PROTECT: Privacy-preserving Contact Tracing for COVID-19 with Homomorphic Encryption.

BACKGROUND Various techniques are employed in order to support contact tracing, which has been shown to be highly effective against the pandemic of coronavirus disease 2019 (COVID-19). To apply the technology, either the quarantine authorities should provide the location history COVID-19 patients, or all people should provide their own location history. This inevitably makes people either the patient location history or personal location history of the public, leading to the privacy protection issue of information release for the public good against privacy exposure risks. OBJECTIVE The objective of this study is to develop an effective contact tracing system without exposing the location information between the user and the quarantine authorities with COVID-19 patient location history. METHODS We propose a new protocol called PRivacy Oriented Technique for Epidemic Contact Tracing (PROTECT) that securely shares the location information of patients with users by using the Brakerski/Fan-Vercauteren (BFV) homomorphic encryption scheme, along with a new secure proximity computation method for it. RESULTS We have developed a mobile application for the end-user and a web service for the quarantine authorities by applying the proposed method and have verified their effectiveness. Proposed application and web service compute the existence of intersections between the encrypted location history of COVID-19 patients released by the quarantine authorities and the user location history saved on the local device. We also show that the developed contact tracing application can identify whether the user is in contact with patients within a reasonable time on smartphones. CONCLUSIONS The developed method that shares the location information encrypted with homomorphic encryption is a new method for contact tracing without exposing the location information of the COVID-19 patients and the users. Homomorphic encryption is difficult to apply to practical issues despite its high security value. This study, however, has designed a system applicable to a reasonable size using the BFV scheme, and developed it to an operable format. The developed application and web service can help contact tracing for not only COVID-19, but also other various epidemics. CLINICALTRIAL