Real-Time Macrophage Phenotype Identification Using Single Walled Carbon Nanotubes and Machine Learning

Abstract

Macrophages play a critical role in the immune system, as both pro and anti-inflammatory immune cells these are the first line of defense from foreign bodies. Although methods for identifying Macrophage Phenotype (M1, M2) currently exist, these methods are limiting to factors like time, cost, operational training, and sample size. Cost and time effective identification of macrophage phenotypes can not only lead to better therapeutic research but also help in creating a platform for real time monitoring of the wound healing process as well as other (macrophage related) disease progression.Single Walled Carbon Nanotubes are ideal optical biosensors due to their intrinsic near-Infrared (nIR) emission, which is highly sensitive, easily tunable and indefinitely photostable. Due to such desirable properties, non-covalent functionalization of these fluorescent nanotubes has resulted in them being used as in vitro and in vivo sensors for numerous applications. Even so, the full extent of nanotube applicability is yet to be explored due to their underdeveloped potential as next generation bioimaging and bio-sensing probes. Here we demonstrate a platform using DNA functionalized Single Walled Carbon Nanotubes and machine learning to identify macrophage phenotype with 99% accuracy and effectively estimate the number of each polarization state in a sample.