Automatic segmentation of hippocampus in hippocampal sparing whole brain radiotherapy: A multitask edge‐aware learning

Abstract

This study aimed to improve the accuracy of the hippocampus segmentation through multitask edge-aware learning. Method: We developed a multitask framework for computerized hippocampus segmentation. We used three-dimensional (3D) U-net as our backbone model with two training objectives: (a) to minimize the difference between the targeted binary mask and the model prediction; and (b) to optimize an auxiliary edge-prediction task which is designed to guide the model detection of the weak boundary of the hippocampus in model optimization. To balance the multiple task objectives, we proposed an improved gradient normalization by adaptively adjusting the weight of losses from different tasks. A total of 247 T1-weighted MRIs including 131 without contrast and 116 with contrast were collected from 247 patients to train and validate the proposed method. Segmentation was quantitatively evaluated with the dice coefficient (Dice), Hausdorff distance (HD), and average Hausdorff distance (AVD). The 3D U-net was used for baseline comparison. We used a Wilcoxon signed-rank test to compare repeated measurements (Dice, HD, and AVD) by different segmentations. Results: Through fivefold cross-validation, our multitask edge-aware learning achieved Dice of 0.8483 AE 0.0036, HD of 7.5706 AE 1.2330 mm, and AVD of 0.1522 AE 0.0165 mm, respectively. Conversely, the baseline results were 0.8340 AE 0.0072, 10.4631 AE 2.3736 mm, and 0.1884 AE 0.0286 mm, respectively. With a Wilcoxon signed-rank test, we found that the differences between our method and the baseline were statistically significant (P < 0.05). Conclusion:Our results demonstrated the efficiency of multitask edge-aware learning in hippocampus segmentation for hippocampal sparing whole-brain radiotherapy. The proposed framework may also be useful for other low-contrast small organ segmentations on medical imaging modalities.

CARes‐UNet: Content‐aware residual UNet for lesion segmentation of COVID‐19 from chest CT images

“…The features in the downsampling path were concatenated with the features in the upsampling path by skip connection to provide the added information without the downsampling information abstraction. 28 Finally, the channel number was reduced to 64, and the final segmentation result comes out through a 1 × 1 convolution operator followed by a sigmoid layer to map the result into probability. Technical details of our network and loss function will be illustrated later.…”

“…When the channel number reaches 1024, the feature map with minimal size was upsampled via a content‐aware upsampling operation. The features in the downsampling path were concatenated with the features in the upsampling path by skip connection to provide the added information without the downsampling information abstraction 28 . Finally, the channel number was reduced to 64, and the final segmentation result comes out through a 1 × 1 convolution operator followed by a sigmoid layer to map the result into probability.…”

“…Motivated by Qiu et al., 28 we defined the loss function as a combination of a weighted Dice loss and a weighted binary cross‐entropy loss for each segmentation supervision, that is, where is the weight, and we set it to 1 in our experiments. The is where is Dice coefficient between and , and are the prediction result and the corresponding ground‐truth, respectively.…”

“…Motivated by Qiu et al, 28 we defined the loss function L seg as a combination of a weighted Dice loss L dice and a weighted binary cross-entropy loss L BCE for each segmentation supervision, that is,…”

Blockchain + Radiotherapy: The Scenarios of the 5G‐Enabled Smart Radiotherapy Era

“…Furthermore, in cases where the anatomical structures, such as the centers of growth in the pediatric skeleton 12 and hippocampal tissues, 13 are anatomically difficult to distinguish and conventional methods struggle to outline and evaluate these challenging organs, AI‐assisted delineation can offer more tissue sparing, enhance patient quality of life, and subsequently elevate the benefits of RT.…”