Loza Tadesse1,2,3,Chi-Sing Ho3,Baba Ogunlade3,Chris Cundy3,Ahmed Shuaibi4,3,Fareeha Safir3,Pierre Khuri-Yakub3,Stefanie S. Jeffrey3,Stefano Ermon3,Amr Essawi5,3,Jennifer Dionne3
UC Berkeley1,Massachusetts Institute of Technology2,Stanford University3,Princeton University4,Cairo University5
Loza Tadesse1,2,3,Chi-Sing Ho3,Baba Ogunlade3,Chris Cundy3,Ahmed Shuaibi4,3,Fareeha Safir3,Pierre Khuri-Yakub3,Stefanie S. Jeffrey3,Stefano Ermon3,Amr Essawi5,3,Jennifer Dionne3
UC Berkeley1,Massachusetts Institute of Technology2,Stanford University3,Princeton University4,Cairo University5
Bacterial bloodstream infections account for over 40% of death in hospitals and are one of the most expensive medical conditions in the US. Current diagnostic methods are slow and costly, due to the long bacterial culturing step. Our work utilized Raman spectroscopy for rapid culture-free, sensitive, and specific bacterial identification and antibiotic susceptibility testing. Here, I will present three major milestones that bring Raman closer to clinical application by using machine learning and nanophotonics. First, we achieve high (>99%) species level classification accuracies across 30 major disease-causing bacterial species. Second, we showcase the first of its kind demonstration of a versatile and antibiotic co-incubation free susceptibility testing. Third, we develop a simple liquid well setup for clinical sample handling with uniform Raman spectral enhancement using gold nanorods. Overall, our work opens the door for clinical translation of novel spectroscopy based diagnostic tools for identifying bacterial infections, viral infections such as COVID-19 virus, early cancer detection and drug susceptibility testing by merging machine learning and nanophotonics.