The Internet of Things (IoT) refers to a system of interconnected, internet-connected devices and sensors that allows the collection and dissemination of data. The data provided by these sensors may include outliers or exhibit anomalous behavior as a result of attack activities or device failure, for example. However, the majority of existing outlier detection algorithms rely on labeled data, which is frequently hard to obtain in the IoT domain. More crucially, the IoT’s data volume is continually increasing, necessitating the requirement for predicting and identifying the classes of future data. In this study, we propose an unsupervised technique based on a deep Variational Auto-Encoder (VAE) to detect outliers in IoT data by leveraging the characteristic of the reconstruction ability and the low-dimensional representation of the input data’s latent variables of the VAE. First, the input data are standardized. Then, we employ the VAE to find a reconstructed output representation from the low-dimensional representation of the latent variables of the input data. Finally, the reconstruction error between the original observation and the reconstructed one is used as an outlier score. Our model was trained only using normal data with no labels in an unsupervised manner and evaluated using Statlog (Landsat Satellite) dataset. The unsupervised model achieved promising and comparable results with the state-of-the-art outlier detection schemes with a precision of ≈90% and an F1 score of 79%.

Alzheimer’s Disease (AD) is characterized by the gradual degeneration and decline of brain cells, leading to irreversible neurological changes. This study investigates advanced image enhancement techniques for improving AD diagnosis using brain MRI. The methods used include CLAHE (Contrast Limited Adaptive Histogram Equalization) to enhance local image contrast and ESRGAN (Enhanced Super-Resolution Generative Adversarial Networks) to improve image resolution. These preprocessing methods improve MRI images and classification accuracy. An ensemble model of MobileNetV2 and DenseNet121, two efficient deep-learning models with feature extraction capabilities, were used as classifiers. This approach addresses challenges in AD diagnosis by leveraging deep learning for more accurate classification of brain tissue images. The model achieved an accuracy of 80.31% for MobileNetV2 and 89.22% for DenseNet121 when no enhancements were utilized. The model achieved accuracies of 92.34% and 89.38% for MobileNetV2 and DenseNet121, respectively, when enhancement methods were utilized, indicating strong dependability with the Kaggle dataset of MRI images. These findings underscore the efficacy of advanced image processing and deep learning in the early and accurate detection of Alzheimer’s disease.

Health insurance has become a crucial component of people’s lives as the occurrence of health problems rises. Unaffordable healthcare problems for individuals with little income might be a problem. In the case of a medical emergency, health insurance assists individuals in affording the costs of healthcare services and protects them financially against the possibility of debt. Security, privacy, and fraud risks may impact the numerous benefits of health insurance. In recent years, health insurance fraud has been a contentious topic due to the substantial losses it causes for individuals, commercial enterprises, and governments. Therefore, there is a need to develop mechanisms for identifying health insurance fraud incidents. Furthermore, a large quantity of highly sensitive electronic health insurance data are generated on a daily basis, which attracts fraudulent users. Motivated by these facts, we propose a smart healthcare insurance framework for fraud detection and prevention (SHINFDP) that leverages the capabilities of cutting-edge technologies including blockchain, 5G, cloud, and machine learning (ML) to enhance the health insurance process. The proposed framework is evaluated using mathematical modeling and an industrial focus group. In addition, a case study was demonstrated to illustrate the SHINFDP’s applicability in enhancing the security and effectiveness of health insurance. The findings indicate that the SHINFDP aids in the detection of healthcare fraud at early stages. Furthermore, the results of the focus group show that SHINFDP is adaptable and simple to comprehend. The case study further strengthens the findings and also describes the implications of the proposed solution in a real setting.