Deep learning-enhanced diabetic retinopathy image classification.

Objective: Diabetic retinopathy (DR) can sometimes be treated and prevented from causing irreversible vision loss if caught and treated properly. In this work, a deep learning (DL) model is employed to accurately identify all five stages of DR. Methods: The suggested methodology presents two examples, one with and one without picture augmentation. A balanced dataset meeting the same criteria in both cases is then generated using augmentative methods. The DenseNet-121-rendered model on the Asia Pacific Tele-Ophthalmology Society (APTOS) and dataset for diabetic retinopathy (DDR) datasets performed exceptionally well when compared to other methods for identifying the five stages of DR. Results: Our propose model achieved the highest test accuracy of 98.36%, top-2 accuracy of 100%, and top-3 accuracy of 100% for the APTOS dataset, and the highest test accuracy of 79.67%, top-2 accuracy of 92.%76, and top-3 accuracy of 98.94% for the DDR dataset. Additional criteria (precision, recall, and F1-score) for gauging the efficacy of the proposed model were established with the help of APTOS and DDR. Conclusions: It was discovered that feeding a model with higher-quality photographs increased its efficiency and ability for learning, as opposed to both state-of-the-art technology and the other, non-enhanced model.

A Trust-Based Model for Secure Routing against RPL Attacks in Internet of Things.

In IoT networks, the de facto Routing Protocol for Low Power and Lossy Networks (RPL) is vulnerable to various attacks. Routing attacks in RPL-based IoT are becoming critical with the increase in the number of IoT applications and devices globally. To address routing attacks in RPL-based IoT, several security solutions have been proposed in literature, such as machine learning techniques, intrusion detection systems, and trust-based approaches. Studies show that trust-based security for IoT is feasible due to its simple integration and resource-constrained nature of smart devices. Existing trust-based solutions have insufficient consideration of nodes' mobility and are not evaluated for dynamic scenarios to satisfy the requirements of smart applications. This research work addresses the Rank and Blackhole attacks in RPL considering the static as well as mobile nodes in IoT. The proposed Security, Mobility, and Trust-based model (SMTrust) relies on carefully chosen trust factors and metrics, including mobility-based metrics. The evaluation of the proposed model through simulation experiments shows that SMTrust performs better than the existing trust-based methods for securing RPL. The improvisation in terms of topology stability is 46%, reduction in packet loss rate is 45%, and 35% increase in throughput, with only 2.3% increase in average power consumption.

A Web-Based Model to Predict a Neurological Disorder Using ANN.

Dementia is a condition in which cognitive ability deteriorates beyond what can be anticipated with natural ageing. Characteristically it is recurring and deteriorates gradually with time affecting a person's ability to remember, think logically, to move about, to learn, and to speak just to name a few. A decline in a person's ability to control emotions or to be social can result in demotivation which can severely affect the brain's ability to perform optimally. One of the main causes of reliance and disability among older people worldwide is dementia. Often it is misunderstood which results in people not accepting it causing a delay in treatment. In this research, the data imputation process, and an artificial neural network (ANN), will be established to predict the impact of dementia. based on the considered dataset. The scaled conjugate gradient algorithm (SCG) is employed as a training algorithm. Cross-entropy error rates are so minimal, showing an accuracy of 95%, 85.7% and 89.3% for training, validation, and test. The area under receiver operating characteristic (ROC) curve (AUC) is generated for all phases. A Web-based interface is built to get the values and make predictions.

Detection of Skin Cancer Based on Skin Lesion Images Using Deep Learning.

An increasing number of genetic and metabolic anomalies have been determined to lead to cancer, generally fatal. Cancerous cells may spread to any body part, where they can be life-threatening. Skin cancer is one of the most common types of cancer, and its frequency is increasing worldwide. The main subtypes of skin cancer are squamous and basal cell carcinomas, and melanoma, which is clinically aggressive and responsible for most deaths. Therefore, skin cancer screening is necessary. One of the best methods to accurately and swiftly identify skin cancer is using deep learning (DL). In this research, the deep learning method convolution neural network (CNN) was used to detect the two primary types of tumors, malignant and benign, using the ISIC2018 dataset. This dataset comprises 3533 skin lesions, including benign, malignant, nonmelanocytic, and melanocytic tumors. Using ESRGAN, the photos were first retouched and improved. The photos were augmented, normalized, and resized during the preprocessing step. Skin lesion photos could be classified using a CNN method based on an aggregate of results obtained after many repetitions. Then, multiple transfer learning models, such as Resnet50, InceptionV3, and Inception Resnet, were used for fine-tuning. In addition to experimenting with several models (the designed CNN, Resnet50, InceptionV3, and Inception Resnet), this study's innovation and contribution are the use of ESRGAN as a preprocessing step. Our designed model showed results comparable to the pretrained model. Simulations using the ISIC 2018 skin lesion dataset showed that the suggested strategy was successful. An 83.2% accuracy rate was achieved by the CNN, in comparison to the Resnet50 (83.7%), InceptionV3 (85.8%), and Inception Resnet (84%) models.

Earlier Detection of Brain Tumor by Pre-Processing Based on Histogram Equalization with Neural Network.

MRI is an influential diagnostic imaging technology specifically worn to detect pathological changes in tissues with organs early. It is also a non-invasive imaging method. Medical image segmentation is a complex and challenging process due to the intrinsic nature of images. The most consequential imaging analytical approach is MRI, which has been in use to detect abnormalities in tissues and human organs. The portrait was actualized for CAD (computer-assisted diagnosis) utilizing image processing techniques with deep learning, initially to perceive a brain tumor in a person with early signs of brain tumor. Using AHCN-LNQ (adaptive histogram contrast normalization with learning-based neural quantization), the first image is preprocessed. When compared to extant techniques, the simulation outcome shows that this proposed method achieves an accuracy of 93%, precision of 92%, and 94% of specificity.

Hybrid HCNN-KNN Model Enhances Age Estimation Accuracy in Orthopantomography.

Age estimation in dental radiographs Orthopantomography (OPG) is a medical imaging technique that physicians and pathologists utilize for disease identification and legal matters. For example, for estimating post-mortem interval, detecting child abuse, drug trafficking, and identifying an unknown body. Recent development in automated image processing models improved the age estimation's limited precision to an approximate range of +/- 1 year. While this estimation is often accepted as accurate measurement, age estimation should be as precise as possible in most serious matters, such as homicide. Current age estimation techniques are highly dependent on manual and time-consuming image processing. Age estimation is often a time-sensitive matter in which the image processing time is vital. Recent development in Machine learning-based data processing methods has decreased the imaging time processing; however, the accuracy of these techniques remains to be further improved. We proposed an ensemble method of image classifiers to enhance the accuracy of age estimation using OPGs from 1 year to a couple of months (1-3-6). This hybrid model is based on convolutional neural networks (CNN) and K nearest neighbors (KNN). The hybrid (HCNN-KNN) model was used to investigate 1,922 panoramic dental radiographs of patients aged 15 to 23. These OPGs were obtained from the various teaching institutes and private dental clinics in Malaysia. To minimize the chance of overfitting in our model, we used the principal component analysis (PCA) algorithm and eliminated the features with high correlation. To further enhance the performance of our hybrid model, we performed systematic image pre-processing. We applied a series of classifications to train our model. We have successfully demonstrated that combining these innovative approaches has improved the classification and segmentation and thus the age-estimation outcome of the model. Our findings suggest that our innovative model, for the first time, to the best of our knowledge, successfully estimated the age in classified studies of 1 year old, 6 months, 3 months and 1-month-old cases with accuracies of 99.98, 99.96, 99.87, and 98.78 respectively.

Detection of COVID-19 Based on Chest X-rays Using Deep Learning.

The coronavirus disease (COVID-19) is rapidly spreading around the world. Early diagnosis and isolation of COVID-19 patients has proven crucial in slowing the disease's spread. One of the best options for detecting COVID-19 reliably and easily is to use deep learning (DL) strategies. Two different DL approaches based on a pertained neural network model (ResNet-50) for COVID-19 detection using chest X-ray (CXR) images are proposed in this study. Augmenting, enhancing, normalizing, and resizing CXR images to a fixed size are all part of the preprocessing stage. This research proposes a DL method for classifying CXR images based on an ensemble employing multiple runs of a modified version of the Resnet-50. The proposed system is evaluated against two publicly available benchmark datasets that are frequently used by several researchers: COVID-19 Image Data Collection (IDC) and CXR Images (Pneumonia). The proposed system validates its dominance over existing methods such as VGG or Densnet, with values exceeding 99.63% in many metrics, such as accuracy, precision, recall, F1-score, and Area under the curve (AUC), based on the performance results obtained.

A new spatial spherical pattern model into interactive cartography pattern: multi-dimensional data via geostrategic cluster.

A growing amount of research conducted in digital, cooperative with advances in Artificial Intelligence, Computer Vision including Machine learning, has managed to the advance of progressive techniques that aim to detect and process affective information contained in multi-modal evidences. This research intends to bring together for theoreticians and practitioners from academic fields, professionals and industries and extends to be visualizing cries such epidemic, votes, social Phenomena in spherical representation interactive model working in the broad range of topics relevant to multi - modal data processing and forensics tools developing. Furthermore, progress has been made in this research besides that in this research conducted progression of mapping claims in present epoch necessitate the capacities of virtual guide of any understandable Geo-Visualization of spatial features that talented to convert the quantities of spatial pattern into cartography. The enlargement of a novel approaches fit for visualization of spatial pattern constituencies Starting exclusive Input Set of object O, set associated with feature F for regenerating Output the set C , interested region I special target C Even so, as indicated by the construction of the prototype as listed earlier in this thread, does it have the incentive for improvements: Representation could be used by Google Earth can Using Project enhancement representation whereby provides a 3D or 4D interaction with life measures with a view to cartography. In addition, the initiative suggests that a tool not accessible for disseminating information to the public can be addressed by the use of online mapping, which fuses with trends visualization for political circles and electors. But as mentioned above the framework is developed and it's also possible in the current example, for improvements: The project's representation 3D or 4D interacting Earth can use measures of life Earth From the map viewpoint. That's what that says. That means that. Which just means. Developers have concerns that. So it. Designers concern about that. This study supports the new, multi - demission and deployed countries in conjunction with another data is processed. Comprehensive, well-interpreted source data for the Data like Malaysia Jabatan Pendaftaran (JPN).