Pneumonia is an acute lung infection that has caused many fatalitiesglobally. Radiologists often employ chest X-rays to identify pneumoniasince they are presently the most effective imaging method for this ***-aided d...
详细信息
Pneumonia is an acute lung infection that has caused many fatalitiesglobally. Radiologists often employ chest X-rays to identify pneumoniasince they are presently the most effective imaging method for this ***-aided diagnosis of pneumonia using deep learning techniques iswidely used due to its effectiveness and performance. In the proposed method,the Synthetic Minority Oversampling Technique (SMOTE) approach is usedto eliminate the class imbalance in the X-ray dataset. To compensate forthe paucity of accessible data, pre-trained transfer learning is used, and anensemble Convolutional Neural Network (CNN) model is developed. Theensemble model consists of all possible combinations of the MobileNetv2,Visual Geometry Group (VGG16), and DenseNet169 models. MobileNetV2and DenseNet169 performed well in the Single classifier model, with anaccuracy of 94%, while the ensemble model (MobileNetV2+DenseNet169)achieved an accuracy of 96.9%. Using the data synchronous parallel modelin Distributed Tensorflow, the training process accelerated performance by98.6% and outperformed other conventional approaches.
The usage of machine learning and deep learning algorithms have necessitated Artificial Intelligence'. AI is aimed at automating things by limiting human interference. It is widely used in IT, healthcare, finance,...
详细信息
The present paper attempts to design an adaptive multi-model predictive control strategy for strongly nonlinear or switched systems with various operating points. The proposed control system guarantees the feasibility...
详细信息
Classification of brain haemorrhage is a challenging task that needs to be solved to help advance medical treatment. Recently, it has been observed that efficient deep learning architectures have been developed to det...
详细信息
Recently, with increased use of mobile phones, it has transformed into a multibillion-dollar Short Message Service or SMS. However, the drop in the cost of messaging services has led to an increased number of unsolici...
详细信息
The thyroid gland, a pivotal regulator of essential physiological functions, orchestrates the production and release of thyroid hormones, playing a vital role in metabolism, growth, development, and overall bodily fun...
详细信息
Quality degradation due to the compression and the transmission of images is a significant threat to multimedia applications. Blind image quality assessment (BIQA) is a principal technique to measure the distortion an...
详细信息
An effective task scheduling method can accommodate user needs, boost resource usage, and boost cloud computing's overall efficiency. However, the unchanging task needs are generally the focus of grid computing...
详细信息
An effective task scheduling method can accommodate user needs, boost resource usage, and boost cloud computing's overall efficiency. However, the unchanging task needs are generally the focus of grid computing's job scheduling, leading to low resource usage. Distributing the dynamic user tasks fairly among all cloud nodes is the goal of load balancing, a relatively new field of study. The primary difficulty with cloud computing is load balancing. By making better use of available resources, load balancing methods improve cloud performance. Load balancing primary goal is to lessen the burden on the environment by cutting down on energy use and carbon emissions. The most crucial characteristics that can both satisfy user needs and maximize resource utilization are used to determine the order of priorities. Existing systems often ignore user priority suggestions in favor of optimal scheduling to improve load balancing. Scheduling that takes into account user-guided priorities uses a data-driven strategy, which helps improve load balancing. Scheduling algorithms that take user priorities into account can optimize load distribution more effectively. The primary objective of this research is to provide a priority based randomized load balancing technique that assigns tasks to virtual machines in a random fashion based on criteria such as the number of users, the amount of time the task takes to run, the type of software being used, the cost of the software, and the amount of available resources. This method maximizes system performance by decreasing response time and resource consumption while increasing metrics like fault tolerance and scalability. This system for scheduling tasks not only accommodates user needs but also achieves excellent resource usage. This research proposes a User Task Priority based Resource Allocation with Multi Class Task Scheduling Strategy and Load Balancing (UPRA-MCTSS-LB) Model for enhancing the cloud service quality. The proposed method res
Anaemia, a condition characterised by reduced haemoglobin levels, exerts a significant global impact, affecting billions of individuals worldwide. According to data from the World Health Organisation (WHO), India exhi...
详细信息
The accurate and early detection of abnormalities in fundus images is crucial for the timely diagnosis and treatment of various eye diseases, such as glaucoma and diabetic retinopathy. The detection of abnormalities i...
详细信息
The accurate and early detection of abnormalities in fundus images is crucial for the timely diagnosis and treatment of various eye diseases, such as glaucoma and diabetic retinopathy. The detection of abnormalities in fundus images using traditional methods is often challenging due to high computational demands, scalability issues, and the requirement of large labeled datasets for effective training. To address these limitations, a new method called triplet-based orchard search (Triplet-OS) has been proposed in this paper. In this study, a GoogleNet (Inception) is utilized for feature extraction of fundus images. Also, the residual network is employed to detect abnormalities in fundus images. The Triplet-OS utilizes the medical imaging technique fundus photography dataset to capture detailed images of the interior surface of the eye, known as the fundus and the fundus includes the retina, optic disk, macula, and blood vessels. To enhance the performance of the Triplet-OS method, the orchard optimization algorithm has been implemented with an initial search strategy for hyperparameter optimization. The performance of the Triplet-OS method has been evaluated based on different metrics such as F1-score, specificity, AUC-ROC, recall, precision, and accuracy. Additionally, the performance of the proposed method has been compared with existing methods. Few-shot learning refers to a process where models can learn from just a small number of examples. This method has been applied to reduce the dependency on deep learning [1]. The goal is for machines to become as intelligent as humans. Today, numerous computing devices, extensive datasets, and advanced methods such as CNN and LSTM have been developed. AI has achieved human-like performance and, in many fields, surpasses human abilities. AI has become part of our daily lives, but it generally relies on large-scale data. In contrast, humans can often apply past knowledge to quickly learn new tasks [2]. For example, if given
暂无评论