Figure - available from: Computational Intelligence and Neuroscience
This content is subject to copyright. Terms and conditions apply.
Source publication
Breast cancer is a lethal illness that has a high mortality rate. In treatment, the accuracy of diagnosis is crucial. Machine learning and deep learning may be beneficial to doctors. The proposed backbone network is critical for the present performance of CNN-based detectors. Integrating dilated convolution, ResNet, and Alexnet increases detection...
Citations
... It also needs topic expertise as well as the support of a radiologist. Deep learning (DL) models, on the other hand, may extract features from an input dataset and generate an adaptive learning process depending on the goal output [14]. The DL techniques significantly accelerate data processing and pattern extraction activities, enabling for the reuse of approaches Various studies have been conducted to look at breast cancer images from various angles [15]. ...
The most common type of cancer that results in death is breast cancer. In the world, millions of people struggle with this disease. Breast cancer can affect men and women but women are more affected. For awareness, it is necessary to understand the sign and symptoms of breast cancer. The most common sign is an abnormal lump in the breast. But there may be many reasons of develop abnormal lumps. Computer-Aided Diagnosis (CAD) is extensively used in pathological image analysis to help pathologists enhance diagnosis efficiency, accuracy, and consistency. Recent studies have looked into deep learning methodologies to improve the effectiveness of pathological CAD.
... VGG-16 is made up of 16 convolutional layers as compared with AlexNet's 5 convolutional layers and 3 fully connected layers. VGG-16 uses the highest filter size of 4×, while AlexNet utilizes the smallest filter size of 3 × 3 [34]. ...
In computer vision, the convolutional neural network (CNN) is a very popular model used for emotion recognition. It has been successfully applied to detect various objects in digital images with remarkable accuracy. In this paper, we extracted learned features from a pre-trained CNN and evaluated different machine learning (ML) algorithms to perform classification. Our research looks at the impact of replacing the standard SoftMax classifier with other ML algorithms by applying them to the FC6, FC7, and FC8 layers of Deep Convolutional Neural Networks (DCNNs). Experiments were conducted on two well-known CNN architectures, AlexNet and VGG-16, using a dataset of masked facial expressions (MLF-W-FER dataset). The results of our experiments demonstrate that Support Vector Machine (SVM) and Ensemble classifiers outperform the SoftMax classifier on both AlexNet and VGG-16 architectures. These algorithms were able to achieve improved accuracy of between 7% and 9% on each layer, suggesting that replacing the classifier in each layer of a DCNN with SVM or ensemble classifiers can be an efficient method for enhancing image classification performance. Overall, our research demonstrates the potential for combining the strengths of CNNs and other machine learning (ML) algorithms to achieve better results in emotion recognition tasks. By extracting learned features from pre-trained CNNs and applying a variety of classifiers, we provide a framework for investigating alternative methods to improve the accuracy of image classification.
... This makes it a good candidate for large-scale IoT installations. Extensive simulations in a realistic setting, including the healthcare industry, demonstrated the hybrid MFO-DNN algorithm's utility in terms of reaction time, resource utilisation, and load balancing [32][33]. Finally, when employed for task scheduling in cloud-IoHT systems, the HMFO algorithm provides a feasible way for enhancing resource allocation and ensuring efficient performance. ...
Keeping a proper level of task dependency throughout the scheduling process is critical to achieving the aim of decreasing the make-span rate in Internet of Health Things (IoHT) projects. We provide a smart model strategy for effective task scheduling in the IoHT environment for e-healthcare systems by merging hybrid moth flame optimisation (HMFO) with cloud computing. The HMFO algorithm guarantees that all available resources are distributed evenly, resulting in improved quality of service (QoS). We study the Google cluster dataset to learn about the scheduling behaviours of cloud-based jobs in order to train our model. After training, an HMFO model may be used to plan activities in real time. To assess the success of our strategy, we run simulations in the CloudSim environment, taking into account crucial parameters such as resource utilisation, reaction time, and energy consumption. According to a comparative analysis, our hybrid HMFO system surpasses the alternatives in terms of reaction time, average run duration, and cost savings. Our method has proven to be effective due to the favourable effects it has had on response rates, prices, and run times. Combining IoT and cloud computing has the potential to improve healthcare delivery in a variety of ways. One unique strategy we offer for scheduling IOHT jobs is to combine a deep neural network (DNN) algorithm with the MFO technique. Job scheduling in electronic healthcare systems can be optimised with the help of our hybrid MFO-DNN algorithm by taking into account a variety of different objectives, the most important of which are lowering response times while improving resource utilisation and maintaining consistent load balances. The MFO approach searches the search space and provides early solutions, while the DNN algorithm refines and improves those first findings. In comprehensive simulations conducted in a real-world hospital setting, the hybrid MFO-DNN technique outperformed existing scheduling algorithms in terms of reaction time, resource utilisation, and load balancing. The simulated healthcare environments were as true to life as was feasible. The suggested technique has been demonstrated to be both dependable and scalable, making it appropriate for use in large-scale IOHT deployments. This study considerably enhances the state of the art in IOHT task scheduling in E healthcare systems by developing a hybrid optimisation technique that takes advantage of the strengths of both MFO and DNN. The findings indicate that this strategy has the potential to improve the quality and efficiency of healthcare delivery, which helps patients receive care that is both effective and timely.
As one gets older, the likelihood of this happening increases. Minor cognitive impairment may serve as an early warning sign of dementia, according to popular belief. Some of the claims stated here are not supported by sources. Alzheimer’s disease and other kinds of neurodegeneration frequently manifest their first signs as dementia. People with poor intellect may struggle to discern between a sick and a healthy reference group. Our system can efficiently extract as much data from the scan as was feasible by using the entire three-dimensional magnetic resonance imaging images as input. The multichannel meta-discourse learning approach improves the system’s and network’s categorization abilities as well as their ability to make large generalizations. One option for achieving this goal is to broaden the number of available educational routes. The binarization loss and the unchaperoned contrastive loss are combined into a single loss to achieve this. Combining the binarization loss and the unchaperoned contrastive loss results is a realistic method that may help you achieve your aim. We ran many experiments using the acquired data to put our system through its paces and verify its value. These findings show that our method might be used to diagnose Alzheimer's disease, dementia, and mild cognitive impairment (MCI). Individual adaptation to their preferred mode of learning is one method that, if implemented, has the potential to significantly boost the value of educational experiences. By creating a fresh perspective on education, the network may be able to assist us in classifying data in a more accurate and appropriate manner in a wider range of scenarios.KeywordsAlzheimer’s diseaseMild cognitive impairmentMagnetic resonance imagingRegions of interestDeep learning
Convolutional neural networks (CNN) can no longer make a significant contribution to Alzheimer’s disease diagnosis because there is insufficient data to work with. We have built a cutting-edge deep learning system and are currently putting it to use to increase the effectiveness of the work we are doing to achieve this goal. To achieve the highest level of performance, we combine the advantages of fully stacked bidirectional long short-term memory (FSBi-LSTM) with those of three-dimensional convolutional neural networks. These two methods of data storage are stacked one on top of the other. Before interpreting the MRI and PET images, it is critical to train a three-dimensional convolutional neural network. This must be completed to proceed to the next stage. The essential qualities of the deep features can be agreed upon. Before any further inquiry into the matter can proceed, this must be done. Here is only one example of how this method may be applied. Even if only one individual is made aware of this, the ramifications might be terrible. Lastly, we compared our findings to those of an Alzheimer’s disease neuroimaging research study to show definitively that our technique is beneficial in Alzheimer’s disease management. According to our observations, our approach surpasses other theoretically comparable algorithms published in academic literature. These algorithms were evaluated based on their ability to tackle the same issue. This is true regardless of whether our technique is technically equivalent to other published methods: cases of pMCI can be distinguished from NC with a success rate of 94.82%; cases of sMCI can be distinguished from NC with an 86.30% success rate; and cases of Alzheimer’s Disease (AD) can be distinguished from NC with an 86.30% success rate. This result was obtained despite the fact that there was inadequate imaging evidence to back it up.KeywordsConvolution neural networkFully stacked bidirectionalLong short-term memoryDeep learningRecurrent neural networks
