Shokofeh Anari

Automated brain tumor segmentation provides precise evaluations of tumor dimensions and progression across time, which is beneficial for longitudinal studies and treatment oversight. This study provides a novel brain tumor segmentation method that combines a two-path UNet architecture with an upgraded EfficientNetV2 model due to an attention mechanism. By utilizing the advantages of both designs, our technique handles 3D MRI data efficiently and increases segmentation accuracy. By using two distinct encoding pathways, the UNet model is improved and can extract a greater variety of information from MRI images. The pretrained EfficientNetV2 model is also included to further improve the model’s capabilities. By including an attention mechanism, the model is able to concentrate on pertinent features, which further enhances segmentation performance. Using the BraTS 2020 dataset and the clinical dataset, we assessed the performance of the two-path UNet with EfficientNetV2B0 and EfficientNetV2S backbones. The real-clinical dataset comprises MRI scans from 99 patients acquired at Ankara Bilkent City Hospital. The findings showed that particular input combinations produced superior segmentation results, and some models performed exceptionally well at correctly categorizing various tumor kinds. This emphasizes how these models may be useful in clinical settings and how crucial it is to choose the right input configurations and model backbones to get the best segmentation results. Furthermore, the Grad-CAM technique improved the models’ reliability and interpretability, providing crucial insights into their decision-making processes.

This study conducts a comprehensive analysis of various configurations of the UNet + ConvNeXt Tiny architecture for breast tumor segmentation. We assess the influence of data augmentation, skip connections, attention mechanisms, and dropout rates on segmentation performance utilizing the BUSI dataset, subsequently conducting cross-dataset evaluation on the BUS-UCLM dataset. Our findings indicate that the integration of attention mechanisms markedly improves metrics including F1 score, precision, IoU, recall, and Dice coefficient. Data augmentation alone did not consistently enhance model performance and occasionally caused instability, however its integration with attention processes frequently resulted in significant improvements. Dropout was helpful in mitigating overfitting, with a rate of 0.5 consistently yielding enhanced generalization in both in-domain and cross-domain assessments, especially when combined with strategically positioned attention modules. Models using attention mechanisms and a dropout rate of 0.5 attained the optimal equilibrium between accuracy and recall, emphasizing the necessity of balancing regularization intensity with architectural intricacy. Nevertheless, our results underscore the potential for certain augmentation techniques—particularly those that employ aggressive geometric transformations—to distort anatomical structures in ultrasound images, resulting in a decline in the quality of segmentation. This highlights the necessity for meticulous, domain-specific selection of augmentation strategies in medical imaging. This study emphasizes the necessity for meticulous selection of architectural elements and regularization techniques to enhance medical image segmentation and offers direction for future investigations in resilient deep learning applications for clinical imaging.

This study utilizes the Breast Ultrasound Image (BUSI) dataset to present a deep learning technique for breast tumor segmentation based on a modified UNet architecture. To improve segmentation accuracy, the model integrates attention mechanisms, such as the Convolutional Block Attention Module (CBAM) and Non-Local Attention, with advanced encoder architectures, including ResNet, DenseNet, and EfficientNet. These attention mechanisms enable the model to focus more effectively on relevant tumor areas, resulting in significant performance improvements. Models incorporating attention mechanisms outperformed those without, as reflected in superior evaluation metrics. The effects of Dice Loss and Binary Cross-Entropy (BCE) Loss on the model’s performance were also analyzed. Dice Loss maximized the overlap between predicted and actual segmentation masks, leading to more precise boundary delineation, while BCE Loss achieved higher recall, improving the detection of tumor areas. Grad-CAM visualizations further demonstrated that attention-based models enhanced interpretability by accurately highlighting tumor areas. The findings denote that combining advanced encoder architectures, attention mechanisms, and the UNet framework can yield more reliable and accurate breast tumor segmentation. Future research will explore the use of multi-modal imaging, real-time deployment for clinical applications, and more advanced attention mechanisms to further improve segmentation performance.

This study introduces a sophisticated neural network structure for segmenting breast tumors. It achieves this by combining a pretrained Vision Transformer (ViT) model with a UNet framework. The UNet architecture, commonly employed for biomedical image segmentation, is further enhanced with depthwise separable convolutional blocks to decrease computational complexity and parameter count, resulting in better efficiency and less overfitting. The ViT, renowned for its robust feature extraction capabilities utilizing self-attention processes, efficiently captures the overall context within images, surpassing the performance of conventional convolutional networks. By using a pretrained ViT as the encoder in our UNet model, we take advantage of its extensive feature representations acquired from extensive datasets, resulting in a major enhancement in the model’s ability to generalize and train efficiently. The suggested model has exceptional performance in segmenting breast cancers from medical images, highlighting the advantages of integrating transformer-based encoders with efficient UNet topologies. This hybrid methodology emphasizes the capabilities of transformers in the field of medical image processing and establishes a new standard for accuracy and efficiency in activities related to tumor segmentation.

This study introduces a sophisticated neural network structure for segmenting breast tumors. It achieves this by combining a pretrained Vision Transformer (ViT) model with a U-Net framework. The U-Net architecture, commonly employed for biomedical image segmentation, is further enhanced with Depthwise Separable Convolutional Blocks to decrease computational complexity and parameter count, resulting in better efficiency and less overfitting. The Vision Transformer, renowned for its robust feature extraction capabilities utilizing self-attention processes, efficiently captures the overall context within images, surpassing the performance of conventional convolutional networks. By using a pretrained ViT as the encoder in our U-Net model, we take advantage of its extensive feature representations acquired from extensive datasets, resulting in a major enhancement in the model’s ability to generalize and train efficiently. The suggested model has exceptional performance in segmenting breast cancers from medical images, highlighting the advantages of integrating transformer-based encoders with efficient U-Net topologies. This hybrid methodology emphasizes the capabilities of transformers in the field of medical image processing and establishes a new standard for accuracy and efficiency in activities related to tumor segmentation.

Thyroid nodule is one of the common life-threatening diseases, and it had an increasing trend over the last years. Ultrasound imaging is a commonly used diagnostic method for detecting and characterizing thyroid nodules. However, assessing the entire slide images is time-consuming and challenging for the experts. For assessing ultrasound images in a meaningful manner, there is a need for automated, trustworthy, and objective approaches. The recent advancements in deep learning have revolutionized many aspects of computer-aided diagnosis (CAD) and image analysis tools that address the problem of diagnosing thyroid nodules. In this study, we explained the objectives of deep learning in thyroid cancer imaging and conducted a literature review on its potential, limits, and current application in this area. We gave an overview of recent progress in thyroid cancer diagnosis using deep learning methods and discussed various challenges and practical problems that might limit the growth of deep learning and its integration into clinical workflow.