Jennifer A. Dionne, Associate Professor of Materials Science and Engineering
Niaz Banaei, Associate Professor of Pathology and Medicine
Stefano Ermon, Assistant Professor of Computer Science
Sanjiv Sam Gambhir, Professor of Radiology, and by courtesy, Bioengineering and Materials Science and Engineering
Stefanie Jeffrey, Professor of Surgery and Chief of Surgical Oncology Research
Manu Prakash, Assistant Professor of Bioengineering
Sindy Tang, Assistant Professor of Mechanical Engineering
How do we achieve effective yet affordable healthcare everywhere?
Bacterial infections such as meningitis, tuberculosis, and pneumonia may not often make headlines, but are responsible for more deaths than AIDS and many cancers and rank among the most expensive medical conditions to treat. Unfortunately, diagnosis can take days, even in state-of-the-art laboratories, increasing patient mortality and accelerating the spread of infectious disease. Such slow diagnostics also promotes the misuse of antibiotics, and consequently, the evolution of antibiotic-resistant pathogens.
Here, we propose a new technology that uses direct optical characterization to rapidly detect and identify pathogens in whole blood. Unlike traditional culture-based methods, our technology promises to detect and identify a single bacterium in a large (>10mL) volume of whole-blood. To do so, our platform relies on Surface-Enhanced Raman Scattering (SERS) – inelastic photon scattering with single-molecule sensitivity enabled by metallic nanostructures. Because of the unique molecular structure of a pathogen’s cell membrane, each bacterial species has a specific SERS signature that can be used for identification. This SERS signature can also be used to monitor changes to cell membrane structure and cell viability upon antibiotic exposure, facilitating real-time antibiotic susceptibility testing.
To enable sensitive and fast pathogen identification, we collect a blood sample, mix it with metallic nanoparticles, and then split the sample into microdroplets using inkjet-type biological printing. Each microdroplet contains one cell and a homogeneous dispersion of SERS-optimized nanoparticles. Once printed, a customized SERS scanner screens the microdroplet array for bacteria. Millions of droplets are simultaneously imaged while machine learning algorithms identify the presence or absence of bacteria, as well as the species, strain, and antibiotic susceptibility of the pathogen.
Since our proposed technology does not rely on culturing, it reduces the time required for pathogen identification from days to minutes. It is also label-free and does not require cell-specific tagging; hence it can be generalized to all bacteria. Moreover, this assay relies on an easy-to-use automated printing-scanning platform that does not require specially trained personnel or advanced laboratory setups to operate. Thus, this platform can be deployed in both rural and urban clinics without significant financial and time overhead.
Our interdisciplinary team, including investigators from the Schools of Engineering, Medicine, and Business, will develop the technology, conduct clinical trials, and deploy the assay to urban and rural clinics. We will also develop novel engineering and business strategies to keep costs low, distributing the economic impact between developed and developing nations. By democratizing fast, accurate pathogen diagnosis, our platform promises to reduce healthcare costs and improve millions of lives.