Target Convergence Analysis of Cancer-Inspired Swarms for Early Disease Diagnosis and Targeted Collective Therapy

Sensing and perception is generally a challenging aspect of decision-making. In the nanoscale, however, these processes face further complications due to the physical limitations of devising the nanomachines with more limited perception, more noise, and fewer sensors. There is, hence, higher dependence on swarm sensing and perception of many nanomachines. Here, taking hardware and software bioinspiration, we propose Chemo-Mechanical Cancer-Inspired Swarm Perception (CMCISP) based on online nano fuzzy haptic feedback for early disease diagnosis and targeted therapy. Particularly, we use epithelial cancer cell’s scaffold as a carrier, its properties as a distributed perception mechanism, and its motility patterns as the swarm movements such as anti-durotaxis, blebbing, and chemotaxis. We implement the in-silico model of CMCISP using a hybrid computational framework of the cellular Potts model, swarm intelligence, and fuzzy decision-making. Furthermore, the target convergence of CMCISP is analytically proved using swarm control theory. Finally, several numerical experiments and validations for cancer target therapy, cellular stiffness measurement, anti-durotaxis movement, and robustness analysis are also conducted and compared with a mathematical chemotherapy model and authors’ previous works on targeted therapy. Results show improvements of up to 57.49% in early cancer detection, 26.64% in target convergence, and 68.01% in increased normoxic cell density. The study also reveals the strategy’s robustness to environmental/sensory noise by applying six SNR levels of 0, 2, 5, 10, 30, and 50 dB, with an average diagnosis error of only 0.98% and at most 2.51%.