| Abstract |
Recently, a wide variety of real-world problems are solved via cloud computing, deep learning, machine learning, artificial intelligence, and the Internet of Things (IoT). These methodologies are concerned with different areas like smart cities, agriculture, transportation systems, and healthcare systems. The existing researchers focused on the health care monitoring application along with IoT and cloud computing. It met several shortcomings in case of computational complexity, cost, time, and improper health care data storage and so forth. To overcome these challenges, the novel IoT-enabled secure healthcare monitoring model is proposed in this study. At first, several sensors are deployed in the human body thereby collecting patients' data with respect to vital parameters like body temperature deviation. The patient's health record dataset consists of 10 attributes namely phone numbers, marital status, address, age, name, heartbeat rate, oxygen level, smoking, temperature, and blood pressure. The data size reduction and normalization are performed in the IoT medical sensor dataset which contains the redundant or irrelevant attributes that are eliminated during pre-processing. The artificial hummingbird (AHB) algorithm-based convolutional neural network (AHB-CNN) model performs both feature extraction and classification of cancer disease. The AHB-CNN model classifies whether the patient is prone to cancer or not based on the sensor input collected. The received results are then sent to the hospital management for analysis. The Rivest-Shamir-Adleman (RSA) encryption method is mostly used in this study due to the major benefits it gives in terms of asymmetric encryption, ease of use, simpler deployment, and high security associated with factoring large prime numbers. A modified RSA algorithm is used in this article which uses the double encryption-decryption process and “n” prime numbers to enhance the security of the conventional RSA algorithm. The data is always encrypted during transit before being stored in the cloud or for further processing. Depending upon the experimental consequences, the proposed method established superior performances compared to other state-of-art techniques. © 2022 John Wiley & Sons, Ltd. |
