Diabetic Retinopathy (DR) refers to damage to the retina caused by diabetes, which can cause visual impairments or potentially lead to blindness. The process of manually identifying diabetic retinopathy is slow and ca...
详细信息
ISBN:
(数字)9798331516284
ISBN:
(纸本)9798331516291
Diabetic Retinopathy (DR) refers to damage to the retina caused by diabetes, which can cause visual impairments or potentially lead to blindness. The process of manually identifying diabetic retinopathy is slow and can easily be affected by human mistakes because of the eye's complex anatomy. This paper aims to determine the optimal model for accurately staging diabetic retinopathy (DR) across five DR categories. The proposed system involves image pre-processing, segmentation and classification. A well-structured preprocessing framework was implemented, integrating Median Filtering technique to decrease noise and Gamma Correction for improved image quality. data augmentation is performed through methods like flipping, cropping rotating, and translation of fundus images. In this study, a hybrid optimization strategy is proposed by combining the Adam optimizer with Simulated Annealing (SA) and Cosine Annealing for training a U-Net model for diabetic retinopathy (DR) lesion segmentation. Canny Edge Detection is employed to discover edges correctly within the image. For DR categorisation in this work, enhanced Resnet 18 combined with MISH activation function is used. The suggested model makes use of residual blocks in conjunction with identity and convolutional blocks. Experiments on the APTOS dataset show that the proposed model outperforms state-of-the-art techniques. Accuracy, precision, recall, and F1 score are all improved by the suggested Resnet18 model, which receives scores of 99.45%, 98.64%, 98.63%, and 98.63%, respectively.
Our work proposes a comprehensive solution for predicting Metaverse network traffic, addressing the growing demand for intelligent resource management in eXtended Reality (XR) services. We first introduce a state-of-t...
详细信息
Task offloading management in 6G vehicular networks is crucial for maintaining network efficiency, particularly as vehicles generate substantial data. Integrating secure communication through authentication introduces...
详细信息
Proof of Authority (PoA) plays a pivotal role in blockchains for reaching consensus. Clique, which selects consensus nodes to generate blocks with a pre-determined order, is the most popular implementation of PoA due ...
详细信息
Among today's youth, heart disease has emerged as a major health concern. Many forms of cardiac disease can be traced back to bad habits and clinical characteristics. Many other models have been proposed by resear...
Among today's youth, heart disease has emerged as a major health concern. Many forms of cardiac disease can be traced back to bad habits and clinical characteristics. Many other models have been proposed by researchers to aid in the diagnosis of various conditions, however the vast majority either has poor accuracy performance or are overly complex to implement in real-time clinical settings. In addition, these models tend to function correctly for smaller class sets, but their performance degrades with an increase in class count. To facilitate answer these issues, this study suggests creating a novel, effective, and extensible strategy for disease prediction. This employs a deep learning technique, which boosts multiple sets of clinical data using a combination of StackingCV classifier-bagging and feature-boosting. Feature variance is increased by augmenting these vectors with a boosting of classifiers at the outset. AdaBoost strategy (ABA), Support-Vecto-Classifier (SVC), and Extreme-Gradient-Boosting (XGBoost) are then applied to these characteristics to help improve accuracy performance at the cost of minuscule increases in computational delays. Single illness classifier accuracy was 93.44% and multiple disease classifier accuracy was 60.5% when the model was tested using datasets from the University of California, Cleveland. When contrasted to various state-of-the-art methods and classifiers, such as a Deep-Neural-Network (DNN), Random-Forest (RF), XGBoost, &SVC, the suggested model was shown to be at least 3.9% more accurate. Thus, it is appropriate for use in live medical settings.
Exploring spatial contextual information is a well-adopted approach to achieving better semantic segmentation performance. However, most existing methods neglect the class association between the neighboring pixels. I...
详细信息
Exploring spatial contextual information is a well-adopted approach to achieving better semantic segmentation performance. However, most existing methods neglect the class association between the neighboring pixels. In this paper, we propose a CACINet, which consists of a Semantic Affinity Module (SAM) and a Class Association Module (CAM), to generate class-aware contextual information among pixels on a fine-grained level. SAM analyzes the affiliation of any two given pixels belonging to the same or different class. It produces intra-class and inter-class pixel contextual information. CAM classifies the image into different class regions globally and then it encodes the pixel based on the degree of affiliation of the pixels with each class in the image. In this way, it augments the class affiliation of the pixels into the corresponding context calculation. Comprehensive experiments demonstrate that the proposed method achieves competitive performance on two semantic segmentation benchmarks: ADE20K and PASCAL-Context.
This work provides a novel and sustainable advertisement display system that uses trained models, such as CNN and SSD, to adapt to the age and gender preferences of bystanders. By integrating a webcam with the adverti...
详细信息
ISBN:
(数字)9798350375442
ISBN:
(纸本)9798350375459
This work provides a novel and sustainable advertisement display system that uses trained models, such as CNN and SSD, to adapt to the age and gender preferences of bystanders. By integrating a webcam with the advertising screen, the system is able to identify the age and gender of passersby based on the faces that are detected and processed by trained models. Customers can personalize adverts to certain groups by accessing the classification results via a website, which is accessed and stored in real-time on a database server. The system has a motion sensor for energy economy and runs continuously, updating every two seconds. With 91.7% and 95.5% accuracy rates for age and gender classification, respectively, the technology guarantees accurate content delivery. It also has the capacity to show news and guidance in crowded places, thereby fostering societal awareness and knowledge. The work well analyzes the gender detection using conventional methods and reveled it significance.
The research investigates the optimization of multi-robot coordination in agricultural automation through Particle Swarm Optimization. The methodology unfolds in four key aspects: firstly, by harnessing swarm intellig...
详细信息
ISBN:
(数字)9798350384369
ISBN:
(纸本)9798350384376
The research investigates the optimization of multi-robot coordination in agricultural automation through Particle Swarm Optimization. The methodology unfolds in four key aspects: firstly, by harnessing swarm intelligence principles, the approach enhances coordination among multiple robots in agricultural settings, fostering collaboration and dynamic task allocation in response to real-time changes. Secondly, the integration of Particle Swarm Optimization (PSO) introduces a powerful optimization technique, enabling the algorithm to dynamically adapt the swarm's configuration. This adaptability optimizes task allocation and individual robot positions, enhancing overall efficiency. Thirdly, the methodology systematically tackles scalability, adaptability, and robustness challenges in agricultural automation. By synergizing swarm intelligence and PSO, it ensures efficient scalability with an increasing number of robots and responsiveness to evolving agricultural demands. The holistic design makes it well-suited for practical implementation in real-world agricultural scenarios. A series of experiments varying PSO parameters were conducted, revealing nuanced relationships between iterations, swarm size, and coefficients and their impact on convergence time and fitness metrics. The outcomes demonstrate the trade-offs involved in selecting these parameters for efficient coordination in agricultural tasks. A comparative analysis with ABO and ACO highlights PSO's superior performance, achieving remarkable fitness value of 98% in the context of robotic swarm applications.
Over the last 10 years, data mining has become more important, and the field of healthcare research has seen a significant uptick in activity. Most of the applications that were presented may be grouped into two disti...
详细信息
Over the last 10 years, data mining has become more important, and the field of healthcare research has seen a significant uptick in activity. Most of the applications that were presented may be grouped into two distinct classifications: decision support benches and policy formulation. It is still difficult to come across books in the field of healthcare that are worth reading. This review article provides an overview of the current health sector study using a variety of DM approaches and algorithms to look at various diseases such as cancer, diabetes, HIV, and skin co-related disorders and their accurate assessment. Additionally, a startling discovery that was provided to define the article is also provided.
Recent advances in vision transformers (ViTs) have achieved outstanding performance in visual recognition tasks, including image classification and detection. ViTs can learn global representations with their self-atte...
Recent advances in vision transformers (ViTs) have achieved outstanding performance in visual recognition tasks, including image classification and detection. ViTs can learn global representations with their self-attention mechanism, but they are usually heavy-weight and unsuitable for resource-constrained devices. In this paper, we propose a novel linear feature attention (LFA) module to reduce computation costs for vision transformers and combine efficient mobile CNN modules to form a parameter-efficient and high-performance CNN-ViT hybrid model, called LightFormer, which can serve as a general-purpose backbone to learn both global and local representation. Comprehensive experiments demonstrate that LightFormer achieves competitive performance across different visual recognition tasks. On the ImageNet-1K dataset, LightFormer achieves top-1 accuracy of 78.5% with 5.5 million parameters. Our model also performs well when transferred to object detection and semantic segmentation tasks. On the MS COCO dataset, LightFormer attains mAP of 33.2 within the YOLOv3 framework, and on the Cityscapes dataset, with only a simple all-MLP decoder, LightFormer achieves mIoU of 78.5 and FPS of 15.3, surpassing state-of-the-art lightweight segmentation networks.
暂无评论