scholarly journals Generalised Wasserstein Dice Score for Imbalanced Multi-class Segmentation Using Holistic Convolutional Networks

2018 ◽  
pp. 64-76 ◽  
Author(s):  
Lucas Fidon ◽  
Wenqi Li ◽  
Luis C. Garcia-Peraza-Herrera ◽  
Jinendra Ekanayake ◽  
Neil Kitchen ◽  
...  
Keyword(s):  
Convolutional Networks ◽  
Class Segmentation

scholarly journals Deep Learning based Segmentation of Lumbar Vertebrae from CT Images

10.29007/vt7v ◽  
2018 ◽  
Author(s):  
Rens Janssens ◽  
Guoyan Zheng
Keyword(s):  
Automatic Segmentation ◽  
Lumbar Vertebrae ◽  
Ct Images ◽  
Lumbar Region ◽  
Dice Coefficient ◽  
Convolutional Networks ◽  
Surface Distance ◽  
Fully Convolutional Networks ◽  
Symmetric Surface ◽  
Class Segmentation

We present a method to address the challenging problem of automatic segmentation of lumbar vertebrae from CT images acquired with varying fields of view. Our method is based on cascaded 3D Fully Convolutional Networks (FCNs) consisting of a localization FCN and a segmentation FCN. More specifically, in the first step we train a regression 3D FCN (we call it “LocalizationNet”) to find the bounding box of the lumbar region. After that, a 3D U-net like FCN (we call it “SegmentationNet”) is then developed, which after training, can perform a pixel-wise multi-class segmentation to map a cropped lumber region volumetric data to its volume-wise labels. Evaluated on publicly available datasets, our method achieved an average Dice coefficient of 95.77 ± 0.81% and an average symmetric surface distance of 0.37 ± 0.06 mm.

scholarly journals HIFUNet: Multi-Class Segmentation of Uterine Regions From MR Images Using Global Convolutional Networks for HIFU Surgery Planning

2020 ◽  
Vol 39 (11) ◽  
pp. 3309-3320
Author(s):  
Chen Zhang ◽  
Huazhong Shu ◽  
Guanyu Yang ◽  
Faqi Li ◽  
Yingang Wen ◽  
...  
Keyword(s):  
Mr Images ◽  
Surgery Planning ◽  
Convolutional Networks ◽  
Class Segmentation

scholarly journals Loss Weightings for Improving Imbalanced Brain Structure Segmentation Using Fully Convolutional Networks

Healthcare ◽  
2021 ◽  
Vol 9 (8) ◽  
pp. 938
Author(s):  
Takaaki Sugino ◽  
Toshihiro Kawase ◽  
Shinya Onogi ◽  
Taichi Kin ◽  
Nobuhito Saito ◽  
...  
Keyword(s):  
Brain Structure ◽  
Class Imbalance ◽  
Cross Entropy ◽  
Loss Functions ◽  
Mr Images ◽  
Distance Map ◽  
Convolutional Networks ◽  
Fully Convolutional Networks ◽  
Frequency Weighting ◽  
Class Segmentation

Brain structure segmentation on magnetic resonance (MR) images is important for various clinical applications. It has been automatically performed by using fully convolutional networks. However, it suffers from the class imbalance problem. To address this problem, we investigated how loss weighting strategies work for brain structure segmentation tasks with different class imbalance situations on MR images. In this study, we adopted segmentation tasks of the cerebrum, cerebellum, brainstem, and blood vessels from MR cisternography and angiography images as the target segmentation tasks. We used a U-net architecture with cross-entropy and Dice loss functions as a baseline and evaluated the effect of the following loss weighting strategies: inverse frequency weighting, median inverse frequency weighting, focal weighting, distance map-based weighting, and distance penalty term-based weighting. In the experiments, the Dice loss function with focal weighting showed the best performance and had a high average Dice score of 92.8% in the binary-class segmentation tasks, while the cross-entropy loss functions with distance map-based weighting achieved the Dice score of up to 93.1% in the multi-class segmentation tasks. The results suggested that the distance map-based and the focal weightings could boost the performance of cross-entropy and Dice loss functions in class imbalanced segmentation tasks, respectively.

Sign in / Sign up

Export Citation Format

Share Document