MCE-Net: polyp segmentation with multiple branch series-parallel attention and channel interaction via edge distribution guidance.

Objective.Accurate polyp segmentation is vital for diagnosing colorectal cancer. However, it is still challenging for accurate polyp segmentation and several bottlenecks exist, such as incomplete boundary, localization bias and lack of micro blocks along with large fragmented boundaries in uncertain regions.Approach.To address the above issues, a novel polyp segmentation network with multiple branch series-parallel attention (MBSA) and channel interaction via edge distribution guidance is proposed. Initially, the edge distribution guidance strategy is proposed to generate the edge distribution following Cauchy distribution to capture complementary edges with sufficient details. Subsequently, a MBSA module is put forward to extract features from various receptive fields to pinpoint tiny polyps by a multiple kernel dilated convolution block, while combining semantics of different dimensions to filter out noise and refining the details of micro target. Ultimately, the channel interaction model is proposed to improve the segmentation accuracy of the polyps in uncertain area by splitting channels into groups and conducts group-wise interaction to excavate subtle clues contained in different channels.Main results.Extensive experimental results demonstrate that the proposed method is superior over the state-of-the-art methods with the mean dice of 0.8972, 0.9420, 0.8312, 0.8064 and 0.9214 on five public polyp datasets.Significance.The proposed method improves the integrity of the margins and internal details for polyp segmentation, which will provide a powerful aid for doctors to achieve accurate judgments, reducing the likelihood of colorectal cancer and improving the survival chances of patients.

Resampling-based cost loss attention network for explainable imbalanced diabetic retinopathy grading.

Diabetic retinopathy (DR) is currently considered to be one of the most common diseases that cause blindness. However, DR grading methods are still challenged by the presence of imbalanced class distributions, small lesions, low accuracy of small sample classes and poor explainability. To address these issues, a resampling-based cost loss attention network for explainable imbalanced diabetic retinopathy grading is proposed. First, the progressively-balanced resampling strategy is put forward to create a balanced training data by mixing the two sets of samples obtained from instance-based sampling and class-based sampling. Subsequently, a neuron and normalized channel-spatial attention module (Neu-NCSAM) is designed to learn the global features with 3-D weights and a weight sparsity penalty is applied to the attention module to suppress irrelevant channels or pixels, thereby capturing detailed small lesion information. Thereafter, a weighted loss function of the Cost-Sensitive (CS) regularization and Gaussian label smoothing loss, called cost loss, is proposed to intelligently penalize the incorrect predictions and thus to improve the grading accuracy of small sample classes. Finally, the Gradient-weighted Class Activation Mapping (Grad-CAM) is performed to acquire the localization map of the questionable lesions in order to visually interpret and understand the effect of our model. Comprehensive experiments are carried out on two public datasets, and the subjective and objective results demonstrate that the proposed network outperforms the state-of-the-art methods and achieves the best DR grading results with 83.46%, 60.44%, 65.18%, 63.69% and 92.26% for Kappa, BACC, MCC, F1 and mAUC, respectively.