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. 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. Through fivefold cross-validation, our multitask edge-aware learning achieved Dice of 0.8483±0.0036, HD of 7.5706±1.2330mm, and AVD of 0.1522±0.0165mm, respectively. Conversely, the baseline results were 0.8340±0.0072, 10.4631±2.3736mm, and 0.1884±0.0286mm, respectively. With a Wilcoxon signed-rank test, we found that the differences between our method and the baseline were statistically significant (P<0.05). 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.