The detection and resection of small polyp objects in colonoscopy images is of great significance for the prevention of colorectal cancer. At present, blurred edges, variable lesion shapes, and intra-class dissimilarity pose challenges for accurately segmenting small polyp objects. In recent years, many deep learning methods based on convolutional neural networks (CNNs) have been proposed and successfully applied to polyp segmentation tasks. However, these methods still have two limitations: (1) Limited ability to mine boundary detail information, (2) Insufficient ability to capture rich global context information, and (3) Introduced additional complex feature extraction operations. To alleviate these challenges, we propose a Fourier transform-multiscale feature fusion network (FTMF-Net) for segmentation of small polyp objects. The core idea includes two points: (1) Fourier transform module extracts more detailed boundary information, and (2) Multiscale feature fusion module enriches global semantic feature information. FTMF-Net mainly has the following advantages: (1) The proposed model has excellent performance for small polyp object segmentation, (2) This method greatly reduces the complexity of the model without significantly increasing the number of network parameters, and (3) The network is relatively simple and easy to understand. Extensive experiments with eleven state-of-the-art (SOTA) methods on five small polyp object datasets show that our proposed FTMF-Net has superior segmentation performance.

Breast lesion segmentation in ultrasound images is of great importance since it can help us to characterize and localize lesion regions. However, low-quality imaging, blurred boundary, and variable lesion shapes bring challenges to accurate segmentation. In recent years, many U-Net variants have been proposed and successfully applied to breast lesion segmentation. However, these methods suffer from two limitations: (1) Ignoring the ability to capture rich global context information, and (2) Introducing extra complex operations. To alleviate these challenges, we propose a multiscale nonlocal feature extraction network (MNFE-Net) for accurately segmenting breast lesions. The core idea includes three points: (1) Parallel Encoder models long-range dependencies, (2) Multiscale Feature Module refines local features without introducing extra complex operations, and (3) Global Feature Guidance Module extracts global semantic information. MNFE-Net mainly has the following advantages: (1) The method has excellent performance for segmentation of malignant breast lesions, (2) The Parallel Encoder increases network parameters without significantly decreasing inference speed, and (3) The method is easy to understand and execute. Extensive experiment results with six state-of-the-art (SOTA) methods on three public breast ultrasound datasets demonstrate the superior segmentation performance of our proposed MNFE-Net.