TY - JOUR TI - Security through physical dynamics in medical and manufacturing platforms DO - https://doi.org/doi:10.7282/t3-zszr-9p85 PY - 2019 AB - Portable medical diagnostic or point-of-care (POC) devices enable the transition from reactive, clinical-based healthcare to preventive, patient-centered management. POC devices have been shown to have accuracy and performance equivalent of laboratory equipment. However, this does not remove medical practitioner’s involvement in result analysis. The diagnostic results would be exchanged between the patients and medical practitioners. In this framework, a trustworthy and usable healthcare requires not only effective diagnostics but also lightweight user privacy-preserving capabilities. On the other hand, Additive Manufacturing (AM) or 3D printing has been found applicable in manufacturing safety-critical parts and medical implants. AM is projected to reach 50% market potential by 2038. Due to its potential expansion, AM has become an attractive target to the attackers. Initiatives have been undertaken to study the impact of malicious attacks to critical components. Correspondingly, we develop an end-to-end malicious attack detection in AM in this study. This thesis focuses on the developing of solutions for diagnostic information and user privacy protection leveraging the physical system designs of biomedical device and the malicious detection in manufacturing platform. The thesis will focus on three major tasks: information protection, user privacy protection, and malicious attacks detection. In information protection, we introduce a diagnostic information protection for impedance flow cytometry. The encryption scheme is developed leveraging the design of the microfluidic device. The sensor of a microfluidic device is designed to be mechanically re-configurable to enable the encryption of information. In user privacy protection, we present a protection scheme leveraging functionality of impedance flow cytometry. In this scheme, we perform a domain specific user authentication by embedding the synthetic microbeads in the test device as authentication strings. This alternative method removes the authentication burden from users and protects their privacy by preventing them from linking personal information to the test results. Applying the similar physical design concept, we present the solution for malicious attack detection in additive manufacturing or 3D printing. The scheme incorporates real-time tracking of instrument and post production material analysis to reconstruct the physical design model for verification and detection of malicious modification. This allows the end user to accurately verify and manage the 3D printed models in real-time. Furthermore, we present a design of portable malicious material detection device in additive manufacturing. The design utilizes the lock-in amplifier architecture to detect the change of material during printing. The portable device can be used in real-time malicious detection of material modification in traditional 3D printing. KW - Cyber physical security KW - Electrical and Computer Engineering KW - Three-dimensional printing -- Software -- Security measures LA - English ER -