Trusted Sensor Signal Protection for Confidential Point-of-Care Medical Diagnostic

Abstract

Trustworthy and usable healthcare requires not only effective disease diagnostic procedures to ensure delivery of rapid and accurate outcomes, but also maintaining the confidentiality of patient’s medical test results. This paper presents a diagnostic data protection scheme for cytometry-based point-of-care systems. Our solution consists in a biomarker detection sensor integrated with a smartphone to provide users with easy-to-use real-time diagnostic capabilities, thereby, reducing the need for in-person clinical visits. The proposed hardware-level trusted sensing framework obfuscates the measured analog signals that relate to patient’s blood cell counts. The diagnostic outcome, based on the cell counts, is protected through an encryption scheme, before sending out the data through the smartphone to the cloud for analysis. The outcome of the analysis is then sent back to the device for decryption and user notifications. The proposed data protection scheme is realized for a prototype consisting of a biosensor connected to a smartphone. A smartphone app and cloud-based service that perform the analysis have also been implemented. This design guarantees the user’s privacy by considering the smartphone and the cloud server possibly untrusted: the proposed setup assumes a curious but honest security model. The proposed encryption scheme infringes no overhead while the decryption and diagnostic calculations are <1 s in average. An evaluation of the strength of the signal obfuscation mechanism via a general case study is also presented. This design provides a domain specific encryption scheme for analog measurements with a small trusted computing base and provides an alternative to digital ciphers.