Advanced Privacy Preserving Model for Smart Healthcare Using Deep Learning
Pinkal Jain, Harish Kumar Shakya, Ajay Lala · 2023
The rise of intelligent healthcare, the rise of the internet of medical things, the corona virus pandemic, and the rise of big data-based medical applications. The generated biomedical data is extremely private and confidential. Unfortunately, traditional medicine systems are unable to handle the enormous volume of biomedical data. Consequently, data is usually the cloud is used to store and share data. Afterward, several uses for the shared data are made such as investigation and the finding of novel information. Biomedical information typically appears in written form (such as test results, medication recommendations, and diagnoses). Unfortunately, such data is exposed to a variety of security risks and attacks, including privacy and confidentiality breaches. Despite significant advancements in the security of biological data, most current methods produce significant delays and are unable to handle real-time responses. In order to enhance the healthcare system, this study presents an innovative, fog-enabled, privacy-preserving paradigm termed “r sanitizer.” This work is based on a Convolutional Neural Network with Bidirectional Long Short Term Memory (LSTM) and recognizes medical entities efficiently. The test results show that with a recall of 91.15%, precision of 92.6%, and a F1 score of 92%, r sanitizer beats the most recent models. Comparing the sanitization model to the state-of-the-art reveals a 28.77%improvement in privacy preservation.