Thesis (M.Sc.) - Cairo University - Faculty of Pharmacy - Department of Microbiology and Immunology

Invasive aspergillosis is a challenging infection that requires convenient, efficient, and cost-effective diagnostics. This study addresses the potential of infrared spectroscopy to satisfy this clinical need with the aid of machine learning by creating and assessing models that identify the presence of Aspergillus species in human blood plasma. For training of the machine learning models, clinical isolates of three Aspergillus species (A. fumigatus, A. flavus, and A. niger) in addition to Penicillium chrysogenum have been collected and identified by their macroscopic and microscopic morphology, and confirmed by MALDI-TOF MS. Two models, based on Partial Least Squares-Discriminant Analysis (PLS-DA), have been trained by a set of infrared spectral data of 9 Aspergillus-spiked (three of each collected Aspergillus species) and 7 Aspergillus-free plasma samples (including 2 samples spiked with P. chrysogenum), and a set of 200 spectral data simulated by oversampling these 16 samples. Two further models have also been trained by the same sets but with auto-scaling performed prior to PLS-DA.These models were assessed using 45 mock samples, simulating the challenging samples of patients at risk of invasive aspergillosis, including the presence of other common bloodstream pathogens (5 tested) and drugs (9 tested) as potential confounders.The pathogen tested as potential confounders are clinical isolates of Candida albicans, Klebsiella pneumoniae, Pseudomonas aeruginosa, Escherichia coli, and Staphylococcus epidermidis that have been collected and identified by conventional phenotypic method including culture morphology, microscopic examination, and biochemical reactions

Issued also as CD