Dual Focal Loss to address class imbalance in semantic segmentation

U-Net-Based Models for Precise Brain Stroke Segmentation

Ischemic stroke, a widespread neurological condition with a substantial mortality rate, necessitates accurate delineation of affected regions to enable proper evaluation of patient outcomes. However, such precision is complicated by factors like variable lesion sizes, noise interference, and the overlapping intensity characteristics of different tissue structures. This research addresses these issues by focusing on the segmentation of Diffusion Weighted Imaging (DWI) scans from the ISLES 2022 dataset and conducting a comparative assessment of three advanced deep learning models: the U-Net framework, its U-Net++ extension, and the Attention U-Net. Applying consistent evaluation criteria specifically, Intersection over Union (IoU), Dice Similarity Coefficient (DSC), and recall the Attention U-Net emerged as the superior choice, establishing record high values for IoU (0.8223) and DSC (0.9021). Although U-Net achieved commendable recall, its performance lagged behind that of U-Net++ in other critical measures. These findings underscore the value of integrating attention mechanisms to achieve more precise segmentation. Moreover, they highlight that the Attention U-Net model is a reliable candidate for medical imaging tasks where both accuracy and efficiency hold paramount importance, while U Net and U Net++ may still prove suitable in certain niche scenarios.

DeepGlioSeg: advanced glioma MRI data segmentation with integrated local-global representation architecture

Introduction

Glioma segmentation is vital for diagnostic decision-making, monitoring disease progression, and surgical planning. However, this task is hindered by substantial heterogeneity within gliomas and imbalanced region distributions, posing challenges to existing segmentation methods.

Methods

To address these challenges, we propose the DeepGlioSeg network, a U-shaped architecture with skip connections for continuous contextual feature integration. The model includes two primary components. First, a CTPC (CNN-Transformer Parallel Combination) module leverages parallel branches of CNN and Transformer networks to fuse local and global features of glioma images, enhancing feature representation. Second, the model computes a region-based probability by comparing the number of pixels in tumor and background regions and assigns greater weight to regions with lower probabilities, thereby focusing on the tumor segment. Test-time augmentation (TTA) and volume-constrained (VC) post-processing are subsequently applied to refine the final segmentation outputs.

Results

Extensive experiments were conducted on three publicly available glioma MRI datasets and one privately owned clinical dataset. The quantitative and qualitative findings consistently show that DeepGlioSeg achieves superior segmentation performance over other state-of-the-art methods.

Discussion

By integrating CNN- and Transformer-based features in parallel and adaptively emphasizing underrepresented tumor regions, DeepGlioSeg effectively addresses the challenges associated with glioma heterogeneity and imbalanced region distributions. The final pipeline, augmented with TTA and VC post-processing, demonstrates robust segmentation capabilities. The source code for this work is publicly available at https://github.com/smallboy-code/Brain-tumor-segmentation.

Improved Grain Boundary Reconstruction Method Based on Channel Attention Mechanism

The grain size of metal materials has a significant impact on their macroscopic properties. However, original metallographic images often suffer from issues such as substantial noise, missing grain boundaries, low contrast, and blurred edges. These challenges hinder the accurate extraction of complete grain boundaries, limiting the precision of grain size measurement and material performance prediction. Therefore, effectively reconstructing incomplete grain boundaries is particularly crucial. This paper proposes a grain boundary reconstruction and grain size measurement method based on an improved channel attention mechanism. A generative adversarial network (GAN) serves as the backbone, with a custom-designed channel attention module embedded in the generator. Combined with a global context attention mechanism, the method captures the global contextual information of the image, enhancing the network's semantic understanding and reconstruction accuracy for regions with missing grain boundaries. During the image reconstruction process, the method effectively leverages long-range feature correlations within the image, significantly improving network performance. To address the Mode Collapse observed during experiments, the loss function is optimized using Focal Loss, balancing the ratio of positive and negative samples and improving network robustness. Compared with other attention modules, the improved channel attention module significantly enhances the performance of the generative network. Experimental results demonstrate that the generative network based on this module outperforms comparable modules in terms of MIoU (86.25%), Accuracy (95.06%), and Precision (86.54%). The grain boundary reconstruction method based on the improved channel attention mechanism not only effectively improves the accuracy of grain boundary reconstruction but also significantly enhances the generalization ability of the network. This provides reliable technical support for the characterization of the microstructure and the performance prediction of metal materials.

Research on injection molded parts defect detection algorithm based on multiplicative feature fusion and improved attention mechanism

Injection molded parts are increasingly utilized across various industries due to their cost-effectiveness, lightweight nature, and durability. However, traditional defect detection methods for these parts often rely on manual visual inspection, which is inefficient, expensive, and prone to errors. To enhance the accuracy of defect detection in injection molded parts, a new method called MRB-YOLO, based on the YOLOv8 model, has been proposed. This method introduces several key improvements: (1) the MAFHead, a four-detection head based on multiplicative feature fusion, which replaces the original detection head to enhance feature representation; (2) the RepGFPN-SE module, a re-parameterized generalized feature pyramid network that improves detection of small objects by replacing the original C2f. module; (3) and the BiNorma module, employing a bi-level routing attention mechanism to optimize the training process by reducing input distribution changes across layers. The effectiveness of the MRB-YOLO model was validated through ablation and contrast experiments using a specially constructed dataset of injection molded parts defects. The results demonstrated an accuracy of 88.8%, a recall rate of 86.8%, and a mean average precision (mAP) of 91.5%. Compared to the YOLOv8n model, the MRB-YOLO model shows an increase in accuracy by 8.2%, in recall rate by 17.2%, and in mAP by 11.8%. These findings confirm that the MRB-YOLO model meets the requirements for accurate detection of defects in injection molded parts.

BCL-Former: Localized Transformer Fusion with Balanced Constraint for polyp image segmentation

Polyp segmentation remains challenging for two reasons: (a) the size and shape of colon polyps are variable and diverse; (b) the distinction between polyps and mucosa is not obvious. To solve the above two challenging problems and enhance the generalization ability of segmentation method, we propose the Localized Transformer Fusion with Balanced Constraint (BCL-Former) for Polyp Segmentation. In BCL-Former, the Strip Local Enhancement module (SLE module) is proposed to capture the enhanced local features. The Progressive Feature Fusion module (PFF module) is presented to make the feature aggregation smoother and eliminate the difference between high-level and low-level features. Moreover, the Tversky-based Appropriate Constrained Loss (TacLoss) is proposed to achieve the balance and constraint between True Positives and False Negatives, improving the ability to generalize across datasets. Extensive experiments are conducted on four benchmark datasets. Results show that our proposed method achieves state-of-the-art performance in both segmentation precision and generalization ability. Also, the proposed method is 5%-8% faster than the benchmark method in training and inference. The code is available at: https://github.com/sjc-lbj/BCL-Former.

Rethinking automatic segmentation of gross target volume from a decoupling perspective

Accurate and reliable segmentation of Gross Target Volume (GTV) is critical in cancer Radiation Therapy (RT) planning, but manual delineation is time-consuming and subject to inter-observer variations. Recently, deep learning methods have achieved remarkable success in medical image segmentation. However, due to the low image contrast and extreme pixel imbalance between GTV and adjacent tissues, most existing methods usually obtained limited performance on automatic GTV segmentation. In this paper, we propose a Heterogeneous Cascade Framework (HCF) from a decoupling perspective, which decomposes the GTV segmentation into independent recognition and segmentation subtasks. The former aims to screen out the abnormal slices containing GTV, while the latter performs pixel-wise segmentation of these slices. With the decoupled two-stage framework, we can efficiently filter normal slices to reduce false positives. To further improve the segmentation performance, we design a multi-level Spatial Alignment Network (SANet) based on the feature pyramid structure, which introduces a spatial alignment module into the decoder to compensate for the information loss caused by downsampling. Moreover, we propose a Combined Regularization (CR) loss and Balance-Sampling Strategy (BSS) to alleviate the pixel imbalance problem and improve network convergence. Extensive experiments on two public datasets of StructSeg2019 challenge demonstrate that our method outperforms state-of-the-art methods, especially with significant advantages in reducing false positives and accurately segmenting small objects. The code is available at https://github.com/shijun18/GTV_AutoSeg.

Multi-objective pedestrian tracking method based on YOLOv8 and improved DeepSORT

A multi-objective pedestrian tracking method based on you only look once-v8 (YOLOv8) and the improved simple online and real time tracking with a deep association metric (DeepSORT) was proposed with the purpose of coping with the issues of local occlusion and ID dynamic transformation that frequently arise when tracking target pedestrians in real complex traffic scenarios. To begin with, in order to enhance the feature extraction network's capacity to learn target feature information in busy traffic situations, the detector implemented the YOLOv8 method with a high level of small-scale feature expression. In addition, the omni-scale network (OSNet) feature extraction network was then put on top of DeepSORT in order to accomplish real-time synchronized target tracking. This increases the effectiveness of picture edge recognition by dynamically fusing the collected feature information at various scales. Furthermore, a new adaptive forgetting smoothing Kalman filtering algorithm (FSA) was created to adapt to the nonlinear condition of the pedestrian trajectory in the traffic scene in order to address the issue of poor prediction attributed to the linear state equation of Kalman filtering once more. Afterward, the original intersection over union (IOU) association matching algorithm of DeepSORT was replaced by the complete-intersection over union (CIOU) association matching algorithm to fundamentally reduce the target pedestrians' omission and misdetection situation and to improve the accuracy of data matching. Eventually, the generalized trajectory feature extractor model (GFModel) was developed to tightly merge the local and global information through the average pooling operation in order to get precise tracking results and further decrease the impact of numerous disturbances on target tracking. The fusion algorithm of YOLOv8 and improved DeepSORT method based on OSNet, FSA and GFModel was named YOFGD. According to the experimental findings, YOFGD's ultimate accuracy can reach 77.9% and its speed can reach 55.8 frames per second (FPS), which is more than enough to fulfill the demands of real-world scenarios.

A non-negative feedback self-distillation method for salient object detection

Self-distillation methods utilize Kullback-Leibler divergence (KL) loss to transfer the knowledge from the network itself, which can improve the model performance without increasing computational resources and complexity. However, when applied to salient object detection (SOD), it is difficult to effectively transfer knowledge using KL. In order to improve SOD model performance without increasing computational resources, a non-negative feedback self-distillation method is proposed. Firstly, a virtual teacher self-distillation method is proposed to enhance the model generalization, which achieves good results in pixel-wise classification task but has less improvement in SOD. Secondly, to understand the behavior of the self-distillation loss, the gradient directions of KL and Cross Entropy (CE) loss are analyzed. It is found that KL can create inconsistent gradients with the opposite direction to CE in SOD. Finally, a non-negative feedback loss is proposed for SOD, which uses different ways to calculate the distillation loss of the foreground and background respectively, to ensure that the teacher network transfers only positive knowledge to the student. The experiments on five datasets show that the proposed self-distillation methods can effectively improve the performance of SOD models, and the average F_β is increased by about 2.7% compared with the baseline network.

Deep learning approaches to landmark detection in tsetse wing images

Morphometric analysis of wings has been suggested for identifying and controlling isolated populations of tsetse (Glossina spp), vectors of human and animal trypanosomiasis in Africa. Single-wing images were captured from an extensive data set of field-collected tsetse wings of species Glossina pallidipes and G. m. morsitans. Morphometric analysis required locating 11 anatomical landmarks on each wing. The manual location of landmarks is time-consuming, prone to error, and infeasible for large data sets. We developed a two-tier method using deep learning architectures to classify images and make accurate landmark predictions. The first tier used a classification convolutional neural network to remove most wings that were missing landmarks. The second tier provided landmark coordinates for the remaining wings. We compared direct coordinate regression using a convolutional neural network and segmentation using a fully convolutional network for the second tier. For the resulting landmark predictions, we evaluate shape bias using Procrustes analysis. We pay particular attention to consistent labelling to improve model performance. For an image size of 1024 × 1280, data augmentation reduced the mean pixel distance error from 8.3 (95% confidence interval [4.4,10.3]) to 5.34 (95% confidence interval [3.0,7.0]) for the regression model. For the segmentation model, data augmentation did not alter the mean pixel distance error of 3.43 (95% confidence interval [1.9,4.4]). Segmentation had a higher computational complexity and some large outliers. Both models showed minimal shape bias. We deployed the regression model on the complete unannotated data consisting of 14,354 pairs of wing images since this model had a lower computational cost and more stable predictions than the segmentation model. The resulting landmark data set was provided for future morphometric analysis. The methods we have developed could provide a starting point to studying the wings of other insect species. All the code used in this study has been written in Python and open sourced.