Touchable computation: Computing-inspired bio-detection

Abstract

We propose a new computing-inspired bio-detection framework called touchable computation (TouchComp). Under the rubric of TouchComp, the best solution in the parameter space is associated with the target to be detected. A population of externally steerable agents locate the optimal solution by moving through the parameter space, whose landscape (objective function) may be altered by these agents but the location of the best solution remains unchanged. Thus, one can infer the parameter space by observing the movement of agents. The term “touchable” emphasizes the framework's similarity to controlling by touching the screen with a finger, where the external field for controlling and tracking acts as the finger. We apply the TouchComp model to cancer detection, where the target is the cancer, the parameter space is the tissue region at high risk of malignancy, and agents are nanorobots loaded with contrast medium molecules for tracking purpose. Given this analogy, we revisit the classical particle swarm optimization (PSO) algorithm and apply it to TouchComp in order to achieve effective cancer detection. The PSO is modified by taking into account realistic in vivo propagation, controlling, and tracking conditions of nanorobots. Finally, we present numerical examples to demonstrate the effectiveness of the proposed computing-inspired bio-detection strategy.