Nanosensor Location Estimation in the Human Circulatory System Using Machine Learning

The human body can be considered a complex natural network due to the variety of interconnections between the different body regions. One example is the network of blood vessels, where artificial communication channels can be rendered using nanosensors that travel in the bloodstream as collectors and carriers of information. Further advancing this vision, in this work we investigate the detection and localization capabilities of flowing nanosensors in the blood flow to report abnormalities in the human body. Specifically, we target the detection quorum sensing molecules and provide a methodology to evaluate its performance. The methodology consists of modeling the traveling path of nanosensors along the vessels through a Markov chain, and the use of machine learning (ML) models to compute their transition probabilities. We illustrate the resulting distribution of nanosensors in the body, which evidences a close match to expected results. We also evaluate their detection and localization capabilities in different body regions revealing their effectiveness to determine the presence of abnormalities in the human vessels.