DSCA-Net: A depthwise separable convolutional neural network with attention mechanism for medical image segmentation.

Explainable multi-module semantic guided attention based network for medical image segmentation

Accurate image segmentation is crucial in medical imaging for quantifying diseases, assessing prognosis, and evaluating treatment outcomes. However, existing methods often fall short in integrating global and local features in a meaningful way, failing to give sufficient attention to abnormal regions and boundary details in medical images. These limitations hinder the effectiveness of segmentation techniques in clinical settings. To address these issues, we propose a novel deep learning-based approach, MIPC-Net, designed for precise boundary segmentation in medical images. Our approach, inspired by radiologists' working patterns, introduces two distinct modules: 1. Mutual Inclusion of Position and Channel Attention (MIPC) Module: To improve boundary segmentation precision, we present the MIPC module. This module enhances the focus on channel information while extracting position features and vice versa, effectively enhancing the segmentation of boundaries in medical images. 2. Skip-Residue Module: To optimize the restoration of medical images, we introduce Skip-Residue, a global residual connection. This module improves the integration of the encoder and decoder by filtering out irrelevant information and recovering the most crucial information lost during the feature extraction process. We evaluate the performance of MIPC-Net on three publicly accessible datasets: Synapse, ISIC2018-Task, and Segpc. The evaluation uses metrics such as the Dice coefficient (DSC) and Hausdorff Distance (HD). Our ablation study confirms that each module contributes to the overall improvement of segmentation quality. Notably, with the integration of both modules, our model outperforms state-of-the-art methods across all metrics. Specifically, MIPC-Net achieves a 2.23 mm reduction in Hausdorff Distance on the Synapse dataset, highlighting the model's enhanced capability for precise image boundary segmentation. The introduction of the novel MIPC and Skip-Residue modules significantly improves feature extraction accuracy, leading to better boundary recognition in medical image segmentation tasks. Our approach demonstrates substantial improvements over existing methods, as evidenced by the results on benchmark datasets.

Medical image segmentation relying on deep learning is important for assisting doctors in treating diseases. In this paper, we propose a DSPM-Net structure for medical image segmentation based on UNeXt which combines the advantages of CNN and MLP. Concretely, our approach creatively designs three components, named Squeeze and Excitation-Deep Understanding Convolutional Kernel (SE-DUCK), Attention Module, and Group Aggregation Feature Fusion Attention Module (GAFA). An encoder with a dual stream pyramid structure is constructed to deeply learn contextual information at different scales by combining SE-DUCK and Attention Module. Equally importantly, our proposed GAFA module effectively fuses high-level and low-level features for accurately segmenting images. We test the segmentation performance on two different medical image datasets. The results demonstrate our method could obtain optimal segmentation performance.

Medical image segmentation is a bottleneck for physicians and radiologists in diagnosis of diseases. Deep learning based convolutional neural networks (CNNs) is used to support decision making in medical diagnosis. Three architectures are analyzed for segmentation of affected tissues. CNN with contraction path classifies the each pixel in the image into region of interest and region of non interest. The affected region is then segmented from RoI. Second architecture is developed with contraction and expansion paths (auto-encoders) to extract the spatial information of the pixels from the input image. The deconvolutional layer extracts the spatial information related to corresponding features but still fails to capture contextual dependent information of high-level features. In the third architecture, attention module with U-Net captures the contextual dependent information. Filter size, learning rate and k-fold cross validation are tuned to improve the accuracy and dice similarity coefficient (DSC). Filter size and k-fold cross validation are varied as 3x3, 5x5 and 7x7 and 3-fold, 5-fold and 10-fold respectively. The attention module helps to extract the spatial information of the high level features which are related to low level features. This gives better segmentation output. U-Net with attention module provides an accuracy 99.07 %, sensitivity 98.57 %, specificity 99.5 % and DSC 91.7%.

Segmenting medical images via explicit–implicit attention aggregation

Linear attention has gained popularity in recent years due to its lower computational complexity compared to SoftMax attention. However, its relatively lower performance has limited its widespread application. To address this issue, we propose a plug-and-play module called Wavelet-Enhanced Linear Attention Mechanism (WLAM), which integrates Discrete Wavelet Transform (DWT) with linear attention. This approach enhances the model's ability to express global contextual information while improving the capture of local features.Firstly, we introduce DWT into the attention mechanism to decompose the input features. The original input features are utilized to generate the query vector Q, while the low-frequency coefficients are used to generate the key K. The high-frequency coefficients undergo convolution to produce the value V. This method effectively embeds global and local information into different components of the attention mechanism, thereby enhancing the model's perception of details and overall structure.Secondly, we perform multi-scale convolution on the high-frequency wavelet coefficients and incorporate a Squeeze-and-Excitation (SE) module to enhance feature selectivity. Subsequently, we utilize the Inverse Discrete Wavelet Transform (IDWT) to reintegrate the multi-scale processed information back into the spatial domain, addressing the limitations of linear attention in handling multi-scale and local information.Finally, inspired by certain structures of the Mamba network, we introduce a forget gate and an improved block design into the linear attention framework, inheriting the core advantages of the Mamba architecture. Following a similar rationale, we leverage the lossless downsampling property of wavelet transforms to combine the downsampling module with the attention module, resulting in the Wavelet Downsampling Attention (WDSA) module. This integration reduces the network size and computational load while mitigating information loss associated with downsampling.We apply Wavelet-Enhanced Linear Attention Mechanism (WLAM) to classical networks such as PVT, Swin, and CSwin, achieving significant improvements in performance on image classification tasks. Furthermore, we combine Wavelet Linear Attention with the Wavelet Downsampling Attention (WDSA) module to construct WDLMFormer, which achieves an accuracy of 84.2% on the ImageNet-1K dataset.

CLCU-Net: Cross-level connected U-shaped network with selective feature aggregation attention module for brain tumor segmentation

In recent years, attention mechanisms have shown great potential in various computer vision tasks. However, most existing methods focus on developing more complex attention modules for better performance, which inevitably increases the complexity of the model. To overcome performance and complexity tradeoffs, this paper proposes efficient spatial and channel attention (ESCA), a symmetric, comprehensive, and efficient attention module. By analyzing squeeze-and-excitation (SE), convolutional block attention module (CBAM), coordinate attention (CA), and efficient channel attention (ECA) modules, we abandon the dimension-reduction operation of SE module, verify the negative impact of global max pooling (GMP) on the model, and apply a local cross-channel interaction strategy without dimension reduction to learn attention. We not only care about the channel features of the image, we also care about the spatial location of the target on the image, and we take into account the effectiveness of channel attention, so we designed the symmetric ESCA module. The ESCA module is effective, as demonstrated by its application in the ResNet-50 classification benchmark. With 26.26 M parameters and 8.545 G FLOPs, it introduces a mere 0.14% increment in FLOPs while achieving over 6.33% improvement in Top-1 accuracy and exceeding 3.25% gain in Top-5 accuracy. We perform image classification and object detection tasks on ResNet, MobileNet, YOLO, and other architectures on popular datasets such as Mini ImageNet, CIFAR-10, and VOC 2007. Experiments show that ESCA can achieve great improvement in model accuracy at a very small cost, and it performs well among similar models.

The dynamic unsignalized intersection environment has long been one of the most challenging scenarios for autonomous driving tasks. The crux of the research lies in how to safely and efficiently complete decision-making tasks in such uncertain scenarios. In order to solve the above problems, this paper proposes a decision-making strategy based on risk anticipation and attention mechanisms. The strategy is mainly composed of risk anticipation and attention module. In the risk anticipation module, we consider the distribution of vehicle shapes and motion risk diffusion areas, incorporating these factors into safety considerations, and propose a new risk anticipation index based on the risk anticipation theory, the risk anticipation index is composed of the risk values of road points along the vehicle’s route. Using this risk diffusion model, we can effectively calculate the degree of impact that surrounding risk diffusion zones have on the vehicle itself, thereby effectively preventing dangerous driving behaviors from occurring. In the attention module, we introduce an attention mechanism to enhance the strategy’s ability to extract scene information, improving the vehicle’s understanding of the current scene and allowing it to more effectively focus on potential risks. In practical operation, the attention mechanism and risk prediction module form a tightly integrated system. Based on the road point information along the current vehicle’s route, the risk anticipation module outputs a risk sequence indicator. This sequence serves as the risk anticipation metric and, together with other information, constitutes the current vehicle’s state. The risk sequence indicators output by the risk expectation module serve as network inputs, functioning as the basis for weight adjustment in the attention mechanism. When high-risk scenarios are detected, the attention mechanism dynamically adjusts the weights of each attention head, allocating more resources to feature extraction related to high-risk factors. Finally, the performance of our strategy is validated in an unsignalized intersection scenario using the simulation software CARLA. The proposed method achieves a 12%–35% reduction in collision rate, a 5%–20% decrease in average speed, and a 10%–17% increase in average reward compared to the baseline model for the left-turn task at unsignalized intersections. Although average speed slightly decreased, core safety performance significantly improved. Convergence characteristics also showed superiority: compared to other baseline algorithms, convergence began gradually after training for 400 episodes, with stable performance maintained post-convergence. This demonstrates that the proposed DA-SAC algorithm effectively reduces decision risks and enhances traffic safety.

Ultrasound imaging can distinctly display the morphology and structure of internal organs within the human body, enabling the examination of organs like the breast, liver, and thyroid. It can identify the locations of tumors, nodules, and other lesions, thereby serving as an efficacious tool for treatment detection and rehabilitation evaluation. Typically, the attending physician is required to manually demarcate the boundaries of lesion locations, such as tumors, in ultrasound images. Nevertheless, several issues exist. The high noise level in ultrasound images, the degradation of image quality due to the impact of surrounding tissues, and the influence of the operator’s experience and proficiency on the determination of lesion locations can all contribute to a reduction in the accuracy of delineating the boundaries of lesion sites. In the wake of the advancement of deep learning, its application in medical image segmentation is becoming increasingly prevalent. For instance, while the U-Net model has demonstrated a favorable performance in medical image segmentation, the convolution layers of the traditional U-Net model are relatively simplistic, leading to suboptimal extraction of global information. Moreover, due to the significant noise present in ultrasound images, the model is prone to interference. In this research, we propose an Attention Residual Network model (ARU-Net). By incorporating residual connections within the encoder section, the learning capacity of the model is enhanced. Additionally, a spatial hybrid convolution module is integrated to augment the model’s ability to extract global information and deepen the vertical architecture of the network. During the feature fusion stage of the skip connections, a channel attention mechanism and a multi-convolutional self-attention mechanism are respectively introduced to suppress noisy points within the fused feature maps, enabling the model to acquire more information regarding the target region. Finally, the predictive efficacy of the model was evaluated using publicly accessible breast ultrasound and thyroid ultrasound data. The ARU-Net achieved mean Intersection over Union (mIoU) values of 82.59% and 84.88%, accuracy values of 97.53% and 96.09%, and F1-score values of 90.06% and 89.7% for breast and thyroid ultrasound, respectively.

Segmentation is a fundamental task in medical image analysis. Current state-of-the-art Convolutional Neural Networks on medical image segmentation capture local context information using fixed-shape receptive fields and feature detectors with position-invariant weights, which limits the robustness to the variance of input, such as medical objects of variant sizes, shapes, and domains. In order to capture global context information, we propose High-order Attention (HA), a novel attention module with adaptive receptive fields and dynamic weights. HA allows each pixel to has its own global attention map that models its relationship to all other pixels. In particular, HA constructs the attention map through graph transduction and thus captures high relevant context information at high-order. Consequently, feature maps at each position are selectively aggregated as a weighted sum of feature maps at all positions. We further embed the proposed HA module into an efficient encoder-decoder structure for medical image segmentation, namely High-order Attention Network (HANet). Extensive experiments are conducted on four benchmark sets for three tasks, i.e., REFUGE and Drishti-GS1 for optic disc/cup segmentation, DRIVE for blood vessel segmentation, and LUNA for lung segmentation. The results justify the effectiveness of the new attention module for medical image segmentation.

LCP-Net: A local context-perception deep neural network for medical image segmentation

Accurate segmentation of organs or lesions from medical images is crucial for reliable diagnosis of diseases and organ morphometry. In recent years, convolutional encoder-decoder solutions have achieved substantial progress in the field of automatic medical image segmentation. Due to the inherent bias in the convolution operations, prior models mainly focus on local visual cues formed by the neighboring pixels, but fail to fully model the long-range contextual dependencies. In this article, we propose a novel Transformer-based Attention Guided Network called <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">TransAttUnet</i> , in which the multi-level guided attention and multi-scale skip connection are designed to jointly enhance the performance of the semantical segmentation architecture. Inspired by Transformer, the self-aware attention (SAA) module with Transformer Self Attention (TSA) and Global Spatial Attention (GSA) is incorporated into TransAttUnet to effectively learn the non-local interactions among encoder features. Moreover, we also use additional multi-scale skip connections between decoder blocks to aggregate the upsampled features with different semantic scales. In this way, the representation ability of multi-scale context information is strengthened to generate discriminative features. Benefitting from these complementary components, the proposed TransAttUnet can effectively alleviate the loss of fine details caused by the stacking of convolution layers and the consecutive sampling operations, finally improving the segmentation quality of medical images. Extensive experiments were conducted on multiple medical image segmentation datasets from various imaging modalities, which demonstrate that the proposed method consistently outperforms the existing state-of-the-art methods.

Accurate segmentation of medical images is crucial for disease diagnosis and understanding disease changes. Deep learning methods, utilizing encoder-decoder structures, have demonstrated cutting-edge performance in various medical image segmentation tasks. However, the pooling operation in the encoding stage results in feature loss, which makes the network lack the ability to fuse multi-scale information at different levels, hinders its effective perception of multi-scale information, and leads to poor segmentation performance. Drawing inspiration from the U-shaped network, this study introduces a multi-branch feature hybrid attention and adaptive receptive field network (MFHARFNet) for medical image segmentation. Building upon the encoder-decoder framework, we initially devise a multi-branch feature hybrid attention module (MFHAM) to seamlessly integrate feature maps of varying scales, capturing both fine-grained features and coarse-grained semantics across the entire scale. Furthermore, we redesign the skip connection to amalgamate feature information from different branches in the encoder stage and efficiently transmit it to the decoder, providing the decoder with global context feature maps at different levels. Finally, the adaptive receptive field (ARF) module is introduced in the decoder feature reconstruction stage to adapt and focus on related fields, ensuring the model’s adaptation to different segmentation target features, and achieving different weights for the output of different convolution kernels to improve segmentation performance. We comprehensively evaluate our method on medical image segmentation tasks, by using four public datasets across CT and MRI. Remarkably, MFHARFNet method consistently outperforms other state-of-the-art methods, exceeding UNet by 2.1%, 0.9%, 6.6% and 1.0% on Dice on ATLAS, LiTs, BraTs2019 and Spine and intervertebral disc datasets, respectively. In addition, MFHARFNet minimizes network parameters and computational complexity as much as possible. The source codes are in https://github.com/OneHundred99/MFHARFNet.

To achieve precise and efficient skin cancer segmentation, an innovative SkinSegNet architecture is proposed. Inspired by U‐Net, SkinSegNet uses an encoder–decoder architecture incorporating advanced feature extraction and attention mechanisms. The encoder utilizes convolutional blocks and pooling attention (PA) layers to downsample feature maps and focus on significant regions. At the bottleneck, the proposed feature aggregation block integrates channel‐aware multihead attention (CAMA) and Adaptive Spatial Cross‐Scale Attention (ASCA) modules. These modules let the model capture channel relationships and complex spatial dependencies for accurate segmentation. The decoder reconstructs the segmentation mask through continuous upsampling and skips connections, ensuring fine‐grained spatial detail is preserved. Experiments are carried out on benchmark datasets namely ISIC 2016, ISIC 2017, and ISIC 2018 demonstrating SkinSegNet's superior performance in skin lesion segmentation, achieving state‐of‐the‐art accuracy of 95.12%, 94.01%, and 95.04%, respectively. Furthermore, cross‐dataset experiments show that SkinSegNet performs well on ISIC datasets, demonstrating its strong generalization ability.