International Journal of Imaging Systems and Technology最新文献

Glaucoma is a dangerous disease that can lead to blindness in advanced stages. It has been a hot topic among machine learning researchers as it can be diagnosed quickly and effectively through optic disc segmentation. However, anomalies in the optic disk region significantly complicate this task. There is also a need for a model that produces stable results on different data sets. Therefore, in this study, we propose a novel method that can be used for early diagnosis and treatment of glaucoma. Using deep learning architecture, hypercolumn deep features extracted from retinal fundus images were trained with different classifiers, and their behavior was examined in depth according to varying thresholding values. Global and local thresholding approaches were developed to improve the predictions of the standard classifier. The proposed model achieved Dice scores of 0.9437, 0.9654, and 0.9407 on the DRIONS, DRISHTI, and RIMONE-v3 datasets, respectively. The results obtained at the end of the study show that the proposed model is robust on these datasets and is competitive with other studies in the literature. In conclusion, we showed that the proposed model can be used effectively in glaucoma diagnosis.

Hand fractures, particularly phalangeal fractures, are often subtle, concealed, and complex, posing significant challenges for accurate and efficient diagnosis with traditional radiography. To address the limitations of existing deep learning methods in handling complex anatomy, multi-view uncertainty, and causal discrimination, we propose a novel two-stage context-aware multi-view fusion framework for precise fracture classification and localization. In the first stage, a Context-Aware Multi-View Classification Network (CAMVC-Net) is developed. A Dual-path Semantic-Spatial Alignment (DSSA) module aligns low-level spatial details with high-level semantics, enhancing fine-grained fracture discrimination. A Context Aware Pyramid (CAP) module captures multi-scale context, while a Counterfactual Attention Learning (CAL) loss guides the network to focus on discriminative regions. Multi-view uncertainty is modeled using a Dirichlet distribution, and decision-level fusion is achieved by Dempster's rule to mimic clinical reasoning. In the second stage, a Detail-aware Fracture Localization Network (DFL-Net) is designed. To adapt to irregular fracture geometry, DFL-Net integrates deformable convolutions and incorporates DSSA into the feature pyramid to preserve fine spatial details during deep downsampling. Experiments on the MURA and clinical datasets demonstrate strong performance: the classification model achieved an accuracy of 92.89%, precision of 93.05%, recall of 96.96%, and F1 score of 94.97%. The localization network obtained 73.8% AP₅₀ and 43.8% AP₇₅, with a 10.4% improvement in AP₇₅ over Faster R-CNN. These results indicate that the proposed framework provides an accurate and efficient tool for computer-aided fracture diagnosis, with the potential to reduce misdiagnosis and improve clinical decision-making efficiency.

In CT imaging, the size, shape, margin, and density of pulmonary nodules are used to determine whether they are benign or malignant. However, pulmonary nodules often exhibit challenging characteristics in CT images, such as irregular shapes, and tiny nodules are prone to being overlooked. In addition, the density of nodules may be similar to that of surrounding tissues. When the nodules are small or close to the pulmonary wall, their resolution is low, making them difficult to distinguish. Because of these characteristics, it is quite challenging to segment pulmonary nodules automatically. This research proposes CCEMSS-Unet++, a medical image segmentation network that enhances feature fusion between local details and global context through a multi-scale design. It includes the CCEMS module and the SE attention mechanism. The SE module mitigates false negatives, particularly for small nodules; the CCEMS module is used to extract local information and connect global context, improving segmentation accuracy. This study used two datasets: the publicly available LIDC-IDRI dataset and a dataset collected from the CT Department of Yan'an University Affiliated Hospital. Compared with Unet++, the proposed model improves IoU and Dice by 2.62% and 2.15% on the LIDC-IDRI dataset, and by 5.10% and 7.11% on our dataset, respectively. In terms of segmentation accuracy and generalization, the experimental findings demonstrate that this approach outperforms other networks, including handling nodules with low resolution. The efficacy of the proposed enhancement method is further supported by the ablation experiments.

Precise segmentation of critical retinal structures, including the optic cup, optic disc, and vascular bifurcation sites, is essential for the early identification and management of severe ocular conditions such as glaucoma, diabetic retinopathy, hypertensive retinopathy, and age-related macular degeneration. This study proposes a hybrid multi-attention residual-based UNet (MRUNet+), a deep learning (DL) model developed to address the complex nature of retinal image processing using an advanced attention mechanism to enhance segmentation precision. MRUNet+ was trained on many high-quality datasets, including DRIVE, CHASE_DB1, Drishti-GS1, RIM-ONE and REFUGE1, enabling it to address prevalent imaging issues, such as variations in contrast, resolution and the presence of abnormalities. The model has exhibited markedly enhanced performance compared to recent methodologies, with improved accuracy metrics, Dice coefficients and intersection over union (IoU), particularly in low-contrast or complex anatomical regions. Due to its adaptability and precision, MRUNet+ has the potential to function as a pivotal instrument in clinical settings, providing automated retinal assessments that aid in the prompt detection of various retinal and vascular pathologies, thereby improving patient management.

Accurate grading of invasive ductal carcinoma (IDC) remains hard even for the trained pathologist. The patient's therapeutic schedule depends on the observed grade, emphasizing the importance of accurate tissue grading. We propose a novel architecture called Twin Tokens Talking-heads Class Attention Net (T2TCAN), designed to grade IDC images. Based on the transformer architecture, this design enhances the model's ability to capture local and global tissue features. This model was trained on the PathoIDCG dataset, outperforming the existing methods in terms of standard metrics. Further, introducing class and twin tokens later in the model, along with the talking-heads class attention and per-channel scaling, made the T2TCAN model efficient at distinguishing the tissue grades. Our model outperformed conventional CNN-based IDC grading approaches. We report the highest 98.82% accuracy, AUC (0.99), precision (98.83%), recall (98.80%), and F1-score (98.81%) on the PathoIDCG dataset. We plotted class attention maps for the TCT layers to provide insights into the model's decision-making process. These attention maps are also helpful for pathologists to understand the discriminative features that significantly contribute to prognostic scoring. Further, our model also performed well on a separate IDC grading dataset, producing state-of-the-art results and thus proving its efficacy. The designed model claims its adaptability and suitability for integration into the healthcare workflow for IDC grading.

The Mamba model performs excellently in natural image processing but faces limitations in analyzing whole slide images (WSIs) for cancer prediction and subtype classification in digital pathology—pathological images feature highly irregular lesion spatial distributions (especially complex small-lesion associations), while Mamba's inherent unidirectional/limited-direction scanning cannot effectively model such multi-dimensional spatial dependencies, failing to capture key pathological structural features. To address this, we propose HCSMIL, a Mamba-based optimized framework tailored to pathological image clinical analysis. It comprehensively captures local lesion spatial topology via multi-directional contextual modeling and integrates a multi-scale pyramid structure to extract global lesion distribution features, jointly enhancing diagnostic accuracy. Validation on authoritative datasets (Camelyon16, TCGA-LUNG, TCGA-Kidney) shows HCSMIL significantly outperforms existing mainstream methods: on TCGA-LUNG, accuracy (ACC), F1 score, and AUC are 0.66%, 1.42%, and 1.25% higher than the second-best method; on TCGA-Kidney, these metrics increase by 1.47%, 0.09%, and 1.00%; on Camelyon16, ACC is 0.77% higher. Notably, HCSMIL achieves an 84% small-lesion recognition rate, substantially exceeding TransMIL (70.59%) and MambaMIL (64.71%), fully demonstrating its strength in capturing complexly distributed lesions and providing reliable technical support for cancer diagnosis.

Lung tumor segmentation using machine learning and artificial intelligence techniques leverages the diagnosis precision through accurate localization of the infections. The prominent factors are the features that reflect the infected region concealed through patterns, boundaries, and edges. In this article, a novel Feature-Derivative Pixel Segmentation (FDPS) method is introduced to improve the tumor segmentation accuracy influenced by the disparity pixel distribution problem. This proposed method is assisted by tuneable recurrent learning (TRL) to vary the feature derivative count for varying segments. The learning inputs are modifiable using different extracted feature derivatives under parity and disparity pixel distributions. By identifying the maximum disparity pixels, the tuneable inputs for the recurrent learning are decided. The computation layer of the learning process identifies the maximum related regions identified under parity and disparity features. Such regions are segmented from multiple pixel distribution points until the image size. This process is therefore iterated to identify maximum conjoined features under different infected regions. The proposed method improves the accuracy by 9.63%, the true positive rate by 10.85%, and reduces the classification error by 10.06% for the maximum regions.

Diabetic retinopathy (DR) is a leading cause of blindness among individuals with diabetes. Timely diagnosis and precise classification of DR are essential for patients. However, the traditional diagnostic methods have limitations in precision, mainly relying on doctors' experiences and subjective judgments on DR images. Therefore, an efficient network model, named SDCSCF-Net, is proposed based on deep learning for DR diagnosis and classification. Firstly, the SE_Double_Conv (SDC) module is designed by integrating the Squeeze-and-Excitation (SE) attention mechanism into the first Double_Conv block of the encoder structure of U-Net to enhance feature representation and suppress redundant information. Secondly, a novel attention mechanism, spatial channel fusion attention (SCFA) module, is proposed to enhance the model's focus on lesion areas and the relationship between the channels, making the model more effectively distinguish subtle differences between adjacent DR classes. Finally, the proposed model is evaluated on the APTOS 2019 dataset, which contains 3662 fundus images. The results show that the proposed model demonstrates superior classification performance for DR compared to other existing approaches, and its accuracy, precision, recall, and F1-score for binary classification of DR are 99.18%, 99.47%, 98.98%, and 99.19%, respectively. For the five-class classification task, the model achieves an accuracy of 84.72%, a precision of 84.12%, a recall of 84.72%, and an F1-score of 84.02%. All the evaluation metrics are obtained from the testing phase of the model. In addition, the Grad-CAM technology is utilized to visualize the key lesion areas concerned by the model and further verifies the effectiveness of the proposed model. It is beneficial to promote the research and practical application in the intelligent diagnosis of DR.