A vessel-guided multi-task deep learning framework with visual interpretability for simultaneous retinal vessel segmentation and multi-disease classification from fundus images.

Retinal vessel segmentation is the main step in the analysis of fundus images. However, gray-scales’ uneven distribution, complex structure, and serious noise interference bring difficulties of automatic retinal vessels segmentation on fundus images. To solve these problems, we present an effective retinal vessel segmentation model, SDUW-Net, in this paper. The same scale dense connection is designed to improve U-Net’s structure and remove different scales dense connections to accelerate the training speed. We use skip connections to merges the features between shallow layers and deep layers to retain more features that may be lost in the process of down sampling and convolution. Experimental results on the DRIVE dataset show that the retinal vessels can be effectively segmented out by our proposed SDUW-Net, which has AUC of 0.9811 with low computation and short training time.KeywordsRetinal vessels segmentationSame scale dense connectionSkip connection

We propose a lightweight multi-task deep learning architecture tailored for simultaneous retinal disease classification and blood vessel segmentation. Leveraging a shared Mo-bileNetV2 encoder, the model branches into task-specific heads for multi-class diabetic retinopathy (DR) grading and binary vessel segmentation. It is trained on the APTOS 2019 dataset for classification and the DRIVE dataset for segmentation using a label-masked, unified training strategy. The proposed model achieves 76.67% classification accuracy and a 0.5868 Dice score on segmentation, demonstrating competitive performance with significantly reduced computational complexity. Designed for real-time deployment on mobile devices, this model addresses the critical need for scalable, low-resource ophthalmic screening tools. Future work explores the integration of uncertainty estimation to enhance clinical interpretability and decision support.

A unified modeling framework for lane change intention recognition and vehicle status prediction

The retinal vessel segmentation task plays an important role in clinical diagnosis and treatment, especially in cardiovascular diseases such as diabetic retinopathy and hypertensive retinopathy. Recently, the fully convolutional neural network, as a popular learning-based segmentation method, has been demonstrated to yield highly segmentation performance in vessel wall segmentation tasks. However, the major network factors affecting the performance of segmentation are still not obvious. This paper uses the single-factor control variable method to investigates the effects of network architectures (FCN network and U-Net network) and other network factors (pooling times, patch size, number of skip connection, and network depth) on retinal blood vessel segmentation. Our experiments are performed on two public fundus image database DRIVE and STARE. The results show that U-net is better than FCN and skip connections, proper pooling times, dilated convolution is vital to obtain better performance of retinal vessel segmentation.

Purpose Retinal blood vessel image segmentation is an important step in ophthalmological analysis. However, it is difficult to segment small vessels accurately because of low contrast and complex feature information of blood vessels. The objective of this study is to develop an improved retinal blood vessel segmentation structure (WA-Net) to overcome these challenges. Methods This paper mainly focuses on the width of deep learning. The channels of the ResNet block were broadened to propagate more low-level features, and the identity mapping pathway was slimmed to maintain parameter complexity. A residual atrous spatial pyramid module was used to capture the retinal vessels at various scales. We applied weight normalization to eliminate the impacts of the mini-batch and improve segmentation accuracy. The experiments were performed on the DRIVE and STARE datasets. To show the generalizability of WA-Net, we performed cross-training between datasets. Results The global accuracy and specificity within datasets were 95.66% and 96.45% and 98.13% and 98.71%, respectively. The accuracy and area under the curve of the interdataset diverged only by 1%∼2% compared with the performance of the corresponding intradataset. Conclusion All the results show that WA-Net extracts more detailed blood vessels and shows superior performance on retinal blood vessel segmentation tasks.

ADMET property prediction via multi-task graph learning under adaptive auxiliary task selection

Accurate segmentation of retinal vessels is critical to the mechanism, diagnosis, and treatment of many ocular pathologies. Due to the poor contrast and inhomogeneous background of fundus imaging and the complex structure of retinal fundus images, this makes accurate segmentation of blood vessels from retinal images still challenging. In this paper, we propose an effective framework for retinal vascular segmentation, which is innovative mainly in the retinal image pre-processing stage and segmentation stage. First, we perform image enhancement on three publicly available fundus datasets based on the multiscale retinex with color restoration (MSRCR) method, which effectively suppresses noise and highlights the vessel structure creating a good basis for the segmentation phase. The processed fundus images are then fed into an effective Reverse Fusion Attention Residual Network (RFARN) for training to achieve more accurate retinal vessel segmentation. In the RFARN, we use Reverse Channel Attention Module (RCAM) and Reverse Spatial Attention Module (RSAM) to highlight the shallow details of the channel and spatial dimensions. And RCAM and RSAM are used to achieve effective fusion of deep local features with shallow global features to ensure the continuity and integrity of the segmented vessels. In the experimental results for the DRIVE, STARE and CHASE datasets, the evaluation metrics were 0.9712, 0.9822 and 0.9780 for accuracy (Acc), 0.8788, 0.8874 and 0.8352 for sensitivity (Se), 0.9803, 0.9891 and 0.9890 for specificity (Sp), area under the ROC curve(AUC) was 0.9910, 0.9952 and 0.9904, and the F1-Score was 0.8453, 0.8707 and 0.8185. In comparison with existing retinal image segmentation methods, e.g. UNet, R2UNet, DUNet, HAnet, Sine-Net, FANet, etc., our method in three fundus datasets achieved better vessel segmentation performance and results.

The segmentation of fundus vascular images is a key step in retinal image processing and analysis. It plays an important role in the early prevention and diagnosis of eye diseases and blood system diseases. In view of the low accuracy of retinal blood vessel segmentation caused by the complex features of fundus images, the presence of soft, hard penetrant and bleeding noises, the low contrast between retinal blood vessels and background, uneven changes in width and curvature, small shape and blurred edges, this paper proposes a new network model RCU-Net for retinal vessel segmentation. Specifically, this paper proposes CBAM-Conv convolution block and RA-Conv convolution block based on visual attention mechanism (CBAM) and residual connection idea, which can make the network better use of the information at the spatial and channel levels. Experiments were carried out in public datasets DRIVE and CHASE_DB1. The accuracy, sensitivity, F1-Score and AUC of RCU-Net network on DRIVE datasets reached 96.75%, 82.02%, 81.45% and 98.33% respectively. Compared with the classical U-Net network, the proposed model can effectively improve the performance of retinal vessel segmentation.

Advances in artificial intelligence (AI) in general and Natural Language Processing (NLP) in particular are paving the new way forward for the automated detection and prediction of mental health disorders among the population. Recent research in this area has prioritized predictive accuracy over model interpretability by relying on deep learning methods. However, prioritizing predictive accuracy over model interpretability can result in a lack of transparency in the decision-making process, which is critical in sensitive applications such as healthcare. There is thus a growing need for explainable AI (XAI) approaches to psychiatric diagnosis and prediction. The main aim of this work is to address a gap by conducting a systematic investigation of XAI approaches in the realm of automatic detection of mental disorders from language behavior leveraging textual data from social media. In pursuit of this aim, we perform extensive experiments to evaluate the balance between accuracy and interpretability across predictive mental health models. More specifically, we build BiLSTM models trained on a comprehensive set of human-interpretable features, encompassing syntactic complexity, lexical sophistication, readability, cohesion, stylistics, as well as topics and sentiment/emotions derived from lexicon-based dictionaries to capture multiple dimensions of language production. We conduct extensive feature ablation experiments to determine the most informative feature groups associated with specific mental health conditions. We juxtapose the performance of these models against a "black-box" domain-specific pretrained transformer adapted for mental health applications. To enhance the interpretability of the transformers models, we utilize a multi-task fusion learning framework infusing information from two relevant domains (emotion and personality traits). Moreover, we employ two distinct explanation techniques: the local interpretable model-agnostic explanations (LIME) method and a model-specific self-explaining method (AGRAD). These methods allow us to discern the specific categories of words that the information-infused models rely on when generating predictions. Our proposed approaches are evaluated on two public English benchmark datasets, subsuming five mental health conditions (attention-deficit/hyperactivity disorder, anxiety, bipolar disorder, depression and psychological stress).

Retinal fundus image is a crucial tool for ophthalmologists to diagnose eye-related diseases. These images provide visual information of the interior layer of the retina structures such as optic disc, optic cup, blood vessels and macula that can assist ophthalmologist in determining the health of an eye. Segmentation of blood vessels in fundus images is one of the most fundamental phase in detecting diseases such as diabetic retinopathy. However, the ambiguity of the retina structures in the retinal fundus images presents a challenge for researcher to segment the blood vessels. Extensive pre-processing and training of the images is necessary for precise segmentation, which is very intricate and laborious. This paper proposes the implementation of object-oriented-based metadata (OOM) structures of each pixel in the retinal fundus images. These structures comprise of additional metadata towards the conventional red, green, and blue data for each pixel within the images. The segmentation of the blood vessels in the retinal fundus images are performed by considering these additional metadata that enunciates the location, color spaces, and neighboring pixels of each individual pixel. From the results, it is shown that accurate segmentation of retinal fundus blood vessels can be achieved by purely employing straightforward thresholding method via the OOM structures without extensive pre-processing image processing technique or data training.

With improvements in fundus imaging technology and the increasing use of digital images in screening and diagnosis, the issue of automated analysis of retinal images is gaining more serious attention. We consider the problem of retinal vessel segmentation, a key issue in automated analysis of digital fundus images. We propose a texture-based vessel segmentation algorithm based on the notion of textons. Using a weak statistical learning approach, we construct textons for retinal vasculature by designing filters that are specifically tuned to the structural and photometric properties of retinal vessels. We evaluate the performance of the proposed approach using a standard database of retinal images. On the DRIVE data set, the proposed method produced an average performance of 0.9568 specificity at 0.7346 sensitivity. This compares well with the best-published results on the data set 0.9773 specificity at 0.7194 sensitivity [Proc. SPIE5370, 648 (2004)].

Accurate retinal vessel segmentation from Optical Coherence Tomography Angiography (OCTA) images is vital in ophthalmic medicine, particularly for the early diagnosis and monitoring of diseases, such as diabetic retinopathy and hypertensive retinopathy. The retinal vascular system exhibits complex characteristics, including branching, crossing, and continuity, which are crucial for precise segmentation and subsequent medical analysis. However, traditional pixel-wise vessel segmentation methods focus on learning how to effectively divide each pixel into different categories, relying mainly on local features, such as intensity and texture, and often neglecting the intrinsic structural properties of vessels. This can cause suboptimal segmentation accuracy and robustness, particularly when handling low-contrast, noisy, or pathologicalimages. This study aims to integrate structural priors into a segmentation framework. Prior embeddings are used to guide the segmentation process, which encode the typical morphology and topological structure of blood vessels. Incorporating these embeddings can improve the accuracy of retinal vessel segmentation, particularly in challenging areas such as small vessels and regions with ambiguous boundaries. This approach could help to preserve the integrity and continuity of the vascular structure, resulting in more reliable and precisesegmentation. This study adopts a generative image segmentation framework. A structured representation in a latent embedding space is presented to explore retinal vessel priors. On this basis, a prior-driven retinal vessel segmentation network is introduced. First, the retinal vessel priors from ground truth data are learned, which are encoded as embedding tokens through a residual quantization reconstruction network. The learned priors are stored in a codebook. In our network, a raw OCTA image is transformed into semantic features using an encoder. Each semantic feature is subsequently represented by a set of embedding tokens from the codebook. Finally, the retinal vessels are reconstructed, preserving the integrity and continuity of the vascular structures using the learned structuralpriors. The performance of our proposed network was assessed across three OCTA datasets: publically available ROSE-1 and ROSE-2, and the private dataset OCTA-Z. Both quantitative and qualitative evaluations revealed that our network outperformed current state-of-the-art methods. In particular, our approach achieved the average Dice scores of 77.63, 71.01, and 81.11% for ROSE-1, ROSE-2, and OCTA-Z,respectively. The experimental results demonstrated that the proposed network effectively learned and utilized implicit vessel priors for OCTA vessel segmentation. Extensive evaluations indicated that our network surpassed the current state-of-the-art methods. Specifically, our reconstruction approach of applying latent prior tokens offered a promising solution for the pattern representation of retinal vessels. In the future, we will extend this method to support more comprehensive retinal structure segmentation and eye-related disease classification tasks. Additionally, we plan to integrate diverse and detailed prior knowledge, such as anatomical and pathological information, to enhance the performance and versatility of ourapproach.

This paper deals with the retinal blood vessel segmentation in fundus video-sequences acquired by experimental fundus video camera. Quality of acquired video-sequences is relatively low and fluctuates across particular frames. Especially, due to the low resolution, poor signal-to-noise ratio, and varying illumination conditions within the frames, application of standard image processing methods might be difficult in such experimental fundus images. In this study, we tried two methods for the segmentation of retinal vessels - matched filtering and Hessian-based approach, originally developed for vessel segmentation in standard fundus images. We showed that modified versions of these two approaches, combined with support vector machine (SVM), can be used also for segmentation in experimental low-quality fundus video-sequences. The SVM classifier trained and consecutively tested on the database of high-resolution images achieved classification accuracy over 94 % and thus revealed a possible applicability of the proposed method on low-quality data. Then, testing on low-quality video-sequences revealed sufficiently large reliability in term of segmentation stability within the sequence with the interframe variability in image quality.

Automatic segmentation of medical pictures is a significant stage to extricate valuable data that can assist doctors to detect infections. Eye and systemic infections are known to show themselves in retinal vasculature. Retinal image segmentation is the undertaking of sectioning vessels in retina imagery. Segmentation of retinal vessel is one of the significant strides in retinal image analysis. The order that has been put together for the segmentation of vessel is the chief treatment, cinematography, feature selection, classifier development, testing, education and training in a variety of information and the evaluation of the presentation of the different measures. Many machine learning techniques have already been given by researchers to segment the blood vessels of the images of the retinal fundus of the eyes. A large portion of the accessible retinal vessels division strategies are inclined to more unfortunate outcomes when managing testing circumstances, such as, recognizing low-contrast miniature vessels, vessels with focal reflex, and vessels within the sight of pathologies. We classified the different methods of vascular segmentation of the controlled and uncontrolled practices. Moreover, in this work, we have reviewed and discussed various retinal image segmentation state-of-the-art techniques based on machine learning.

Trajectory regression, which aims to predict the travel time of arbitrary trajectories on road networks, attracts significant attention in various applications of traffic systems these years. In this paper, we tackle this problem with a multitask learning (MTL) framework. To take the temporal nature of the problem into consideration, we divide the regression problem into a set of sub-tasks of distinct time periods, then the problem can be treated in a multi-task learning framework. Further, we propose a novel regularization term in which we exploit the block sparse structure to augment the robustness of the model. In addition, we incorporate the spatial smoothness over road links and thus achieve a spatial-temporal framework. An accelerated proximal algorithm is adopted to solve the convex but non-smooth problem, which will converge to the global optimum. Experiments on both synthetic and real data sets demonstrate the effectiveness of the proposed method.