DLUNet: Semi-supervised Learning based Dual-Light UNet for Multi-organ Segmentation
This work addresses the challenge of reducing annotation effort for medical imaging segmentation, though it appears incremental as it builds upon existing UNet architectures with semi-supervised techniques.
The paper tackles the labor-intensive problem of manual ground truth annotation for abdominal multi-organ segmentation in CT data by developing a semi-supervised learning method called DLUNet, which achieves an average Dice Similarity Coefficient (DSC) of 0.8718 on the validation set.
The manual ground truth of abdominal multi-organ is labor-intensive. In order to make full use of CT data, we developed a semi-supervised learning based dual-light UNet. In the training phase, it consists of two light UNets, which make full use of label and unlabeled data simultaneously by using consistent-based learning. Moreover, separable convolution and residual concatenation was introduced light UNet to reduce the computational cost. Further, a robust segmentation loss was applied to improve the performance. In the inference phase, only a light UNet is used, which required low time cost and less GPU memory utilization. The average DSC of this method in the validation set is 0.8718. The code is available in https://github.com/laihaoran/Semi-SupervisednnUNet.