Vessel Wall Characterization Using Quantitative MR Imaging.

MR imaging's exceptional capabilities in vascular imaging stem from its ability to visualize and quantify vessel wall features, such as plaque burden, composition, and biomechanical properties. The application of advanced MR imaging techniques, including two-dimensional and three-dimensional black-blood MR imaging, T1 and T2 relaxometry, diffusion-weighted imaging, and dynamic contrast-enhanced MR imaging, wall shear stress, and arterial stiffness, empowers clinicians and researchers to explore the intricacies of vascular diseases. This array of techniques provides comprehensive insights into the development and progression of vascular pathologies, facilitating earlier diagnosis, targeted treatment, and improved patient outcomes in the management of vascular health.

Attention-based deep learning framework for automatic fundus image processing to aid in diabetic retinopathy grading.

BACKGROUND AND OBJECTIVE: Early detection and grading of Diabetic Retinopathy (DR) is essential to determine an adequate treatment and prevent severe vision loss. However, the manual analysis of fundus images is time consuming and DR screening programs are challenged by the availability of human graders. Current automatic approaches for DR grading attempt the joint detection of all signs at the same time. However, the classification can be optimized if red lesions and bright lesions are independently processed since the task gets divided and simplified. Furthermore, clinicians would greatly benefit from explainable artificial intelligence (XAI) to support the automatic model predictions, especially when the type of lesion is specified. As a novelty, we propose an end-to-end deep learning framework for automatic DR grading (5 severity degrees) based on separating the attention of the dark structures from the bright structures of the retina. As the main contribution, this approach allowed us to generate independent interpretable attention maps for red lesions, such as microaneurysms and hemorrhages, and bright lesions, such as hard exudates, while using image-level labels only. METHODS: Our approach is based on a novel attention mechanism which focuses separately on the dark and the bright structures of the retina by performing a previous image decomposition. This mechanism can be seen as a XAI approach which generates independent attention maps for red lesions and bright lesions. The framework includes an image quality assessment stage and deep learning-related techniques, such as data augmentation, transfer learning and fine-tuning. We used the architecture Xception as a feature extractor and the focal loss function to deal with data imbalance. RESULTS: The Kaggle DR detection dataset was used for method development and validation. The proposed approach achieved 83.7 % accuracy and a Quadratic Weighted Kappa of 0.78 to classify DR among 5 severity degrees, which outperforms several state-of-the-art approaches. Nevertheless, the main result of this work is the generated attention maps, which reveal the pathological regions on the image distinguishing the red lesions and the bright lesions. These maps provide explainability to the model predictions. CONCLUSIONS: Our results suggest that our framework is effective to automatically grade DR. The separate attention approach has proven useful for optimizing the classification. On top of that, the obtained attention maps facilitate visual interpretation for clinicians. Therefore, the proposed method could be a diagnostic aid for the early detection and grading of DR.

Deep learning supported echocardiogram analysis: A comprehensive review.

An echocardiogram is a sophisticated ultrasound imaging technique employed to diagnose heart conditions. The transthoracic echocardiogram, one of the most prevalent types, is instrumental in evaluating significant cardiac diseases. However, interpreting its results heavily relies on the clinician's expertise. In this context, artificial intelligence has emerged as a vital tool for helping clinicians. This study critically analyzes key state-of-the-art research that uses deep learning techniques to automate transthoracic echocardiogram analysis and support clinical judgments. We have systematically organized and categorized articles that proffer solutions for view classification, enhancement of image quality and dataset, segmentation and identification of cardiac structures, detection of cardiac function abnormalities, and quantification of cardiac functions. We compared the performance of various deep learning approaches within each category, identifying the most promising methods. Additionally, we highlight limitations in current research and explore promising avenues for future exploration. These include addressing generalizability issues, incorporating novel AI approaches, and tackling the analysis of rare cardiac diseases.

Quantitative characterization of breast lesions and normal fibroglandular tissue using compartmentalized diffusion-weighted model: comparison of intravoxel incoherent motion and restriction spectrum imaging.

BACKGROUND: To compare the compartmentalized diffusion-weighted models, intravoxel incoherent motion (IVIM) and restriction spectrum imaging (RSI), in characterizing breast lesions and normal fibroglandular tissue. METHODS: This prospective study enrolled 152 patients with 157 histopathologically verified breast lesions (41 benign and 116 malignant). All patients underwent a full-protocol preoperative breast MRI, including a multi-b-value DWI sequence. The diffusion parameters derived from the mono-exponential model (ADC), IVIM model (Dt, Dp, f), and RSI model (C1, C2, C3, C1C2, F1, F2, F3, F1F2) were quantitatively measured and then compared among malignant lesions, benign lesions and normal fibroglandular tissues using Kruskal-Wallis test. The Mann-Whitney U-test was used for the pairwise comparisons. Diagnostic models were built by logistic regression analysis. The ROC analysis was performed using five-fold cross-validation and the mean AUC values were calculated and compared to evaluate the discriminative ability of each parameter or model. RESULTS: Almost all quantitative diffusion parameters showed significant differences in distinguishing malignant breast lesions from both benign lesions (other than C2) and normal fibroglandular tissue (all parameters) (all P < 0.0167). In terms of the comparisons of benign lesions and normal fibroglandular tissues, the parameters derived from IVIM (Dp, f) and RSI (C1, C2, C1C2, F1, F2, F3) showed significant differences (all P < 0.005). When using individual parameters, RSI-derived parameters-F1, C1C2, and C2 values yielded the highest AUCs for the comparisons of malignant vs. benign, malignant vs. normal tissue and benign vs. normal tissue (AUCs = 0.871, 0.982, and 0.863, respectively). Furthermore, the combined diagnostic model (IVIM + RSI) exhibited the highest diagnostic efficacy for the pairwise discriminations (AUCs = 0.893, 0.991, and 0.928, respectively). CONCLUSIONS: Quantitative parameters derived from the three-compartment RSI model have great promise as imaging indicators for the differential diagnosis of breast lesions compared with the bi-exponential IVIM model. Additionally, the combined model of IVIM and RSI achieves superior diagnostic performance in characterizing breast lesions.

A novel SpaSA based hyper-parameter optimized FCEDN with adaptive CNN classification for skin cancer detection.

Skin cancer is the most prevalent kind of cancer in people. It is estimated that more than 1 million people get skin cancer every year in the world. The effectiveness of the disease's therapy is significantly impacted by early identification of this illness. Preprocessing is the initial detecting stage in enhancing the quality of skin images by removing undesired background noise and objects. This study aims is to compile preprocessing techniques for skin cancer imaging that are currently accessible. Researchers looking into automated skin cancer diagnosis might use this article as an excellent place to start. The fully convolutional encoder-decoder network and Sparrow search algorithm (FCEDN-SpaSA) are proposed in this study for the segmentation of dermoscopic images. The individual wolf method and the ensemble ghosting technique are integrated to generate a neighbour-based search strategy in SpaSA for stressing the correct balance between navigation and exploitation. The classification procedure is accomplished by using an adaptive CNN technique to discriminate between normal skin and malignant skin lesions suggestive of disease. Our method provides classification accuracies comparable to commonly used incremental learning techniques while using less energy, storage space, memory access, and training time (only network updates with new training samples, no network sharing). In a simulation, the segmentation performance of the proposed technique on the ISBI 2017, ISIC 2018, and PH2 datasets reached accuracies of 95.28%, 95.89%, 92.70%, and 98.78%, respectively, on the same dataset and assessed the classification performance. It is accurate 91.67% of the time. The efficiency of the suggested strategy is demonstrated through comparisons with cutting-edge methodologies.

Imaging segmentation mechanism for rectal tumors using improved U-Net.

OBJECTIVE: In radiation therapy, cancerous region segmentation in magnetic resonance images (MRI) is a critical step. For rectal cancer, the automatic segmentation of rectal tumors from an MRI is a great challenge. There are two main shortcomings in existing deep learning-based methods that lead to incorrect segmentation: 1) there are many organs surrounding the rectum, and the shape of some organs is similar to that of rectal tumors; 2) high-level features extracted by conventional neural networks often do not contain enough high-resolution information. Therefore, an improved U-Net segmentation network based on attention mechanisms is proposed to replace the traditional U-Net network. METHODS: The overall framework of the proposed method is based on traditional U-Net. A ResNeSt module was added to extract the overall features, and a shape module was added after the encoder layer. We then combined the outputs of the shape module and the decoder to obtain the results. Moreover, the model used different types of attention mechanisms, so that the network learned information to improve segmentation accuracy. RESULTS: We validated the effectiveness of the proposed method using 3773 2D MRI datasets from 304 patients. The results showed that the proposed method achieved 0.987, 0.946, 0.897, and 0.899 for Dice, MPA, MioU, and FWIoU, respectively; these values are significantly better than those of other existing methods. CONCLUSION: Due to time savings, the proposed method can help radiologists segment rectal tumors effectively and enable them to focus on patients whose cancerous regions are difficult for the network to segment. SIGNIFICANCE: The proposed method can help doctors segment rectal tumors, thereby ensuring good diagnostic quality and accuracy.

Dual path parallel hierarchical diagnosis model for intracranial tumors based on multi-feature entropy weight.

The grading diagnosis of intracranial tumors is a key step in formulating clinical treatment plans and surgical guidelines. To effectively grade the diagnosis of intracranial tumors, this paper proposes a dual path parallel hierarchical model that can automatically grade the diagnosis of intracranial tumors with high accuracy. In this model, prior features of solid tumor mass and intratumoral necrosis are extracted. Then the optimal division of the data set is achieved through multi-feature entropy weight. The multi-modal input is realized by the dual path structure. Multiple features are superimposed and fused to achieve the image grading. The model has been tested on the actual clinical medical images provided by the Second Affiliated Hospital of Dalian Medical University. The experiment shows that the proposed model has good generalization ability, with an accuracy of 0.990. The proposed model can be applied to clinical diagnosis and has practical application prospects.

High-resolution free-breathing hepatobiliary phase MRI of the liver using XD-GRASP.

PURPOSE: To evaluate the performance of high-resolution free-breathing (FB) hepatobiliary phase imaging of the liver using the eXtra-Dimension Golden-angle RAdial Sparse Parallel (XD-GRASP) MRI technique. METHODS: Fifty-eight clinical patients (41 males, mean age = 52.9 ± 12.9) with liver lesions who underwent dynamic contrast-enhanced MRI with a liver-specific contrast agent were prospectively recruited for this study. Both breath-hold volumetric interpolated examination (BH-VIBE) imaging and FB imaging were performed during the hepatobiliary phase. FB images were acquired using a stack-of-stars golden-angle radial sequence and were reconstructed using the XD-GRASP method. Two experienced radiologists blinded to acquisition schemes independently scored the overall image quality, liver edge sharpness, hepatic vessel clarity, conspicuity of lesion, and overall artifact level of each image. The non-parametric paired two-tailed Wilcoxon signed-rank test was used for statistical analysis. RESULTS: Compared to BH-VIBE images, XD-GRASP images received significantly higher scores (P < 0.05) for the liver edge sharpness (4.83 ± 0.45 vs 4.29 ± 0.46), the hepatic vessel clarity (4.64 ± 0.67 vs 4.15 ± 0.56) and the conspicuity of lesion (4.75 ± 0.53 vs 4.31 ± 0.50). There were no significant differences (P > 0.05) between BH-VIBE and XD-GRASP images for the overall image quality (4.61 ± 0.50 vs 4.74 ± 0.47) and the overall artifact level (4.13 ± 0.44 vs 4.05 ± 0.61). CONCLUSION: Compared to conventional BH-VIBE MRI, FB radial acquisition combined with XD-GRASP reconstruction facilitates higher spatial resolution imaging of the liver during the hepatobiliary phase. This enhancement can significantly improve the visualization and evaluation of the liver.

An interpretable dual attention network for diabetic retinopathy grading: IDANet.

Diabetic retinopathy (DR) is the most prevalent cause of visual impairment in adults worldwide. Typically, patients with DR do not show symptoms until later stages, by which time it may be too late to receive effective treatment. DR Grading is challenging because of the small size and variation in lesion patterns. The key to fine-grained DR grading is to discover more discriminating elements such as cotton wool, hard exudates, hemorrhages, microaneurysms etc. Although deep learning models like convolutional neural networks (CNN) seem ideal for the automated detection of abnormalities in advanced clinical imaging, small-size lesions are very hard to distinguish by using traditional networks. This work proposes a bi-directional spatial and channel-wise parallel attention based network to learn discriminative features for diabetic retinopathy grading. The proposed attention block plugged with a backbone network helps to extract features specific to fine-grained DR-grading. This scheme boosts classification performance along with the detection of small-sized lesion parts. Extensive experiments are performed on four widely used benchmark datasets for DR grading, and performance is evaluated on different quality metrics. Also, for model interpretability, activation maps are generated using the LIME method to visualize the predicted lesion parts. In comparison with state-of-the-art methods, the proposed IDANet exhibits better performance for DR grading and lesion detection.

DctViT: Discrete Cosine Transform meet vision transformers.

Vision transformers (ViTs) have become one of the dominant frameworks for vision tasks in recent years because of their ability to efficiently capture long-range dependencies in image recognition tasks using self-attention. In fact, both CNNs and ViTs have advantages and disadvantages in vision tasks, and some studies suggest that the use of both may be an effective way to balance performance and computational cost. In this paper, we propose a new hybrid network based on CNN and transformer, using CNN to extract local features and transformer to capture long-distance dependencies. We also proposed a new feature map resolution reduction based on Discrete Cosine Transform and self-attention, named DCT-Attention Down-sample (DAD). Our DctViT-L achieves 84.8% top-1 accuracy on ImageNet 1K, far outperforming CMT, Next-ViT, SpectFormer and other state-of-the-art models, with lower computational costs. Using DctViT-B as the backbone, RetinaNet can achieve 46.8% mAP on COCO val2017, which improves mAP by 2.5% and 1.1% with less calculation cost compared with CMT-S and SpectFormer as the backbone.

Automated curation of large-scale cancer histopathology image datasets using deep learning.

BACKGROUND: Artificial intelligence (AI) has numerous applications in pathology, supporting diagnosis and prognostication in cancer. However, most AI models are trained on highly selected data, typically one tissue slide per patient. In reality, especially for large surgical resection specimens, dozens of slides can be available for each patient. Manually sorting and labelling whole-slide images (WSIs) is a very time-consuming process, hindering the direct application of AI on the collected tissue samples from large cohorts. In this study we addressed this issue by developing a deep-learning (DL)-based method for automatic curation of large pathology datasets with several slides per patient. METHODS: We collected multiple large multicentric datasets of colorectal cancer histopathological slides from the United Kingdom (FOXTROT, N = 21,384 slides; CR07, N = 7985 slides) and Germany (DACHS, N = 3606 slides). These datasets contained multiple types of tissue slides, including bowel resection specimens, endoscopic biopsies, lymph node resections, immunohistochemistry-stained slides, and tissue microarrays. We developed, trained, and tested a deep convolutional neural network model to predict the type of slide from the slide overview (thumbnail) image. The primary statistical endpoint was the macro-averaged area under the receiver operating curve (AUROCs) for detection of the type of slide. RESULTS: In the primary dataset (FOXTROT), with an AUROC of 0.995 [95% confidence interval [CI]: 0.994-0.996] the algorithm achieved a high classification performance and was able to accurately predict the type of slide from the thumbnail image alone. In the two external test cohorts (CR07, DACHS) AUROCs of 0.982 [95% CI: 0.979-0.985] and 0.875 [95% CI: 0.864-0.887] were observed, which indicates the generalizability of the trained model on unseen datasets. With a confidence threshold of 0.95, the model reached an accuracy of 94.6% (7331 classified cases) in CR07 and 85.1% (2752 classified cases) for the DACHS cohort. CONCLUSION: Our findings show that using the low-resolution thumbnail image is sufficient to accurately classify the type of slide in digital pathology. This can support researchers to make the vast resource of existing pathology archives accessible to modern AI models with only minimal manual annotations.

Intra-patient and inter-observer image quality analysis in liver MRI study with gadoxetic acid using two different multi-arterial phase techniques.

PURPOSE: To evaluate intra-patient and interobserver agreement in patients who underwent liver MRI with gadoxetic acid using two different multi-arterial phase (AP) techniques. METHODS: A total of 154 prospectively enrolled patients underwent clinical gadoxetic acid-enhanced liver MRI twice within 12 months, using two different multi-arterial algorithms: CAIPIRINHA-VIBE and TWIST-VIBE. For every patient, breath-holding time, body mass index, sex, age were recorded. The phase without contrast media and the APs were independently evaluated by two radiologists who quantified Gibbs artefacts, noise, respiratory motion artefacts, and general image quality. Presence or absence of Gibbs artefacts and noise was compared by the McNemar's test. Respiratory motion artefacts and image quality scores were compared using Wilcoxon signed rank test. Interobserver agreement was assessed by Cohen kappa statistics. RESULTS: Compared with TWIST-VIBE, CAIPIRINHA-VIBE images had better scores for every parameter except higher noise score. Triple APs were always acquired with TWIST-VIBE but failed in 37% using CAIPIRINHA-VIBE: 11% have only one AP, 26% have two. Breath-holding time was the only parameter that influenced the success of multi-arterial techniques. TWIST-VIBE images had worst score for Gibbs and respiratory motion artefacts but lower noise score. CONCLUSION: CAIPIRINHA-VIBE images were always diagnostic, but with a failure of triple-AP in 37%. TWIST-VIBE was successful in obtaining three APs in all patients. Breath-holding time is the only parameter which can influence the preliminary choice between CAIPIRINHA-VIBE and TWIST-VIBE algorithm. ADVANCES IN KNOWLEDGE: If the patient is expected to perform good breath-holds, TWIST-VIBE is preferable; otherwise, CAIPIRINHA-VIBE is more appropriate.

Ultrashort VIBE-Dixon versus Cine and late gadolinium enhancement MRI sequences for cardiac thrombus detection.

PURPOSE: To determine the diagnostic accuracy of volumetric interpolated breath-hold examination sequences with fat suppression in Dixon technique (VIBE-Dixon) for cardiac thrombus detection. METHOD: From our clinical database, we retrospectively identified consecutive patients between 2014 and 2022 who had definite diagnosis or exclusion of cardiac thrombus confirmed by an independent adjudication committee, serving as the reference standard. All patients received 2D-Cine plus 2D-Late-Gadolinium-Enhancement (Cine + LGE) and VIBE-Dixon sequences. Two blinded readers assessed all images for the presence of cardiac thrombus. The diagnostic accuracy of Cine + LGE and VIBE-Dixon was determined and compared. RESULTS: Among 141 MRI studies (116 male, mean age: 61 years) mean image examination time was 28.8 ± 3.1 s for VIBE-Dixon and 23.3 ± 2.5 min for Cine + LGE. Cardiac thrombus was present in 49 patients (prevalence: 35 %). For both readers sensitivity for thrombus detection was significantly higher in VIBE-Dixon compared with Cine + LGE (Reader 1: 96 % vs.73 %, Reader 2: 96 % vs. 78 %, p < 0.01 for both readers), whereas specificity did not differ significantly (Reader 1: 96 % vs. 98 %, Reader 2: 92 % vs. 93 %, p > 0.1). Overall diagnostic accuracy of VIBE-Dixon was higher than for Cine + LGE (95 % vs. 89 %, p = 0.02) and was non-inferior to the reference standard (Delta ≤ 5 % with probability > 95 %). CONCLUSIONS: Biplanar VIBE-Dixon sequences, acquired within a few seconds, provided a very high diagnostic accuracy for cardiac thrombus detection. They could be used as stand-alone sequences to rapidly screen for cardiac thrombus in patients not amenable to lengthy acquisition times.

Multimodal Action Quality Assessment.

Action quality assessment (AQA) is to assess how well an action is performed. Previous works perform modelling by only the use of visual information, ignoring audio information. We argue that although AQA is highly dependent on visual information, the audio is useful complementary information for improving the score regression accuracy, especially for sports with background music, such as figure skating and rhythmic gymnastics. To leverage multimodal information for AQA, i.e., RGB, optical flow and audio information, we propose a Progressive Adaptive Multimodal Fusion Network (PAMFN) that separately models modality-specific information and mixed-modality information. Our model consists of with three modality-specific branches that independently explore modality-specific information and a mixed-modality branch that progressively aggregates the modality-specific information from the modality-specific branches. To build the bridge between modality-specific branches and the mixed-modality branch, three novel modules are proposed. First, a Modality-specific Feature Decoder module is designed to selectively transfer modality-specific information to the mixed-modality branch. Second, when exploring the interaction between modality-specific information, we argue that using an invariant multimodal fusion policy may lead to suboptimal results, so as to take the potential diversity in different parts of an action into consideration. Therefore, an Adaptive Fusion Module is proposed to learn adaptive multimodal fusion policies in different parts of an action. This module consists of several FusionNets for exploring different multimodal fusion strategies and a PolicyNet for deciding which FusionNets are enabled. Third, a module called Cross-modal Feature Decoder is designed to transfer cross-modal features generated by Adaptive Fusion Module to the mixed-modality branch. Our extensive experiments validate the efficacy of the proposed method, and our method achieves state-of-the-art performance on two public datasets. Code is available at https://github.com/qinghuannn/PAMFN.

Sidebands in CEST MR-How to recognize and avoid them.

PURPOSE: Clinical scanners require pulsed CEST sequences to maintain amplifier and specific absorption rate limits. During off-resonant RF irradiation and interpulse delay, the magnetization can accumulate specific relative phases within the pulse train. In this work, we show that these phases are important to consider, as they can lead to unexpected artifacts when no interpulse gradient spoiling is performed during the saturation train. METHODS: We investigated sideband artifacts using a CEST-3D snapshot gradient-echo sequence at 3 T. Initially, Bloch-McConnell simulations were carried out with Pulseq-CEST, while measurements were performed in vitro and in vivo. RESULTS: Sidebands can be hidden in Z-spectra, and their structure becomes clearly visible only at high sampling. Sidebands are further influenced by B0 inhomogeneities and the RF phase cycling within the pulse train. In vivo, sidebands are mostly visible in liquid compartments such as CSF. Multi-pulse sidebands can be suppressed by interpulse gradient spoiling. CONCLUSION: We provide new insights into sidebands occurring in pulsed CEST experiments and show that, similar as in imaging sequences, gradient and RF spoiling play an important role. Gradient spoiling avoids misinterpretations of sidebands as CEST effects especially in liquid environments including pathological tissue or for CEST resonances close to water. It is recommended to simulate pulsed CEST sequences in advance to avoid artifacts.

A reliable diabetic retinopathy grading via transfer learning and ensemble learning with quadratic weighted kappa metric.

The most common eye infection in people with diabetes is diabetic retinopathy (DR). It might cause blurred vision or even total blindness. Therefore, it is essential to promote early detection to prevent or alleviate the impact of DR. However, due to the possibility that symptoms may not be noticeable in the early stages of DR, it is difficult for doctors to identify them. Therefore, numerous predictive models based on machine learning (ML) and deep learning (DL) have been developed to determine all stages of DR. However, existing DR classification models cannot classify every DR stage or use a computationally heavy approach. Common metrics such as accuracy, F1 score, precision, recall, and AUC-ROC score are not reliable for assessing DR grading. This is because they do not account for two key factors: the severity of the discrepancy between the assigned and predicted grades and the ordered nature of the DR grading scale. This research proposes computationally efficient ensemble methods for the classification of DR. These methods leverage pre-trained model weights, reducing training time and resource requirements. In addition, data augmentation techniques are used to address data limitations, improve features, and improve generalization. This combination offers a promising approach for accurate and robust DR grading. In particular, we take advantage of transfer learning using models trained on DR data and employ CLAHE for image enhancement and Gaussian blur for noise reduction. We propose a three-layer classifier that incorporates dropout and ReLU activation. This design aims to minimize overfitting while effectively extracting features and assigning DR grades. We prioritize the Quadratic Weighted Kappa (QWK) metric due to its sensitivity to label discrepancies, which is crucial for an accurate diagnosis of DR. This combined approach achieves state-of-the-art QWK scores (0.901, 0.967 and 0.944) in the Eyepacs, Aptos, and Messidor datasets.

Resilient back-propagation machine learning-based classification on fundus images for retinal microaneurysm detection.

BACKGROUND: The timely diagnosis of medical conditions, particularly diabetic retinopathy, relies on the identification of retinal microaneurysms. However, the commonly used retinography method poses a challenge due to the diminutive dimensions and limited differentiation of microaneurysms in images. PROBLEM STATEMENT: Automated identification of microaneurysms becomes crucial, necessitating the use of comprehensive ad-hoc processing techniques. Although fluorescein angiography enhances detectability, its invasiveness limits its suitability for routine preventative screening. OBJECTIVE: This study proposes a novel approach for detecting retinal microaneurysms using a fundus scan, leveraging circular reference-based shape features (CR-SF) and radial gradient-based texture features (RG-TF). METHODOLOGY: The proposed technique involves extracting CR-SF and RG-TF for each candidate microaneurysm, employing a robust back-propagation machine learning method for training. During testing, extracted features from test images are compared with training features to categorize microaneurysm presence. RESULTS: The experimental assessment utilized four datasets (MESSIDOR, Diaretdb1, e-ophtha-MA, and ROC), employing various measures. The proposed approach demonstrated high accuracy (98.01%), sensitivity (98.74%), specificity (97.12%), and area under the curve (91.72%). CONCLUSION: The presented approach showcases a successful method for detecting retinal microaneurysms using a fundus scan, providing promising accuracy and sensitivity. This non-invasive technique holds potential for effective screening in diabetic retinopathy and other related medical conditions.

MTFN: multi-temporal feature fusing network with co-attention for DCE-MRI synthesis.

BACKGROUND: Dynamic Contrast Enhanced Magnetic Resonance Imaging (DCE-MRI) plays an important role in the diagnosis and treatment of breast cancer. However, obtaining complete eight temporal images of DCE-MRI requires a long scanning time, which causes patients' discomfort in the scanning process. Therefore, to reduce the time, the multi temporal feature fusing neural network with Co-attention (MTFN) is proposed to generate the eighth temporal images of DCE-MRI, which enables the acquisition of DCE-MRI images without scanning. In order to reduce the time, multi-temporal feature fusion cooperative attention mechanism neural network (MTFN) is proposed to generate the eighth temporal images of DCE-MRI, which enables DCE-MRI image acquisition without scanning. METHODS: In this paper, we propose multi temporal feature fusing neural network with Co-attention (MTFN) for DCE-MRI Synthesis, in which the Co-attention module can fully fuse the features of the first and third temporal image to obtain the hybrid features. The Co-attention explore long-range dependencies, not just relationships between pixels. Therefore, the hybrid features are more helpful to generate the eighth temporal images. RESULTS: We conduct experiments on the private breast DCE-MRI dataset from hospitals and the multi modal Brain Tumor Segmentation Challenge2018 dataset (BraTs2018). Compared with existing methods, the experimental results of our method show the improvement and our method can generate more realistic images. In the meanwhile, we also use synthetic images to classify the molecular typing of breast cancer that the accuracy on the original eighth time-series images and the generated images are 89.53% and 92.46%, which have been improved by about 3%, and the classification results verify the practicability of the synthetic images. CONCLUSIONS: The results of subjective evaluation and objective image quality evaluation indicators show the effectiveness of our method, which can obtain comprehensive and useful information. The improvement of classification accuracy proves that the images generated by our method are practical.

Diagnosis of Alzheimer's disease via optimized lightweight convolution-attention and structural MRI.

Alzheimer's disease (AD) poses a substantial public health challenge, demanding accurate screening and diagnosis. Identifying AD in its early stages, including mild cognitive impairment (MCI) and healthy control (HC), is crucial given the global aging population. Structural magnetic resonance imaging (sMRI) is essential for understanding the brain's structural changes due to atrophy. While current deep learning networks overlook voxel long-term dependencies, vision transformers (ViT) excel at recognizing such dependencies in images, making them valuable in AD diagnosis. Our proposed method integrates convolution-attention mechanisms in transformer-based classifiers for AD brain datasets, enhancing performance without excessive computing resources. Replacing multi-head attention with lightweight multi-head self-attention (LMHSA), employing inverted residual (IRU) blocks, and introducing local feed-forward networks (LFFN) yields exceptional results. Training on AD datasets with a gradient-centralized optimizer and Adam achieves an impressive accuracy rate of 94.31% for multi-class classification, rising to 95.37% for binary classification (AD vs. HC) and 92.15% for HC vs. MCI. These outcomes surpass existing AD diagnosis approaches, showcasing the model's efficacy. Identifying key brain regions aids future clinical solutions for AD and neurodegenerative diseases. However, this study focused exclusively on the AD Neuroimaging Initiative (ADNI) cohort, emphasizing the need for a more robust, generalizable approach incorporating diverse databases beyond ADNI in future research.