Concurrent testing of digital microfluidics-based biochips

We present a concurrent testing methodology for detecting catastrophic faults in digital microfluidics-based biochips and investigate the related problems of test planning and resource optimization. We first show that an integer linear programming model can be used to minimize testing time for a given hardware overhead, for example, droplet dispensing sources and capacitive sensing circuitry. Due to the NP-complete nature of the problem, we also develop efficient heuristic procedures to solve this optimization problem. We apply the proposed concurrent testing methodology to a droplet-based microfluidic array that was fabricated and used to perform multiplexed glucose and lactate assays. Experimental results show that the proposed test approach interleaves test application with the biomedical assays and prevents resource conflicts. The proposed method is therefore directed at ensuring high reliability and availability of bio-MEMS and lab-on-a-chip systems, as they are increasingly deployed for safety-critical applications.