International Journal of Imaging Systems and Technology最新文献

Skin lesion segmentation from dermoscopic images must be done accurately and consistently in order to diagnose diseases and arrange treatments. However, when dealing with issues like fuzzy lesion region boundaries, multiscale features, and notable variations in the lesion region's size, shape, and color, existing methods typically have high computational complexity and large parameter counts. They also frequently suffer from decreased segmentation accuracy due to inadequate capture of local features and global information. In this paper, a lightweight deep learning network based on high-performance adaptive attention is proposed to overcome these issues. Specifically, a deep convolutional neural network is introduced to capture local information. Meanwhile, we create a high-performance adaptive attention feature fusion module (EAAF) that uses dynamic feature selection to achieve adaptive fusion of global information with multiscale local features. Furthermore, we created a reverse dynamic feature fusion module (RDFM) at the decoding stage to efficiently fuse features at various levels while taking into account the integrity and specifics of the lesion region to increase the precision of complex lesion region segmentation. We carried out in-depth tests on three publicly accessible datasets International Skin Imaging Collaboration (ISIC)-2016, ISIC-2018, and PH² to assess the method's efficacy and contrasted the outcomes with those of the most advanced techniques; the results confirmed that the suggested approach was superior.

Gastrointestinal (GI) bleeding, arising from various conditions, can be critical if untreated. Wireless capsule endoscopy (WCE) is a highly effective method for detecting GI bleeding, offering full visualization of the GI tract. However, the large number of images generated per patient poses challenges for clinicians, leading to prolonged analysis times and increased risk of human error. This emphasizes the need for computer-aided diagnosis systems. In this study, we introduce SEA-Net (Structured Efficient Attention Network), a novel deep learning network for detecting bleeding regions in WCE images. SEA-Net integrates a Convolutional Block Attention Module (CBAM) with long skip connections to enhance gradient flow and improve blood region localization. The EfficientNet-B4 encoder improves feature extraction efficiency and generalizability. A five-fold cross validation demonstrates consistent performance, while generalization tests, including precision-recall curves, ROC curves, and F1 measure, further validate the model's robustness. Minimal performance degradation was observed when the training data was reduced from 80% to 20%. Experimental results show that SEA-Net achieves a Dice score of 93.64% and an IoU score of 88.61% on a publicly available WCE dataset, outperforming state-of-the-art models and highlighting its strong potential for clinical application.

Heart disease remains a leading global health concern, necessitating accurate and interpretable risk prediction models for effective clinical decision-making. Accurate heart disease risk prediction is crucial for preventive healthcare, yet traditional machine learning models often struggle with the inherent uncertainty and nonlinear patterns in medical data. While deep neural networks (DNNs) excel at feature extraction, their lack of interpretability limits clinical utility, whereas fuzzy inference systems (FIS) offer transparency but lack hierarchical learning capabilities. To bridge this gap, we propose a novel Deep Fuzzy Inference System (DFIS) that integrates DNNs and FIS into a unified architecture, combining the strengths of both approaches. The DFIS leverages a weighted fusion mechanism to combine probabilistic outputs from an Adam Cuckoo Search-optimized DNN and a trapezoidal membership-based FIS, enabling simultaneous high accuracy and interpretability. The performance is evaluated on the Cleveland heart disease dataset; the DFIS achieves 97.2% accuracy, outperforming a standalone DNN (95.4%) and ANFIS (91.7%) under the same experimental conditions while providing clinically actionable risk stratification into normal, less critical, and very critical categories.

This study presents a high performance hybrid deep learning model for the classification of 14 lung diseases using chest X-ray (CXR) images. Manual evaluation of CXR images is labor-intensive and prone to human error. Therefore, automated systems are required to improve diagnostic accuracy and efficiency. Our model integrates ResNet18 and EfficientNet-V2-S architectures, combining residual connections with efficient scaling to achieve high accuracy while maintaining computational efficiency. Trained on the NIH ChestX-ray14 dataset, comprising 112 120 images across 14 disease classes, the model mitigates class imbalances with extensive data augmentation techniques. Achieving an impressive average AUC of 0.872, the model outperforms previous approaches. This performance was enhanced by a refined, anatomically-aware data augmentation strategy that improved the model's robustness and clinical relevance, particularly in challenging disease categories such as Pneumothorax, Emphysema, and Hernia. To further validate its generalizability, the proposed model was tested on three additional datasets for pneumonia, COVID-19, and tuberculosis. The results demonstrate superior performance, achieving an accuracy of 0.958, F1 score of 0.944, and ROC AUC of 0.989 for pneumonia; an accuracy of 0.974, F1 score of 0.969, and ROC AUC of 0.995 for COVID-19; and an accuracy of 0.999, F1 score of 0.999, and ROC AUC of 0.999 for tuberculosis. These outstanding results confirm the robustness and clinical applicability of the model across diverse datasets. This research introduces a reliable and efficient diagnostic tool that enhances the potential of automated lung disease classification. By alleviating radiologists' workload and promoting timely, accurate diagnostic outcomes, the model contributes significantly to medical imaging applications and demonstrates its capacity for practical use in real-world clinical settings.

Macular edema is a retinal disorder that can lead to significant vision loss, underscoring the need for accurate and intelligent automated diagnosis. However, its subtle manifestations in color fundus photography (CFP) pose considerable challenges for conventional deep learning models. In this work, we propose a novel diagnostic framework that integrates Dempster–Shafer (D–S) evidence theory—a principled approach for uncertainty quantification and multi-source information fusion—with the advanced Mamba architecture. The proposed method employs a dual-branch network to selectively enhance and extract discriminative features from both bright and dark regions of fundus images. These features are dynamically aligned and fused via an Adaptive Multi-Branch Feature Synthesis (AMFS) module. To model long-range dependencies and aggregate complementary information from multiple scanning views, we introduce a multi-scan Mamba module, whose outputs are further fused using a principled D–S evidence mechanism. This synergistic integration of mathematical theory and deep learning not only reduces information redundancy but also enables confidence-aware automated decision-making. Extensive experiments on three retinal image datasets—IDRiD, Messidor, and a proprietary clinical cohort—demonstrate that our framework consistently outperforms state-of-the-art methods in terms of accuracy, F1-score, and robustness. These results highlight the promise of combining evidence theory with modern deep learning for challenging medical image analysis tasks.

Electroencephalogram (EEG) plays a vital role in identifying the neural activities and human behaviors. EEG is a high-temporal-resolution signal that is contaminated and influenced by various external factors and artifacts. It is very difficult to identify and analyze the EEG signal when it is contaminated by artifacts. The impact of artifacts in EEG signals became an open challenge in EEG signal processing. Even though many techniques and methods are available to remove the artifacts in EEG, analyzing the contaminated EEG signals remains a significant challenge. The signal is polluted by different artifacts like eye blink, muscle activities, environmental interfaces, and so on. Compared with other artifacts and noises, task-specific artifacts such as motor movement and motor imagery have a greater impact on EEG signals due to their direct inference on cognitive signals, and isolating task-specific artifacts from the EEG signal is difficult. To overcome these challenges, a novel hybrid task-specific Dynamic Multistage k-Means Clustering algorithm (TS-DMKC) has been proposed that detects and extracts the processed EEG signal, which will be helpful for all the EEG applications. Initially, the motor movement and motor imagery artifacts are distinguished using the Fast ICA by applying the filtering threshold. Then, a multistage k-means clustering algorithm is employed to isolate the artifacts in different clusters dynamically, and the cluster effectiveness is calculated by using the silhouette score. This proposed novel hybrid algorithm will be useful in various EEG applications like neuroscience, medical diagnosis, brain–computer interface, authentication, brain signal analysis, and gaming. The experimental results demonstrate that the proposed algorithm TS-DMKC achieved a classification accuracy of 94.2% using the Physionet motor movement/imagery dataset, reducing task-dependent variability and offering more robust and reliable EEG systems.

Accurate diagnosis of diseases relies on the subjective experience of physicians. Automatic image segmentation technology can assist doctors in quickly locating lesion areas, thereby improving diagnostic efficiency. In the field of medical image segmentation, the current mainstream models are mostly variants of Swin-Unet, which perform well on large-sample datasets. However, due to the window constraints of the Swin-Transformer architecture, these models often face performance limitations when processing small-sample datasets. In contrast, convolutional-based Unet variant models exhibit better segmentation performance under small-sample conditions but demonstrate poorer segmentation results for medical images with blurred boundaries. To tackle these challenges, this paper introduces a novel boundary information-enhanced architecture, the Hybrid-Parallel-Attention and Pyramid-Edge-Extraction Enhanced UNet (HPPE-Unet). Based on an encoder-decoder architecture of convolutional neural networks, the model incorporates a multi-scale feature extraction strategy and an optimized skip-connection mechanism. It integrates two key modules: the Pyramid Edge Extraction (PEE) module and the Hybrid Parallel Attention (HP) module. The PEE module is applied at each stage of the encoder and decoder, significantly enhancing the model's ability to capture subtle boundary structures through multi-scale feature fusion and boundary reinforcement mechanisms. The HP employs a residual parallel structure combining spatial (SA) and channel attention (CA) blocks, effectively bridging the semantic gap between features in the encoding and decoding stages and addressing the issue of information loss in skip connections between the encoder and decoder. The proposed method was evaluated on an aortic dissection dataset provided by a tertiary hospital. The results show that the method achieved a Dice similarity coefficient (DSC) of 97.65% and a Mean intersection over union (Miou) of 97.92% in the segmentation task. On the public Automated Cardiac Diagnostic Challenge (ACDC) dataset, the DSC reached 90.39%. These results demonstrate that the proposed method holds significant practical value for clinical disease diagnosis.

Classical self-training methods for graph convolutional networks (GCNs) assume that both labeled and unlabeled data follow the identical distribution. However, these works do not work well when they are used in medical applications such as classifying autism spectrum disorder (ASD) in multiple centers, where the unlabeled samples from different imaging centers have varying distribution shifts from the labeled samples. To this end, we propose uncertainty-aware graph self-training (UA-GST) by extending graph self-training to a new situation, in which the labeled data come from one imaging center and the unlabeled data come from several other imaging centers. Specifically, an uncertainty-aware mechanism is proposed to select unlabeled centers and adversarial domain adaptation is introduced into graph self-training to reduce domain shift between centers. With the progression of self-training, more pseudo-labeled test samples are gradually included in the training set, and a final model is finally trained on all the labeled and pseudo-labeled samples. Considering the over-confidence issue of the classifier, an evidential GCN is further proposed to estimate the uncertainty of the pseudo-labels using Dempster–Shafer (D–S) evidence theory. It is evaluated on the Autism Brain Imaging Data Exchange (ABIDE). Experimental results verified the effectiveness of the proposed method in classifying ASD in multiple centers, outperforming existing state-of-the-art methods.

The accurate survival prediction in glioblastoma patients remains a major challenge, largely owing to the considerable variability among brain tumor subtypes and their clinical profiles. This research introduces an innovative SimCLR-Twin Critic DDPG framework that integrates self-supervised representation learning with deep reinforcement learning for enhanced prognostic prediction. Tumor subregion segmentations were independently obtained using four advanced models namely nnU-Net, VNet, SwinUNETR, and UNet. Based on the performance of these segmentation models, deep features were extracted from the nnU-Net segmented tumor subregions and combined with handcrafted radiomics features. The combined features were subjected to the feature selection process using BorutaShap, followed by self-supervised pretraining through SimCLR to improve feature representation and generalization. The optimized features were then fed into a Twin Critic DDPG agent designed for regression-based survival time predictions. The proposed method outperformed existing techniques on the BraTS 2020 dataset, achieving the lowest MSE of 31 062.36 and the highest C-index of 0.87 in overall survival prediction. To further confirm robustness and generalizability, the framework was externally validated on the BraTS 2019 dataset, where it achieved a comparable MSE of 31 156.56 and a C-index of 0.86 using fused features. Additionally, LIME-based local interpretability provided clinically relevant explanations for individual predictions, thereby enhancing trust in the AI-driven system. This study highlights the SimCLR-Twin Critic DDPG framework as a robust and interpretable solution for accurate survival prediction in glioblastoma patients.

This research dives into creating optimized neural network architectures tailored for image classification and medical image segmentation, with a particular emphasis on analyzing lung nodules using the LIDC-IDRI dataset. We introduce a hybrid approach to neural architecture search (NAS) that fuses convolutional neural networks (CNNs), transformer-based models, and custom-built graph architectures. Our evolutionary strategies employ genetic operations like crossover and mutation to gradually enhance these architectures, while the NSGA-III algorithm helps us navigate the tricky balance of multiple conflicting objectives. The goal of our method is to boost performance metrics such as the F1 score and classification accuracy, all while keeping an eye on minimizing computational complexity, which we measure in terms of FLOPS, parameter count, and inference time. Our experiments demonstrate competitive performance on the CIFAR-10 and CIFAR-100 datasets, with promising results on the LIDC-IDRI segmentation task. This work aims to push the boundaries of automated model design, contributing to more efficient and effective deep learning architectures in both general and medical imaging contexts.