An Efficient Algorithm for Optimizing the Test Path of Digital Microfluidic Biochips

Digital microfluidic biochips (DMFBs) have been widely used in biochemical experiments with high safety requirements. To ensure the reliability of the experiment, it is necessary to use test droplets to perform off-line and on-line testing for DMFBs. Previous random search algorithms are not fully combined with heuristic information, which increases the randomness of search progress and thus results in suboptimal test paths. To solve this problem, a new test path optimization method based on priority strategy and genetic algorithm is proposed, which reduces the blindness of test path search and improves the convergence effect of the random search algorithm. In this method, priority levels are randomly assigned to the edges of the chip test model. With the fluid constraints, a test droplet moves to the untraversed adjacent edge with the highest priority. If all adjacent edges have been traversed, Floyd algorithm is used to determine the shortest path from the test droplet to untraversed edges, and guide the test droplet to move along the shortest path. After determining the test path according to priority strategy, the priority level on the test path is optimized by genetic algorithm, so that the length of the path is gradually reduced by iteration. In this paper, a single test droplet is used to test given chips. The experimental results show that the proposed algorithm is efficient, and the shortest path length is equal to the length of the Euler path, indicating that the shortest test path has reached the optimal value. Moreover, for the "deadlock" problem of droplets in on-line testing process, we also provide a solution by using backoff operation.