A hybrid attention network with convolutional neural network and transformer for underwater image restoration.
Ontology highlight
ABSTRACT:
SUBMITTER: Jiao Z
PROVIDER: S-EPMC10702994 | biostudies-literature | 2023
REPOSITORIES: biostudies-literature
Publications
A hybrid attention network with convolutional neural network and transformer for underwater image restoration.
Jiao Zhan Z Wang Ruizi R Zhang Xiangyi X Fu Bo B Thanh Dang Ngoc Hoang DNH
PeerJ. Computer science 20231107
The analysis and communication of underwater images are often impeded by various elements such as blur, color cast, and noise. Existing restoration methods only address specific degradation factors and struggle with complex degraded images. Furthermore, traditional convolutional neural network (CNN) based approaches may only restore local color while ignoring global features. The proposed hybrid attention network combining CNN and Transformer focuses on addressing these issues. CNN captures loca ...[more]
Similar Datasets
Project description:BackgroundMedical image segmentation is crucial for improving healthcare outcomes. Convolutional neural networks (CNNs) have been widely applied in medical image analysis; however, their inherent inductive biases limit their ability to capture global contextual information. Vision transformer (ViT) architectures address this limitation by leveraging attention mechanisms to model global relationships; however, they typically require large-scale datasets for effective training, which is challenging in the field of medical imaging due to limited data availability. This study aimed to integrate the advantages of CNN and ViT architectures to improve segmentation performance on small-scale medical image datasets.MethodsIn this study, we established a U-shaped network architecture based on a Transformer-assisted convolutional neural network (TAC-UNet). The TAC-UNet is primarily composed of a hybrid structure integrating CNN and Transformer components. Specifically, the hybrid architecture follows a dual-path design in which the Transformer branch continuously conveys global contextual information to the CNN backbone. This allows the CNN backbone to enhance its global perception while building on the local features it extracts, thereby improving its ability to comprehend complex image structures. A channel cross-attention (CCA) module is also incorporated as a bridge between the encoder and decoder to better reconcile the semantic discrepancies between them.ResultsDetailed experiments on three public datasets were conducted. Specifically, our model was trained on 30 images from the Multi-organ Nucleus Segmentation (MoNuSeg) training dataset, 85 images from the Gland Segmentation (GlaS) training dataset, and 551 images from the Computer Vision Center Colorectal Cancer-Clinic Database (CVC-ClinicDB) dataset. We evaluated the performance of our model on the corresponding test sets. Our TAC-UNet achieved the best Dice scores (80.36%, 90.70%, and 91.81% on the MoNuSeg, GlaS, and CVC-ClinicDB datasets, respectively) of all the models. Compared to other CNN-based, Transformer-based, and hybrid methods, the TAC-UNet demonstrated significantly superior segmentation performance.ConclusionsOur TAC-UNet model showed advanced segmentation performance on small-scale medical image datasets. The detailed experimental results showed the effectiveness of the method. Our model's code is available at: https://github.com/hejlhello/TAC-UNet.
| S-EPMC11651933 | biostudies-literature
Project description:Extracting information from scanned images is a critical task with far-reaching practical implications. Traditional methods often fall short by inadequately leveraging both image and text features, leading to less accurate and efficient outcomes. In this study, we introduce ConBGAT, a cutting-edge model that seamlessly integrates convolutional neural networks (CNNs), Transformers, and graph attention networks to address these shortcomings. Our approach constructs detailed graphs from text regions within images, utilizing advanced Optical Character Recognition to accurately detect and interpret characters. By combining superior extracted features of CNNs for image and Distilled Bidirectional Encoder Representations from Transformers (DistilBERT) for text, our model achieves a comprehensive and efficient data representation. Rigorous testing on real-world datasets shows that ConBGAT significantly outperforms existing methods, demonstrating its superior capability across multiple evaluation metrics. This advancement not only enhances accuracy but also sets a new benchmark for information extraction in scanned image.
| S-EPMC11622835 | biostudies-literature
Project description:ObjectiveEarly diagnosis of osteoporosis is crucial to prevent osteoporotic vertebral fracture and complications of spine surgery. We aimed to conduct a hybrid transformer convolutional neural network (HTCNN)-based radiomics model for osteoporosis screening in routine CT.MethodsTo investigate the HTCNN algorithm for vertebrae and trabecular segmentation, 92 training subjects and 45 test subjects were employed. Furthermore, we included 283 vertebral bodies and randomly divided them into the training cohort (n = 204) and test cohort (n = 79) for radiomics analysis. Area receiver operating characteristic curves (AUCs) and decision curve analysis (DCA) were applied to compare the performance and clinical value between radiomics models and Hounsfield Unit (HU) values to detect dual-energy X-ray absorptiometry (DXA) based osteoporosis.ResultsHTCNN algorithm revealed high precision for the segmentation of the vertebral body and trabecular compartment. In test sets, the mean dice scores reach 0.968 and 0.961. 12 features from the trabecular compartment and 15 features from the entire vertebral body were used to calculate the radiomics score (rad score). Compared with HU values and trabecular rad-score, the vertebrae rad-score suggested the best efficacy for osteoporosis and non-osteoporosis discrimination (training group: AUC = 0.95, 95%CI 0.91-0.99; test group: AUC = 0.97, 95%CI 0.93-1.00) and the differences were significant in test group according to the DeLong test (p < 0.05).ConclusionsThis retrospective study demonstrated the superiority of the HTCNN-based vertebrae radiomics model for osteoporosis discrimination in routine CT.
| S-EPMC10938662 | biostudies-literature
Project description:Face and face mask detection are one of the most popular topics in computer vision literature. Face mask detection refers to the detection of people's faces in digital images and determining whether they are wearing a face mask. It can be of great benefit in different domains by ensuring public safety through the monitoring of face masks. Current research details a range of proposed face mask detection models, but most of them are mainly based on convolutional neural network models. These models have some drawbacks, such as their not being robust enough for low quality images and their being unable to capture long-range dependencies. These shortcomings can be overcome using transformer neural networks. Transformer is a type of deep learning that is based on the self-attention mechanism, and its strong capabilities have attracted the attention of computer vision researchers who apply this advanced neural network architecture to visual data as it can handle long-range dependencies between input sequence elements. In this study, we developed an automatic hybrid face mask detection model that is a combination of a transformer neural network and a convolutional neural network models which can be used to detect and determine whether people are wearing face masks. The proposed hybrid model's performance was evaluated and compared to other state-of-the-art face mask detection models, and the experimental results proved the proposed model's ability to achieve a highest average precision of 89.4% with an execution time of 2.8 s. Thus, the proposed hybrid model is fit for a practical, real-time trial and can contribute towards public healthcare in terms of infectious disease control.
| S-EPMC10280424 | biostudies-literature
Project description:BackgroundCervical cancer remains a critical global health issue, responsible for over 600,000 new cases and 300,000 deaths annually. Pathological imaging of cervical cancer is a crucial diagnostic tool. However, distinguishing specific areas of cellular differentiation remains challenging because of the lack of clear boundaries between cells at various stages of differentiation. To address the limitations of conventional clinical and deep learning (DL) methods, we developed a mobile attention classification network (MacNet) with multiscale features, aiming to increase the accuracy of differentiation classification and quantitatively analyze cervical cancer cell differentiation.MethodsWe investigated the application of MacNet for classifying non-background images into 3 stages of cervical cancer differentiation. The feature maps are processed through the Mobile Convolution Neural Network with Mobile Attention (MCMA) module, which integrates mobile convolutional blocks and mobile attention blocks. MacNet harnesses the benefits of the image pyramid structure and self-attention mechanism, enabling multiscale feature extraction and emulation of clinical pathologist analysis. The final prediction is generated by the adaptive fusion module, which aggregates features into a unified output.ResultsComparative evaluations demonstrated that MacNet outperforms existing models. The proposed method achieved the best classification accuracy of 92.34% among all 7 DL-based models. Specifically, the result achieved by MacNet was 2.62% greater than that of Inception Version 3, 7.9% greater than that of vision transformer, 8.08% greater than that of the visual geometry group network, 3.21% greater than that of Densely Connected Convolutional Network, 2.85% greater than that of shifted window transformer (Swin transformer), 5.4% greater than that of Cross Stage Partial DarkNet, and 5.41% greater than that of Residual Neural Network. At the same time, MacNet also achieved superior results in recall, precision, and F1 score.ConclusionsWe have proposed a lightweight neural network method that innovatively combines attention mechanisms with convolutional neural networks (CNNs) to efficiently utilize multiscale information from histopathological images. This integration enables the precise quantitative display of different stages of differentiation in cervical cancer. By doing so, our method not only enhances diagnostic accuracy but also provides clinicians with a more effective tool for faster and more reliable diagnosis, representing a significant advancement in the field of pathological imaging.
| S-EPMC11744112 | biostudies-literature
Project description:Deep neural networks can improve the quality of fluorescence microscopy images. Previous methods, based on Convolutional Neural Networks (CNNs), require time-consuming training of individual models for each experiment, impairing their applicability and generalization. In this study, we propose a novel imaging-transformer based model, Convolutional Neural Network Transformer (CNNT), that outperforms CNN based networks for image denoising. We train a general CNNT based backbone model from pairwise high-low Signal-to-Noise Ratio (SNR) image volumes, gathered from a single type of fluorescence microscope, an instant Structured Illumination Microscope. Fast adaptation to new microscopes is achieved by fine-tuning the backbone on only 5-10 image volume pairs per new experiment. Results show that the CNNT backbone and fine-tuning scheme significantly reduces training time and improves image quality, outperforming models trained using only CNNs such as 3D-RCAN and Noise2Fast. We show three examples of efficacy of this approach in wide-field, two-photon, and confocal fluorescence microscopy.
| S-EPMC11303381 | biostudies-literature
Project description:Deep neural networks have been applied to improve the image quality of fluorescence microscopy imaging. Previous methods are based on convolutional neural networks (CNNs) which generally require more time-consuming training of separate models for each new imaging experiment, impairing the applicability and generalization. Once the model is trained (typically with tens to hundreds of image pairs) it can then be used to enhance new images that are like the training data. In this study, we proposed a novel imaging-transformer based model, Convolutional Neural Network Transformer (CNNT), to outperform the CNN networks for image denoising. In our scheme we have trained a single CNNT based "backbone model" from pairwise high-low SNR images for one type of fluorescence microscope (instance structured illumination, iSim). Fast adaption to new applications was achieved by fine-tuning the backbone on only 5-10 sample pairs per new experiment. Results show the CNNT backbone and fine-tuning scheme significantly reduces the training time and improves the image quality, outperformed training separate models using CNN approaches such as - RCAN and Noise2Fast. Here we show three examples of the efficacy of this approach on denoising wide-field, two-photon and confocal fluorescence data. In the confocal experiment, which is a 5×5 tiled acquisition, the fine-tuned CNNT model reduces the scan time form one hour to eight minutes, with improved quality.
| S-EPMC11188127 | biostudies-literature
Project description:PurposeAccurate segmentation of liver and liver tumors is critical for radiotherapy. Liver tumor segmentation, however, remains a difficult and relevant problem in the field of medical image processing because of the various factors like complex and variable location, size, and shape of liver tumors, low contrast between tumors and normal tissues, and blurred or difficult-to-define lesion boundaries. In this paper, we proposed a neural network (S-Net) that can incorporate attention mechanisms to end-to-end segmentation of liver tumors from CT images.MethodsFirst, this study adopted a classical coding-decoding structure to realize end-to-end segmentation. Next, we introduced an attention mechanism between the contraction path and the expansion path so that the network could encode a longer range of semantic information in the local features and find the corresponding relationship between different channels. Then, we introduced long-hop connections between the layers of the contraction path and the expansion path, so that the semantic information extracted in both paths could be fused. Finally, the application of closed operation was used to dissipate the narrow interruptions and long, thin divide. This eliminated small cavities and produced a noise reduction effect.ResultsIn this paper, we used the MICCAI 2017 liver tumor segmentation (LiTS) challenge dataset, 3DIRCADb dataset and doctors' manual contours of Hubei Cancer Hospital dataset to test the network architecture. We calculated the Dice Global (DG) score, Dice per Case (DC) score, volumetric overlap error (VOE), average symmetric surface distance (ASSD), and root mean square error (RMSE) to evaluate the accuracy of the architecture for liver tumor segmentation. The segmentation DG for tumor was found to be 0.7555, DC was 0.613, VOE was 0.413, ASSD was 1.186 and RMSE was 1.804. For a small tumor, DG was 0.3246 and DC was 0.3082. For a large tumor, DG was 0.7819 and DC was 0.7632.ConclusionS-Net obtained more semantic information with the introduction of an attention mechanism and long jump connection. Experimental results showed that this method effectively improved the effect of tumor recognition in CT images and could be applied to assist doctors in clinical treatment.
| S-EPMC8381250 | biostudies-literature
Project description:BackgroundUltrasound (US) is widely used in the clinical diagnosis of thyroid nodules. Artificial intelligence-powered US is becoming an important issue in the research community. This study aimed to develop an improved deep learning model-based algorithm to classify benign and malignant thyroid nodules (TNs) using thyroid US images.MethodsIn total, 592 patients with 600 TNs were included in the internal training, validation, and testing data set; 187 patients with 200 TNs were recruited for the external test data set. We developed a Visual Geometry Group (VGG)-16T model, based on the VGG-16 architecture, but with additional batch normalization (BN) and dropout layers in addition to the fully connected layers. We conducted a 10-fold cross-validation to analyze the performance of the VGG-16T model using a data set of gray-scale US images from 5 different brands of US machines.ResultsFor the internal data set, the VGG-16T model had 87.43% sensitivity, 85.43% specificity, and 86.43% accuracy. For the external data set, the VGG-16T model achieved an area under the curve (AUC) of 0.829 [95% confidence interval (CI): 0.770-0.879], a radiologist with 15 years' working experience achieved an AUC of 0.705 (95% CI: 0.659-0.801), a radiologist with 10 years' experience achieved an AUC of 0.725 (95% CI: 0.653-0.797), and a radiologist with 5 years' experience achieved an AUC of 0.660 (95% CI: 0.584-0.736).ConclusionsThe VGG-16T model had high specificity, sensitivity, and accuracy in differentiating between malignant and benign TNs. Its diagnostic performance was superior to that of experienced radiologists. Thus, the proposed improved deep-learning model can assist radiologists to diagnose thyroid cancer.
| S-EPMC8576712 | biostudies-literature
Project description:Deep convolutional neural networks (DCNN) achieve very high accuracy in segmenting various anatomical structures in medical images but often suffer from relatively poor generalizability. Multi-atlas segmentation (MAS), while less accurate than DCNN in many applications, tends to generalize well to unseen datasets with different characteristics from the training dataset. Several groups have attempted to integrate the power of DCNN to learn complex data representations and the robustness of MAS to changes in image characteristics. However, these studies primarily focused on replacing individual components of MAS with DCNN models and reported marginal improvements in accuracy. In this study we describe and evaluate a 3D end-to-end hybrid MAS and DCNN segmentation pipeline, called Deep Label Fusion (DLF). The DLF pipeline consists of two main components with learnable weights, including a weighted voting subnet that mimics the MAS algorithm and a fine-tuning subnet that corrects residual segmentation errors to improve final segmentation accuracy. We evaluate DLF on five datasets that represent a diversity of anatomical structures (medial temporal lobe subregions and lumbar vertebrae) and imaging modalities (multi-modality, multi-field-strength MRI and Computational Tomography). These experiments show that DLF achieves comparable segmentation accuracy to nnU-Net (Isensee et al., 2020), the state-of-the-art DCNN pipeline, when evaluated on a dataset with similar characteristics to the training datasets, while outperforming nnU-Net on tasks that involve generalization to datasets with different characteristics (different MRI field strength or different patient population). DLF is also shown to consistently improve upon conventional MAS methods. In addition, a modality augmentation strategy tailored for multimodal imaging is proposed and demonstrated to be beneficial in improving the segmentation accuracy of learning-based methods, including DLF and DCNN, in missing data scenarios in test time as well as increasing the interpretability of the contribution of each individual modality.
| S-EPMC10009820 | biostudies-literature