), an accurate total liver segmentation is required for activity prescription and absorbed dose calculation. Our goal was to investigate the feasibility of using automatic liver segmentation based on a convolutional neural network (CNN) for CT imaging in SIRT, and the ability of CNN to reduce inter-observer variability of the segmentation. METHODS: A multi-scale CNN was modified for liver segmentation for SIRT patients. The CNN model was trained with 139 datasets from three liver segmentation challenges and 12 SIRT patient datasets from our hospital. Validation was performed on 13 SIRT datasets and 12 challenge datasets. The model was tested on 40 SIRT datasets. One expert manually delineated the livers and adjusted the liver segmentations from CNN for 40 test SIRT datasets. Another expert performed the same tasks for 20 datasets randomly selected from the 40 SIRT datasets. The CNN segmentations were compared with the manual and adjusted segmentations from the experts. The difference between the manual segmentations was compared with the difference between the adjusted segmentations to investigate the inter-observer variability. Segmentation difference was evaluated through dice similarity coefficient (DSC), volume ratio (RV), mean surface distance (MSD), and Hausdorff distance (HD). RESULTS: The CNN segmentation achieved a median DSC of 0.94 with the manual segmentation and of 0.98 with the manually corrected CNN segmentation, respectively. The DSC between the adjusted segmentations is 0.98, which is 0.04 higher than the DSC between the manual segmentations. CONCLUSION: The CNN model achieved good liver segmentations on CT images of good image quality, with relatively normal liver shapes and low tumor burden. 87.5% of the 40 CNN segmentations only needed slight adjustments for clinical use. However, the trained model failed on SIRT data with low dose or contrast, lesions with large density difference from their surroundings, and abnormal liver position and shape. The abovementioned scenarios were not adequately represented in the training data. Despite this limitation, the current CNN is already a useful clinical tool which improves inter-observer agreement and therefore contributes to the standardization of the dosimetry. A further improvement is expected when the CNN will be trained with more data from SIRT patients.
PURPOSE: In selective internal radiation therapy (SIRT), an accurate total liver segmentation is required for activity prescription and absorbed dose calculation. Our goal was to investigate the feasibility of using automatic liver segmentation based on a convolutional neural network (CNN) for CT imaging in SIRT, and the ability of CNN to reduce inter-observer variability of the segmentation. METHODS: A multi-scale CNN was modified for liver segmentation for SIRT patients. The CNN model was trained with 139 datasets from three liver segmentation challenges and 12 SIRT patient datasets from our hospital. Validation was performed on 13 SIRT datasets and 12 challenge datasets. The model was tested on 40 SIRT datasets. One expert manually delineated the livers and adjusted the liver segmentations from CNN for 40 test SIRT datasets. Another expert performed the same tasks for 20 datasets randomly selected from the 40 SIRT datasets. The CNN segmentations were compared with the manual and adjusted segmentations from the experts. The difference between the manual segmentations was compared with the difference between the adjusted segmentations to investigate the inter-observer variability. Segmentation difference was evaluated through dice similarity coefficient (DSC), volume ratio (RV), mean surface distance (MSD), and Hausdorff distance (HD). RESULTS: The CNN segmentation achieved a median DSC of 0.94 with the manual segmentation and of 0.98 with the manually corrected CNN segmentation, respectively. The DSC between the adjusted segmentations is 0.98, which is 0.04 higher than the DSC between the manual segmentations. CONCLUSION: The CNN model achieved good liver segmentations on CT images of good image quality, with relatively normal liver shapes and low tumor burden. 87.5% of the 40 CNN segmentations only needed slight adjustments for clinical use. However, the trained model failed on SIRT data with low dose or contrast, lesions with large density difference from their surroundings, and abnormal liver position and shape. The abovementioned scenarios were not adequately represented in the training data. Despite this limitation, the current CNN is already a useful clinical tool which improves inter-observer agreement and therefore contributes to the standardization of the dosimetry. A further improvement is expected when the CNN will be trained with more data from SIRT patients.
Authors: Moritz Gross; Michael Spektor; Ariel Jaffe; Ahmet S Kucukkaya; Simon Iseke; Stefan P Haider; Mario Strazzabosco; Julius Chapiro; John A Onofrey Journal: PLoS One Date: 2021-12-01 Impact factor: 3.240
Authors: Mi Jin Yun; Dong Young Lee; Yong Jeong; Suhong Kim; Peter Lee; Kyeong Taek Oh; Min Soo Byun; Dahyun Yi; Jun Ho Lee; Yu Kyeong Kim; Byoung Seok Ye Journal: EJNMMI Res Date: 2021-06-10 Impact factor: 3.138
Authors: Elisabeth von Brandis; Håvard B Jenssen; Derk F M Avenarius; Atle Bjørnerud; Berit Flatø; Anders H Tomterstad; Vibke Lilleby; Karen Rosendahl; Tomas Sakinis; Pia K K Zadig; Lil-Sofie Ording Müller Journal: Pediatr Radiol Date: 2022-02-02