HEECCNB: An Efficient IoT-Cloud Architecture for Secure Patient Data Transmission and Accurate Disease Prediction in Healthcare Systems

• Main Authors: Veena, C., Sridevi, Mulagundla, Liyakat, Kazi Kutubuddin Sayyad, Saha, Bishal, Reddy, Sheri Ramchandra, Shirisha, N
• Format: Conference Proceeding
• Language: English
• Subjects: Cloud computing; Computation Overhead and Patient Data; Computer architecture; Data privacy; Data Security; Health Care System; Internet of Things; Internet of Things (IoT); IoT-Cloud; Medical services; Prediction algorithms; Predictive models
• ISSN: 2640-074X
• DOI: 10.1109/ICIIP61524.2023.10537627

The rapid expansion of the Internet of Things (IoT) in various fields has brought about significant advancements in data processing capabilities, in particular with the advent of IoT-cloud-based devices. Among these advancements, the healthcare system stands as one of the most intriguing IoT-Cloud integration uses cases. However, ensuring the security and privacy of patient data has become a critical concern, necessitating the development of robust protective measures. This research addresses the challenges of data security and computation overhead in the context of an IoT-cloud-based health system. A novel strategy dubbed Homomorphic Encryption with Elliptical Curve Cryptography integrated with Naive Bayes (HEECCNB) is put forth to address these problems. This approach is intended to enable precise disease prediction based on the gathered IoT sensor data while simultaneously providing good patient data privacy inside the IoT Health Cloud system. By employing homomorphic encryption, sensitive patient information can be securely processed and analyzed within the cloud environment without the need for decryption, preserving the confidentiality of the data. Furthermore, the integration of Elliptical Curve Cryptography enhances the security of the system by providing efficient and robust cryptographic operations. The proposed HEECCNB approach demonstrates promising results in both data protection and disease prediction. Through experiments conducted on the IoT Health Cloud scheme, the research validates the effectiveness of the projected method in safeguarding patient data and accurately predicting diseases based on IoT sensor data. Hence, the HEECCNB approach presents a compelling solution to the data security and computation overhead challenges faced in IoT-cloud-based health systems. Its successful implementation could significantly enhance the overall functionality and trustworthiness of such systems, facilitating the integration of IoT technologies in healthcare for improved patient outcomes and data privacy.