Automatic segmentation of organs‐at‐risks of nasopharynx cancer and lung cancer by cross‐layer attention fusion network with TELD‐Loss

Abstract

Purpose: Radiotherapy is one of the main treatments of nasopharyngeal cancer (NPC) and lung cancer. Accurate segmentation of organs at risks (OARs) in CT images is a key step in radiotherapy planning for NPC and lung cancer. However, the segmentation of OARs is influenced by the highly imbalanced size of organs, which often results in very poor segmentation results for small and difficult-to-segment organs. In addition, the complex morphological changes and fuzzy boundaries of OARs also pose great challenges to the segmentation task. In this paper, we propose a cross-layer attention fusion network (CLAF-CNN) to solve the problem of accurately segmenting OARs. Methods: In CLAF-CNN, we integrate the spatial attention maps of the adjacent spatial attention modules to make the segmentation targets more accurately focused, so that the network can capture more target-related features. In this way, the spatial attention modules in the network can be learned and optimized together. In addition, we introduce a new Top-K exponential logarithmic Dice loss (TELD-Loss) to solve the imbalance problem in OAR segmentation. The TELD-Loss further introduces a Top-K optimization mechanism based on Dice loss and exponential logarithmic loss, which makes the network pay more attention to small organs and difficult-to-segment organs, so as to enhance the overall performance of the segmentation model. Results: We validated our framework on the OAR segmentation datasets of the head and neck and lung CT images in the StructSeg 2019 challenge. Experiments show that the CLAF-CNN outperforms the state-of -the-art attentionbased segmentation methods in the OAR segmentation task with average Dice coefficient of 79.65% for head and neck OARs and 88.39% for lung OARs. Conclusions: This work provides a new network named CLAF-CNN which contains cross-layer spatial attention map fusion architecture and TELD-Loss for OAR segmentation. Results demonstrated that the proposed method could obtain accurate segmentation results for OARs, which has a potential of improving the efficiency of radiotherapy planning for nasopharynx cancer and lung cancer.

Enhancing Radiotherapy Workflow for Head and Neck Cancer with Artificial Intelligence: A Systematic Review

“…A specific study on small OARs was performed by Liu et al for nasopharyngeal tumors [ 31 ]. The authors validated their model against other state-of-the-art networks on the public StructSeg 2019 challenge dataset, reporting an average DSC = 0.80.…”

“…The current availability of public datasets and scientific challenges, such as the HECKTOR challenge [61], the StructSeg challenge (https://structseg2019.grand-challenge.org (accessed on 4 May 2023)), and the TCGA-HNSC dataset [62], provides an excellent opportunity to benchmark and compare methods. If these datasets and challenges are widely adopted by researchers, as carried out in [18,31], it would be possible to reach higher standards of evidence.…”

Application of FGD-BCEL loss function in segmenting temporal lobes on localized CT images for radiotherapy

“…Liu et al. ( 24 ) proposed a loss function called TELD-loss for automatic segmentation of the OARs for nasopharyngeal and lung cancer, and the results showed that the mean DSC values of the temporal lobes on the left and right side were 0.7873 and 0.5969, respectively. Compared with them, the resultant value in this study was 0.88, which has a sufficient improvement.…”

A Review of Artificial Intelligence Application for Radiotherapy

“…Bladder DSC of 0.94 and eye DSC of 0.91 were reported by Zhou et al 14 The OARs with small volumes and fuzzy boundaries, such as optic chiasma, pose challenges to the segmentation task. To deal with this problem, researchers tried to find solutions by improving their network architecture designs 16 or loss functions. 16 A cross-layer spatial attention map fusion architecture 16 was proposed to enhance the network’s attention to the target area.…”