Blockchain-Enabled Online Diagnostic Platform of Suspected Patients of COVID-19 Like Pandemics
Mahmoud Abouyoussef, Surbhi Bhatia, Poonam Chaudhary, Saurabh Sharma, Muhammad Ismail
Abstract
During times of pandemics, the healthcare system may collapse due to the high demand for healthcare resources. Hence, there is a need for an online-automated platform that enables remote collection of symptoms from suspected patients, accurate and fast diagnostics, and data sharing among different entities within the healthcare system. However, many privacy and scalability challenges face such a platform. To address such challenges, we propose a custom-designed blockchain enabled platform that guarantees privacy-preservation via a mixture of group signature and random numbers that support anonymity of suspected patients and unlinkability of data while enabling mutual interaction between the suspected patient and the platform; provides automatic diagnostics via a deep neural network-based detector that runs on a smart contract within the blockchain; and offers access and administrative authority of the healthcare entities to the database of symptoms and their diagnoses via a consortium-based blockchain architecture. Experimental studies demonstrate a detection accuracy of 90 percent based on a deep convolutional recurrent neural network. A case study of 500 expected patients is examined giving promising results. Every patient can know the test results after only 14 min of submitting the data. The storage requirements are as low as 0.52 MB for each suspected patient and 0.6 MB for each hospital.