Microfluidic technologies are rapidly being designed and integrated into research, clinical, and forensic laboratories as a cost-effective alternative to benchtop techniques. However, a multitude of security concerns, from assay design to data collection and privacy, remain to be addressed. We modeled several attacks on a simple polymerase chain reaction (PCR) protocol, which is integral to many DNA analysis and identification. After identifying potential vulnerabilities, we performed mock attacks using benchtop techniques to demonstrate the catastrophic impact on the outcome of the DNA-identification process. We also propose a countermeasure solution, in which DNA samples are each uniquely barcoded in advance of PCR analysis, demonstrate the feasibility of our approach using benchtop techniques, and discuss how sample barcoding could be mapped onto microfluidic platforms in the future.
Researchers have a surprisingly diverse armamentarium for committing research misconduct — typically to turn a non-significant result into a significant one. This talk will outline some of the common mechanisms for misconduct as well as some of the ways that these mechanisms have been defeated.
Microfluidic technology represents a strong opportunity for providing sensitive, low-cost and rapid diagnostics at the point-of-care. Previous studies on microfluidic-based assays have proven their potential for portable diagnostics that require small volumes of samples and reagents. However, the generation of the precise flow of liquids, the efficient integration of reagents, the high-throughput fabrication of microfluidic chips with sealed channels, and the vulnerability of devices to counterfeiting still represent a significant challenge for commercialization and widespread use of such technologies. In the “Precision Diagnostics” group of IBM Research-Zurich, we have been developing new concepts and a library of microfluidic elements to address some of these challenges. In this presentation, I will share our vision on a portable diagnostic platform, which uses capillary-driven microfluidics and peripherals enabling communication with a smartphone. I will present available technologies and missing components towards this vision and will also discuss our recent efforts on the development of security features to protect rapid diagnostic tests from counterfeiting.
Many microfluidic biochips contain pneumatically-actuated on-chip valves and pumps for manipulating small volumes of fluids. These microfluidic valves and pumps are typically controlled by pneumatic signals (pressure and vacuum) provided by off-chip computer-controlled electronic valves. While the electronic portions of this control system (the computer and electronic valves) can be monitored for errors and attacks using traditional techniques, the pneumatic portions (the pneumatic lines connecting the electronic valves to the biochip and the on-chip pneumatic channels and valves) are vulnerable to attacks and errors that are currently difficult or impossible to detect. In this talk, I will share our recent work developing strategies for detecting errors and attacks in real time in microfluidic chips using on-chip, valve-based pneumatic logic. These pneumatic “circuits” use monolithic membrane valves like transistors to perform complex computations on-chip. We show that a pneumatic “parity bit” calculated on-chip using pneumatic logic can be used to not only detect valve actuation errors but also enable the chip to gracefully recover from such errors. We also demonstrate pneumatic logic checksums and other on-chip techniques for detecting attacks and errors. As microfluidic biochips grow more complex and find more important applications in day-to-day life, pneumatic logic should play a major role in protecting these chips from accidental errors and intentional attacks.
Organ-on-a-chip devices have opened up new possibilities to accelerate the drug discovery process in a cost-efficient manner than the traditional screening methods currently deployed in the pharmaceutical industry. Using patient-derived cells and organoids, these biomimetic microfluidic devices offer a precisely controlled microenvironment that can closely mimic the patient organ’s responses to drug treatment. As a result, patient-specific drugs can rapidly be screened and allows for a “personalized” treatment. If successfully implemented, the era of precision medicine could soon become a reality. However, a massively parallel configuration of organ-on-a-chip devices in a highly automated biofactory-like setting can be highly vulnerable to external or internal manipulations. As a potential scenario, the pressure, temperature, CO2 level or flow rates of cell perfusion could be manipulated or alternatively, the experimental data manipulated or even stolen in the cyberspace. In view of such vulnerability, it would be highly advisable to come up with a more secure design of microfluidic chips that is less prone to manipulation by emulating the innate immune response of the human body. To create such a responsive system, multiple autonomous biosensors can be embedded into the organ-on-a-chip devices that sense the microenvironment continuously and notify if there are any unusual changes beyond threshold values. In this talk, the state of the art in organ-on-a-chip technology will be presented along with several implantable biosensors that could sense and actuate “immune responses” on-chip and finally, strategies how to respond to intentional manipulations discussed.