Retinal images play a pivotal contribution to the diagnosis of various ocular conditions by ophthalmologists. Extensive research was conducted to enable early detection and timely treatment using deep learning algorithms for retinal fundus images. Quick diagnosis and treatment planning can be facilitated by deep learning models’ ability to process images rapidly and deliver outcomes instantly. Our research aims to provide a non-invasive method for early detection and timely eye disease treatment using a Convolutional Neural Network (CNN). We used a dataset Retinal Fundus Multi-disease Image Dataset (RFMiD), which contains various categories of fundus images representing different eye diseases, including Media Haze (MH), Optic Disc Cupping (ODC), Diabetic Retinopathy (DR), and healthy images (WNL). Several pre-processing techniques were applied to improve the model’s performance, such as data augmentation, cropping, resizing, dataset splitting, converting images to arrays, and one-hot encoding. CNNs have extracted extract pertinent features from the input color fundus images. These extracted features are employed to make predictive diagnostic decisions. In this article three CNN models were used to perform experiments. The model’s performance is assessed utilizing statistical metrics such as accuracy, F1 score, recall, and precision. Based on the results, the developed framework demonstrates promising performance with accuracy rates of up to 89.81% for validation and 88.72% for testing using 12-layer CNN after Data Augmentation. The accuracy rate obtained from 20-layer CNN is 90.34% for validation and 89.59% for testing with Augmented data. The accuracy obtained from 20-layer CNN is greater but this model shows overfitting. These accuracy rates suggested that the deep learning model has learned to distinguish between different eye disease categories and healthy images effectively. This study’s contribution lies in providing a reliable and efficient diagnostic system for the simultaneous detection of multiple eye diseases through the analysis of color fundus images.

The advancement and innovations in wireless communication technologies including the Internet of Things have massively changed the paradigms of health-based services. In particular, during the COVID-19 pandemic, the trends of working from home have been promoted. Wireless body area network technology frameworks help sufferers in remotely obtaining scientific remedies from physicians through the Internet without paying a visit to the clinics. IoT sensor nodes are incorporated into the clinical device to allow health workers to consult the patients’ fitness conditions in real time. Insecure wireless communication channels make unauthorized access to fitness-related records and manipulation of IoT sensor nodes attached to the patient’s bodies possible, as a result of security flaws. As a result, IoT-enabled devices are threatened by a number of well-known attacks, including impersonation, replay, man-in-the-middle, and denial-of-service assaults. Modern authentication schemes do solve these issues, but they frequently involve challenging mathematical concepts that raise processing and transmission costs. In this paper, we propose a lightweight, secure, and efficient symmetric key exchange algorithm and remote user authentication scheme. Our research proposal presents a successful privacy-protecting method for remote users and provides protection against known attacks. When compared to conventional options, this technique significantly reduces calculation costs by up to 37.68% and transmission costs by up to 32.55%.

The IPv4 address architecture has been declared ended finally due to the fast growth of the Internet of Things (IoT). IPv6 is becoming a next-generation communication protocol and provides all the requirements that the industry needs. A smart home is an emerging technological revolution in which smart IoT-based physical objects are linked to the Internet to make human life more comfortable. There are several resource-constrained smart devices interconnecting with 6LoWPAN to control the smart home remotely. The communication channels used by cellular communication are vulnerable and increase security threats especially related to authentication. A reliable and portable remote authentication method is critical for ensuring safe communication in the next-generation smart home environment. Recently, many authentication schemes have been proposed, but most of them adopted complex mathematical techniques or protocols that are viewed as heavyweights in the context of computation and communication. This research proposed a lightweight and reliable remote authentication mechanism for the next generation of IoT-based smart homes. Informal and formal security assessments through the AVISPA tool determine the robustness of our proposed scheme. Moreover, we implemented our authentication scheme on a Linux-based client-server network model by using Android programming. In addition, we compared our proposed scheme with existing schemes based on computation and communication costs. Results show that our proposed mechanism is better, more secure, and most suitable for smart home ecosystems.

Email phishing remains a prevalent and sophisticated cyber threat, targeting individuals and organizations by disguising malicious intent in seemingly legitimate communications. Effective classification of phishing and legitimate emails is crucial for cybersecurity. In this study, we investigated various text vectorization techniques and machine learning models to address the challenge of email classification. We utilized three vectorization techniques: TF-IDF, Word2Vec, and Doc2Vec. These techniques were applied to traditional machine learning algorithms, and their performance was evaluated against a proposed stacking model, LogiTriBlend. The dataset comprised 501 phishing and 4090 legitimate emails, undergoing preprocessing steps like stemming, lemmatization, and noise removal. To handle the dataset’s imbalance, Synthetic Minority Over-sampling Technique (SMOTE) was employed. The model combines multiple base learners, including Support Vector Machine (SVM), Logistic Regression, Random Forest, and XGBoost, with a Logistic Regression meta-learner. The experimental results indicated that the LogiTriBlend model achieved an accuracy of 99.34% using Doc2Vec, outperforming Word2Vec and TF-IDF feature extraction methods, which obtained accuracies of 99.12% and 98.80%, respectively. The Doc2Vec method resulting in superior email classification performance. Among the models tested, the proposed stacking model, LogiTriBlend, demonstrated robust results; however, the highest accuracy was consistently achieved using Doc2Vec.