Prospects of Deep Learning with Blockchain for Securing the Digital Radiography Data in Smart Healthcare

Blockchain knowledge is gaining traction in today's information and digital age, and it's expected to become extremely popular among the next generation for reasons that go well beyond bitcoin and other cryptocurrencies. Blockchain technology's potential use in healthcare may be observed through data-sharing, which enables users to pick certain data and limit data access based on user type, and is therefore crucial for the upkeep of EHRs. Despite the broad deployment of Internet of Things (IoT) integrated cloud for remote patient accuracy of sickness prediction remain important problems in today's health care system. Multiple machine learning techniques were used to successfully forecast illnesses at an early stage. Although it offers the possibility of precise categorizations, machine learning is time-consuming and inefficient. This research suggests combining Attribute based Searchable Honey Encryption with SqueezeNet to better analyse sickness and secure IoT-cloud healthcare data. To reduce variability and provide consistency across IoT data, means-mode normalisation is an essential first step. The final output was saved in view the prediction results and relevant patient evidence. We discovered that the projected model has higher accuracy (94%), sensitivity (94%), and specificity (94%), compared to the state-of-the-art models (also 94%).