Enhanced Spatial Fuzzy C-Means Algorithm for Brain Tissue Segmentation in T1 Images.

Magnetic Resonance Imaging (MRI) plays an important role in neurology, particularly in the precise segmentation of brain tissues. Accurate segmentation is crucial for diagnosing brain injuries and neurodegenerative conditions. We introduce an Enhanced Spatial Fuzzy C-means (esFCM) algorithm for 3D T1 MRI segmentation to three tissues, i.e. White Matter (WM), Gray Matter (GM), and Cerebrospinal Fluid (CSF). The esFCM employs a weighted least square algorithm utilizing the Structural Similarity Index (SSIM) for polynomial bias field correction. It also takes advantage of the information from the membership function of the last iteration to compute neighborhood impact. This strategic refinement enhances the algorithm's adaptability to complex image structures, effectively addressing challenges such as intensity irregularities and contributing to heightened segmentation accuracy. We compare the segmentation accuracy of esFCM against four variants of FCM, Gaussian Mixture Model (GMM) and FSL and ANTs algorithms using four various dataset, employing three measurement criteria. Comparative assessments underscore esFCM's superior performance, particularly in scenarios involving added noise and bias fields.The obtained results emphasize the significant potential of the proposed method in the segmentation of MRI images.

MICIL: Multiple-Instance Class-Incremental Learning for skin cancer whole slide images.

Artificial intelligence (AI) agents encounter the problem of catastrophic forgetting when they are trained in sequentially with new data batches. This issue poses a barrier to the implementation of AI-based models in tasks that involve ongoing evolution, such as cancer prediction. Moreover, whole slide images (WSI) play a crucial role in cancer management, and their automated analysis has become increasingly popular in assisting pathologists during the diagnosis process. Incremental learning (IL) techniques aim to develop algorithms capable of retaining previously acquired information while also acquiring new insights to predict future data. Deep IL techniques need to address the challenges posed by the gigapixel scale of WSIs, which often necessitates the use of multiple instance learning (MIL) frameworks. In this paper, we introduce an IL algorithm tailored for analyzing WSIs within a MIL paradigm. The proposed Multiple Instance Class-Incremental Learning (MICIL) algorithm combines MIL with class-IL for the first time, allowing for the incremental prediction of multiple skin cancer subtypes from WSIs within a class-IL scenario. Our framework incorporates knowledge distillation and data rehearsal, along with a novel embedding-level distillation, aiming to preserve the latent space at the aggregated WSI level. Results demonstrate the algorithm's effectiveness in addressing the challenge of balancing IL-specific metrics, such as intransigence and forgetting, and solving the plasticity-stability dilemma.

A Swin transformer encoder-based StyleGAN for unbalanced endoscopic image enhancement.

With the rapid development of artificial intelligence, automated endoscopy-assisted diagnostic systems have become an effective tool for reducing the diagnostic costs and shortening the treatment cycle of patients. Typically, the performance of these systems depends on deep learning models which are pre-trained with large-scale labeled data, for example, early gastric cancer based on endoscopic images. However, the expensive annotation and the subjectivity of the annotators lead to an insufficient and class-imbalanced endoscopic image dataset, and these datasets are detrimental to the training of deep learning models. Therefore, we proposed a Swin Transformer encoder-based StyleGAN (STE-StyleGAN) for unbalanced endoscopic image enhancement, which is composed of an adversarial learning encoder and generator. Firstly, a pre-trained Swin Transformer is introduced into the encoder to extract multi-scale features layer by layer from endoscopic images. The features are subsequently fed into a mapping block for aggregation and recombination. Secondly, a self-attention mechanism is applied to the generator, which adds detailed information of the image layer by layer through recoded features, enabling the generator to autonomously learn the coupling between different image regions. Finally, we conducted extensive experiments on a private intestinal metaplasia grading dataset from a Grade-A tertiary hospital. The experimental results show that the images generated by STE-StyleGAN are closer to the initial image distribution, achieving a Fréchet Inception Distance (FID) value of 100.4. Then, these generated images are used to enhance the initial dataset to improve the robustness of the classification model, and achieved a top accuracy of 86 %.

A transformer-based approach empowered by a self-attention technique for semantic segmentation in remote sensing.

Semantic segmentation of Remote Sensing (RS) images involves the classification of each pixel in a satellite image into distinct and non-overlapping regions or segments. This task is crucial in various domains, including land cover classification, autonomous driving, and scene understanding. While deep learning has shown promising results, there is limited research that specifically addresses the challenge of processing fine details in RS images while also considering the high computational demands. To tackle this issue, we propose a novel approach that combines convolutional and transformer architectures. Our design incorporates convolutional layers with a low receptive field to generate fine-grained feature maps for small objects in very high-resolution images. On the other hand, transformer blocks are utilized to capture contextual information from the input. By leveraging convolution and self-attention in this manner, we reduce the need for extensive downsampling and enable the network to work with full-resolution features, which is particularly beneficial for handling small objects. Additionally, our approach eliminates the requirement for vast datasets, which is often necessary for purely transformer-based networks. In our experimental results, we demonstrate the effectiveness of our method in generating local and contextual features using convolutional and transformer layers, respectively. Our approach achieves a mean dice score of 80.41%, outperforming other well-known techniques such as UNet, Fully-Connected Network (FCN), Pyramid Scene Parsing Network (PSP Net), and the recent Convolutional vision Transformer (CvT) model, which achieved mean dice scores of 78.57%, 74.57%, 73.45%, and 62.97% respectively, under the same training conditions and using the same training dataset.

MRUNet-3D: A multi-stride residual 3D UNet for lung nodule segmentation.

Obtaining an accurate segmentation of the pulmonary nodules in computed tomography (CT) images is challenging. This is due to: (1) the heterogeneous nature of the lung nodules; (2) comparable visual characteristics between the nodules and their surroundings. A robust multi-scale feature extraction mechanism that can effectively obtain multi-scale representations at a granular level can improve segmentation accuracy. As the most commonly used network in lung nodule segmentation, UNet, its variants, and other image segmentation methods lack this robust feature extraction mechanism. In this study, we propose a multi-stride residual 3D UNet (MRUNet-3D) to improve the segmentation accuracy of lung nodules in CT images. It incorporates a multi-slide Res2Net block (MSR), which replaces the simple sequence of convolution layers in each encoder stage to effectively extract multi-scale features at a granular level from different receptive fields and resolutions while conserving the strengths of 3D UNet. The proposed method has been extensively evaluated on the publicly available LUNA16 dataset. Experimental results show that it achieves competitive segmentation performance with an average dice similarity coefficient of 83.47 % and an average surface distance of 0.35 mm on the dataset. More notably, our method has proven to be robust to the heterogeneity of lung nodules. It has also proven to perform better at segmenting small lung nodules. Ablation studies have shown that the proposed MSR and RFIA modules are fundamental to improving the performance of the proposed model.

Evaluating ChatGPT-4's Diagnostic Accuracy: Impact of Visual Data Integration.

BACKGROUND: In the evolving field of health care, multimodal generative artificial intelligence (AI) systems, such as ChatGPT-4 with vision (ChatGPT-4V), represent a significant advancement, as they integrate visual data with text data. This integration has the potential to revolutionize clinical diagnostics by offering more comprehensive analysis capabilities. However, the impact on diagnostic accuracy of using image data to augment ChatGPT-4 remains unclear. OBJECTIVE: This study aims to assess the impact of adding image data on ChatGPT-4's diagnostic accuracy and provide insights into how image data integration can enhance the accuracy of multimodal AI in medical diagnostics. Specifically, this study endeavored to compare the diagnostic accuracy between ChatGPT-4V, which processed both text and image data, and its counterpart, ChatGPT-4, which only uses text data. METHODS: We identified a total of 557 case reports published in the American Journal of Case Reports from January 2022 to March 2023. After excluding cases that were nondiagnostic, pediatric, and lacking image data, we included 363 case descriptions with their final diagnoses and associated images. We compared the diagnostic accuracy of ChatGPT-4V and ChatGPT-4 without vision based on their ability to include the final diagnoses within differential diagnosis lists. Two independent physicians evaluated their accuracy, with a third resolving any discrepancies, ensuring a rigorous and objective analysis. RESULTS: The integration of image data into ChatGPT-4V did not significantly enhance diagnostic accuracy, showing that final diagnoses were included in the top 10 differential diagnosis lists at a rate of 85.1% (n=309), comparable to the rate of 87.9% (n=319) for the text-only version (P=.33). Notably, ChatGPT-4V's performance in correctly identifying the top diagnosis was inferior, at 44.4% (n=161), compared with 55.9% (n=203) for the text-only version (P=.002, χ2 test). Additionally, ChatGPT-4's self-reports showed that image data accounted for 30% of the weight in developing the differential diagnosis lists in more than half of cases. CONCLUSIONS: Our findings reveal that currently, ChatGPT-4V predominantly relies on textual data, limiting its ability to fully use the diagnostic potential of visual information. This study underscores the need for further development of multimodal generative AI systems to effectively integrate and use clinical image data. Enhancing the diagnostic performance of such AI systems through improved multimodal data integration could significantly benefit patient care by providing more accurate and comprehensive diagnostic insights. Future research should focus on overcoming these limitations, paving the way for the practical application of advanced AI in medicine.

Three-dimensional biphase fabric estimation from 2D images by deep learning.

A pruned VGG19 model subjected to Axial Coronal Sagittal (ACS) convolutions and a custom VGG16 model are benchmarked to predict 3D fabric descriptors from a set of 2D images. The data used for training and testing are extracted from a set of 600 3D biphase microstructures created numerically. Fabric descriptors calculated from the 3D microstructures constitute the ground truth, while the input data are obtained by slicing the 3D microstructures in each direction of space at regular intervals. The computational cost to train the custom ACS-VGG19 model increases linearly with p (the number of images extracted in each direction of space), and increasing p does not improve the performance of the model - or only does so marginally. The best performing ACS-VGG19 model provides a MAPE of 2 to 5% for the means of aggregate size, aspect ratios and solidity, but cannot be used to estimate orientations. The custom VGG16 yields a MAPE of 2% or less for the means of aggregate size, distance to nearest neighbor, aspect ratios and solidity. The MAPE is less than 3% for the mean roundness, and in the range of 5-7% for the aggregate volume fraction and the mean diagonal components of the orientation matrix. Increasing p improves the performance of the custom VGG16 model, but becomes cost ineffective beyond 3 images per direction. For both models, the aggregate volume fraction is predicted with less accuracy than higher order descriptors, which is attributed to the bias given by the loss function towards highly-correlated descriptors. Both models perform better to predict means than standard deviations, which are noisy quantities. The custom VGG16 model performs better than the pruned version of the ACS-VGG19 model, likely because it contains 3 times (p = 1) to 28 times (p = 10) less parameters than the ACS-VGG19 model, allowing better and faster cnvergence, with less data. The custom VGG16 model predicts the second and third invariants of the orientation matrix with a MAPE of 2.8% and 8.9%, respectively, which suggests that the model can predict orientation descriptors regardless of the orientation of the input images.

Infrared images help to identify retinal emboli in hemodynamic retinal artery occlusion.

To report a case of stroke with internal carotid artery (ICA) occlusion and retinal artery occlusion (RAO) with multiple emboli identified by infrared images. The patient presented with acute blurred vision and slurred speech since woke up in the morning. Carotid Doppler ultrasonography showed severe left ICA stenosis with the decreased flow. Computed tomography angiography demonstrated left ICA critical stenosis with compensatory flows from the right ICA. There was coexistent RAO with multiple emboli, silver-wiring, and segmentation of the retinal artery. More hyperreflective emboli were uncovered by infrared images of spectral-domain optical coherence tomography. Embolism may accompany with the compensatory flow for ICA occlusion. RAO patients should have thorough carotid evaluations, especially those with multiple retinal emboli.

Development of a multi-fusion convolutional neural network (MF-CNN) for enhanced gastrointestinal disease diagnosis in endoscopy image analysis.

Gastrointestinal (GI) diseases are prevalent medical conditions that require accurate and timely diagnosis for effective treatment. To address this, we developed the Multi-Fusion Convolutional Neural Network (MF-CNN), a deep learning framework that strategically integrates and adapts elements from six deep learning models, enhancing feature extraction and classification of GI diseases from endoscopic images. The MF-CNN architecture leverages truncated and partially frozen layers from existing models, augmented with novel components such as Auxiliary Fusing Layers (AuxFL), Fusion Residual Block (FuRB), and Alpha Dropouts (αDO) to improve precision and robustness. This design facilitates the precise identification of conditions such as ulcerative colitis, polyps, esophagitis, and healthy colons. Our methodology involved preprocessing endoscopic images sourced from open databases, including KVASIR and ETIS-Larib Polyp DB, using adaptive histogram equalization (AHE) to enhance their quality. The MF-CNN framework supports detailed feature mapping for improved interpretability of the model's internal workings. An ablation study was conducted to validate the contribution of each component, demonstrating that the integration of AuxFL, αDO, and FuRB played a crucial part in reducing overfitting and efficiency saturation and enhancing overall model performance. The MF-CNN demonstrated outstanding performance in terms of efficacy, achieving an accuracy rate of 99.25%. It also excelled in other key performance metrics with a precision of 99.27%, a recall of 99.25%, and an F1-score of 99.25%. These metrics confirmed the model's proficiency in accurate classification and its capability to minimize false positives and negatives across all tested GI disease categories. Furthermore, the AUC values were exceptional, averaging 1.00 for both test and validation sets, indicating perfect discriminative ability. The findings of the P-R curve analysis and confusion matrix further confirmed the robust classification performance of the MF-CNN. This research introduces a technique for medical imaging that can potentially transform diagnostics in gastrointestinal healthcare facilities worldwide.

Optimizing vitiligo diagnosis with ResNet and Swin transformer deep learning models: a study on performance and interpretability.

Vitiligo is a hypopigmented skin disease characterized by the loss of melanin. The progressive nature and widespread incidence of vitiligo necessitate timely and accurate detection. Usually, a single diagnostic test often falls short of providing definitive confirmation of the condition, necessitating the assessment by dermatologists who specialize in vitiligo. However, the current scarcity of such specialized medical professionals presents a significant challenge. To mitigate this issue and enhance diagnostic accuracy, it is essential to build deep learning models that can support and expedite the detection process. This study endeavors to establish a deep learning framework to enhance the diagnostic accuracy of vitiligo. To this end, a comparative analysis of five models including ResNet (ResNet34, ResNet50, and ResNet101 models) and Swin Transformer series (Swin Transformer Base, and Swin Transformer Large models), were conducted under the uniform condition to identify the model with superior classification capabilities. Moreover, the study sought to augment the interpretability of these models by selecting one that not only provides accurate diagnostic outcomes but also offers visual cues highlighting the regions pertinent to vitiligo. The empirical findings reveal that the Swin Transformer Large model achieved the best performance in classification, whose AUC, accuracy, sensitivity, and specificity are 0.94, 93.82%, 94.02%, and 93.5%, respectively. In terms of interpretability, the highlighted regions in the class activation map correspond to the lesion regions of the vitiligo images, which shows that it effectively indicates the specific category regions associated with the decision-making of dermatological diagnosis. Additionally, the visualization of feature maps generated in the middle layer of the deep learning model provides insights into the internal mechanisms of the model, which is valuable for improving the interpretability of the model, tuning performance, and enhancing clinical applicability. The outcomes of this study underscore the significant potential of deep learning models to revolutionize medical diagnosis by improving diagnostic accuracy and operational efficiency. The research highlights the necessity for ongoing exploration in this domain to fully leverage the capabilities of deep learning technologies in medical diagnostics.

On image search in histopathology.

Pathology images of histopathology can be acquired from camera-mounted microscopes or whole-slide scanners. Utilizing similarity calculations to match patients based on these images holds significant potential in research and clinical contexts. Recent advancements in search technologies allow for implicit quantification of tissue morphology across diverse primary sites, facilitating comparisons, and enabling inferences about diagnosis, and potentially prognosis, and predictions for new patients when compared against a curated database of diagnosed and treated cases. In this article, we comprehensively review the latest developments in image search technologies for histopathology, offering a concise overview tailored for computational pathology researchers seeking effective, fast, and efficient image search methods in their work.

Coping with test anxiety using imagery rescripting: A two-session randomized controlled trial.

BACKGROUND: Up to 55 % of students experience test anxiety (TA), which is characterized by intense physiological and psychological symptoms before or during exams, such as anxiety, fear of failure, sweating, or increased heart rate. Furthermore, TA increases graduation times and can result in discontinuance of the graduate program all together. Previous research demonstrated the beneficial effects of combining cognitive behavioral therapy with imagery rescripting, however, treatment programs are comparably long. Hence, they do not account for the students´ time-sensitive schedules. Therefore, the present study investigates a two-session short-intervention using imagery rescripting to treat TA. METHODS: 44 students and pupils were randomly assigned to either the two-session imagery rescripting intervention (22 participants) or the waitlist-control condition (22 participants). One week before the intervention clinical interviews were conducted and self-report questionnaires on TA, self-efficacy, symptoms of depression, and intrusive prospective images were completed (T1). The same questionnaires were completed one week (T2) and six months after the intervention (T3). RESULTS: Test anxiety significantly decreased from T1 to T2, as well as from T1 to T3 within the intervention group. Furthermore, there were medium to large within and between group effects for situational test anxiety, self-efficacy, symptoms of depression, as well as prospective intrusive images, showing significant improvements for the intervention group at six months follow-up. LIMITATIONS: The study is limited to the comparably small sample size, as well as the sole usage of self-report measurements. CONCLUSIONS: The presented short-intervention provides a feasible treatment technique, which can be easily applied within school and university counseling centers.

A comparative analysis of pairwise image stitching techniques for microscopy images.

Stitching of microscopic images is a technique used to combine multiple overlapping images (tiles) from biological samples with a limited field of view and high resolution to create a whole slide image. Image stitching involves two main steps: pairwise registration and global alignment. Most of the computational load and the accuracy of the stitching algorithm depend on the pairwise registration method. Therefore, choosing an efficient, accurate, robust, and fast pairwise registration method is crucial in the whole slide imaging technique. This paper presents a detailed comparative analysis of different pairwise registration techniques in terms of execution time and quality. These techniques included feature-based methods such as Harris, Shi-Thomasi, FAST, ORB, BRISK, SURF, SIFT, KAZE, MSER, and deep learning-based SuperPoint features. Additionally, region-based methods were analyzed, which were based on the normalized cross-correlation (NCC) and the combination of phase correlation and NCC. Investigations have been conducted on microscopy images from different modalities such as bright-field, phase-contrast, and fluorescence. The feature-based methods were highly robust to uneven illumination in tiles. Moreover, some features were found to be more accurate and faster than region-based methods, with the SURF features identified as the most effective technique. This study provides valuable insights into the selection of the most efficient and accurate pairwise registration method for creating whole slide images, which is essential for the advancement of computational pathology and biology.

Features Associated with Visible Lamina Cribrosa Pores in Individuals of African Ancestry with Glaucoma: Primary Open-Angle African Ancestry Glaucoma Genetics (POAAGG) Study.

There are scarce data regarding the rate of the occurrence of primary open-angle glaucoma (POAG) and visible lamina cribrosa pores (LCPs) in the eyes of individuals with African ancestry; the potential impact of these features on disease burden remains unknown. We recruited subjects with POAG to the Primary Open-Angle African American Glaucoma Genetics (POAAGG) study. Through regression models, we evaluated the association between the presence of LCPs and various phenotypic features. In a multivariable analysis of 1187 glaucomatous eyes, LCPs were found to be more likely to be present in eyes with cup-to-disc ratios (CDR) of ≥0.9 (adjusted risk ratio (aRR) 1.11, 95%CI: 1.04-1.19, p = 0.005), eyes with cylindrical-shaped (aRR 1.22, 95%CI: 1.11-1.33) and bean pot (aRR 1.24, 95%CI: 1.13-1.36) cups versus conical cups (p < 0.0001), moderate cup depth (aRR 1.24, 95%CI: 1.06-1.46) and deep cups (aRR 1.27, 95%CI: 1.07-1.50) compared to shallow cups (p = 0.01), and the nasalization of central retinal vessels (aRR 1.33, 95%CI: 1.23-1.44), p < 0.0001). Eyes with LCPs were more likely to have a higher degree of African ancestry (q0), determined by means of SNP analysis (aRR 0.96, 95%CI: 0.93-0.99, p = 0.005 for per 0.1 increase in q0). Our large cohort of POAG cases of people with African ancestry showed that LCPs may be an important risk factor in identifying severe disease, potentially warranting closer monitoring by physicians.

Lossless and Near-Lossless Compression Algorithms for Remotely Sensed Hyperspectral Images.

Rapid and continuous advancements in remote sensing technology have resulted in finer resolutions and higher acquisition rates of hyperspectral images (HSIs). These developments have triggered a need for new processing techniques brought about by the confined power and constrained hardware resources aboard satellites. This article proposes two novel lossless and near-lossless compression methods, employing our recent seed generation and quadrature-based square rooting algorithms, respectively. The main advantage of the former method lies in its acceptable complexity utilizing simple arithmetic operations, making it suitable for real-time onboard compression. In addition, this near-lossless compressor could be incorporated for hard-to-compress images offering a stabilized reduction at nearly 40% with a maximum relative error of 0.33 and a maximum absolute error of 30. Our results also show that a lossless compression performance, in terms of compression ratio, of up to 2.6 is achieved when testing with hyperspectral images from the Corpus dataset. Further, an improvement in the compression rate over the state-of-the-art k2-raster technique is realized for most of these HSIs by all four variations of our proposed lossless compression method. In particular, a data reduction enhancement of up to 29.89% is realized when comparing their respective geometric mean values.

Scanning Micro X-ray Fluorescence and Multispectral Imaging Fusion: A Case Study on Postage Stamps.

Scanning micrο X-ray fluorescence (µ-XRF) and multispectral imaging (MSI) were applied to study philately stamps, selected for their small size and intricate structures. The µ-XRF measurements were accomplished using the M6 Jetstream Bruker scanner under optimized conditions for spatial resolution, while the MSI measurements were performed employing the XpeCAM-X02 camera. The datasets were acquired asynchronously. Elemental distribution maps can be extracted from the µ-XRF dataset, while chemical distribution maps can be obtained from the analysis of the multispectral dataset. The objective of the present work is the fusion of the datasets from the two spectral imaging modalities. An algorithmic co-registration of the two datasets is applied as a first step, aiming to align the multispectral and µ-XRF images and to adapt to the pixel sizes, as small as a few tens of micrometers. The dataset fusion is accomplished by applying k-means clustering of the multispectral dataset, attributing a representative spectrum to each pixel, and defining the multispectral clusters. Subsequently, the µ-XRF dataset within a specific multispectral cluster is analyzed by evaluating the mean XRF spectrum and performing k-means sub-clustering of the µ-XRF dataset, allowing the differentiation of areas with variable elemental composition within the multispectral cluster. The data fusion approach proves its validity and strength in the context of philately stamps. We demonstrate that the fusion of two spectral imaging modalities enhances their analytical capabilities significantly. The spectral analysis of pixels within clusters can provide more information than analyzing the same pixels as part of the entire dataset.

High Dynamic Range Image Reconstruction from Saturated Images of Metallic Objects.

This study considers a method for reconstructing a high dynamic range (HDR) original image from a single saturated low dynamic range (LDR) image of metallic objects. A deep neural network approach was adopted for the direct mapping of an 8-bit LDR image to HDR. An HDR image database was first constructed using a large number of various metallic objects with different shapes. Each captured HDR image was clipped to create a set of 8-bit LDR images. All pairs of HDR and LDR images were used to train and test the network. Subsequently, a convolutional neural network (CNN) was designed in the form of a deep U-Net-like architecture. The network consisted of an encoder, a decoder, and a skip connection to maintain high image resolution. The CNN algorithm was constructed using the learning functions in MATLAB. The entire network consisted of 32 layers and 85,900 learnable parameters. The performance of the proposed method was examined in experiments using a test image set. The proposed method was also compared with other methods and confirmed to be significantly superior in terms of reconstruction accuracy, histogram fitting, and psychological evaluation.

Early gastric cancer detection and lesion segmentation based on deep learning and gastroscopic images.

Gastric cancer is a highly prevalent disease that poses a serious threat to public health. In clinical practice, gastroscopy is frequently used by medical practitioners to screen for gastric cancer. However, the symptoms of gastric cancer at different stages of advancement vary significantly, particularly in the case of early gastric cancer (EGC). The manifestations of EGC are often indistinct, leading to a detection rate of less than 10%. In recent years, researchers have focused on leveraging deep learning algorithms to assist medical professionals in detecting EGC and thereby improve detection rates. To enhance the ability of deep learning to detect EGC and segment lesions in gastroscopic images, an Improved Mask R-CNN (IMR-CNN) model was proposed. This model incorporates a "Bi-directional feature extraction and fusion module" and a "Purification module for feature channel and space" based on the Mask R-CNN (MR-CNN). Our study includes a dataset of 1120 images of EGC for training and validation of the models. The experimental results indicate that the IMR-CNN model outperforms the original MR-CNN model, with Precision, Recall, Accuracy, Specificity and F1-Score values of 92.9%, 95.3%, 93.9%, 92.5% and 94.1%, respectively. Therefore, our proposed IMR-CNN model has superior detection and lesion segmentation capabilities and can effectively aid doctors in diagnosing EGC from gastroscopic images.

Correlations between developmental trajectories of brain functional connectivity, neurocognitive functions, and clinical symptoms in patients with attention-deficit hyperactivity disorder.

Several studies on attention-deficit hyperactivity disorder (ADHD) have suggested a developmental sequence of brain changes: subcortico-subcortical connectivity in children, evolving to subcortico-cortical in adolescence, and culminating in cortico-cortical connectivity in young adulthood. This study hypothesized that children with ADHD would exhibit decreased functional connectivity (FC) between the cortex and striatum compared to adults with ADHD, who may show increased FC in these regions. Seventy-six patients with ADHD (26 children, 26 adolescents, and 24 adults) and 74 healthy controls (25 children, 24 adolescents, and 25 adults) participated in the study. Resting state magnetic resonance images were acquired using a 3.0 T Philips Achieva scanner. The results indicated a gradual decrease in the number of subcategories representing intelligence quotient deficits in the ADHD group with age. In adulthood, the ADHD group exhibited lower working memory compared to the healthy control group. The number of regions showing decreased FC from the cortex to striatum between the ADHD and control groups reduced with age, while regions with increased FC from the default mode network and attention network in the ADHD group increased with age. In adolescents and adults, working memory was positively associated with brain activity in the postcentral gyrus and negatively correlated with ADHD clinical symptoms. In conclusion, the findings suggest that intelligence deficits in certain IQ subcategories may diminish as individuals with ADHD age. Additionally, the study indicates an increasing anticorrelation between cortical and subcortical regions with age in individuals with ADHD.

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.