Pathogenic bacteria characterization through portable optical scatter device and machine learning

Abstract

Rapid and accurate detection and characterization of pathogenic bacteria is critical for clinical diagnosis. Most selective clinical procedures are limited by their diagnostic speed, accuracy, and sensitivity challenges. In order to overcome these, we introduce a novel photonics-based, point-of-care device designed for rapid and accurate characterization of bacteria. The device is designed to capture optical scatter signatures unique to various pathogenic bacteria, which are analyzed using advanced clustering and machine learning techniques for characterization. Our preliminary results from controlled experiments show that our device successfully distinguishes bacteria genus with reasonable accuracy.