AugMS-Net:Augmented multiscale network for small cervical tumor segmentation from MRI volumes. (February 2022)

• Record Type: Journal Article
• Authors: Lu, Pengyue; Fang, Faming; Zhang, He; Ling, Lei; Hua, Keqin
• Is Part Of: Computers in biology and medicine. Volume 141(2022)
• Publication Date: 2022-02
• Subjects: Semantic segmentation -- Augmented multiscale -- Deep learning -- Biomedical image; Medicine -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; 610.285
• DOI: 10.1016/j.compbiomed.2021.104774
• Languages: English
• ISSNs: 0010-4825

Abstract

Abstract: Cervical cancer is one of the leading causes of female-specific cancer death. Tumor region segmentation plays a pivotal role in both the clinical analysis and treatment planning of cervical cancer. Due to the heterogeneity and low contrast of biomedical images, current state-of-the-art tumor segmentation approaches are facing the challenge of the insensitive detection of small lesion regions. To tackle this problem, this paper proposes an augmented multiscale network (AugMS-Net) based on 3D U-Net to automatically segment cervical Magnetic Resonance Imaging (MRI) volumes. Since a multiscale strategy is considered one of the most promising algorithms to tackle small object recognition, we introduce a novel 3D module to explore more granular multiscale representations. Besides, we employ a deep multiscale supervision strategy to doubly supervise the side outputs hierarchically. To demonstrate the generalization of our model, we evaluated AugMS-Net on both a cervical dataset from MRI volumes and a liver dataset from Computerized Tomography (CT) volumes. Our proposed AugMS-Net shows superior performance over baseline models, yielding high accuracy while reducing the number of model parameters by nearly 20%. The source code and trained models are available at https://github.com/Cassieyy/AugMS-Net . Highlights: We propose the augmented multiscale network (AugMS-Net) to address the small tumors recognition problem. We introduce the 3D Res2Net module in our 3D network to enhance the model's ability to capture powerful representations. We employ a deep multiscale supervision scheme for the hidden layers to support the prediction accuracy. The experiments show that AugMS-Net achieves modest yet consistent performance gains while reducing about 20% parameters.