A Proposed Pathogen Classification System

Jane Ferris, Dublin Institute of Technology
Aidan Meade, Dublin Institute of Technology
Patrick McHale, Dublin Institute of Technology
Brian MacNamee, Dublin Institute of Technology
John Kelleher, Dublin Institute of Technology

Document Type Conference Paper

Artificial Intelligence & Cognitive Sciences Conference, Dublin 2012.


Approximately 51% of patients that require treatment within hospital Intensive Care Units (ICU) acquire an infection [1]. Rapid identification of the pathogen and medical treatment has a significant impact on survival rates and reduces the associated treatment costs in hospitals and in the community. Bacterial identification with Vibrational Spectroscopy (VS by either Raman and Fourier Transform Infra-Red (FTIR) spectroscopy) has recently been proven to rapidly type bacterial strains using appropriate classification algorithms [2]. This poster details the challenges of classifying common pathogens using VS and machine learning methods to build a pathogen multi-classification system that would aim to deliver very high clinical classification accuracy.