Machine Learning Approaches to Identify Discriminative Signatures of Volatile Organic Compounds (VOCs) from Bacteria and Fungi Using SPME-DART-MS.

Rishikesan Kamaleswaran,Mehak Arora,Joshua M Levy,Annette Esper,Cassandra L Quave,Stephen C Zambrzycki,Jennifer K Frediani,Facundo M Fernández

doi:10.3390/metabo12030232

Abstract

Point-of-care screening tools are essential to expedite patient care and decrease reliance on slow diagnostic tools (e.g., microbial cultures) to identify pathogens and their associated antibiotic resistance. Analysis of volatile organic compounds (VOC) emitted from biological media has seen increased attention in recent years as a potential non-invasive diagnostic procedure. This work explores the use of solid phase micro-extraction (SPME) and ambient plasma ionization mass spectrometry (MS) to rapidly acquire VOC signatures of bacteria and fungi. The MS spectrum of each pathogen goes through a preprocessing and feature extraction pipeline. Various supervised and unsupervised machine learning (ML) classification algorithms are trained and evaluated on the extracted feature set. These are able to classify the type of pathogen as bacteria or fungi with high accuracy, while marked progress is also made in identifying specific strains of bacteria. This study presents a new approach for the identification of pathogens from VOC signatures collected using SPME and ambient ionization MS by training classifiers on just a few samples of data. This ambient plasma ionization and ML approach is robust, rapid, precise, and can potentially be used as a non-invasive clinical diagnostic tool for point-of-care applications.

Highlights

And accurate detection of pathogens in the human body can play a critical role in the early detection of infection, greatly improving prognosis and recovery [1]
Classification of Pathogens as Bacteria or Fungi Based on volatile organic compounds (VOC) Signatures
Unsupervised K-means clustering on Principal Component Anaysis (PCA)-transformed data (Figure 1) showed that all fungi were correctly clustered together, but some bacterial samples were assigned to the fungi cluster

Summary

Introduction

And accurate detection of pathogens in the human body can play a critical role in the early detection of infection, greatly improving prognosis and recovery [1]. There is a need for rapid, sensitive, low-cost, and non-invasive screening tools that can be deployed as point-of-care (POC) devices [4]. These POC tools employed in the proximity of patient care are essential to expedite patient care and decrease reliance on slower diagnostic techniques to identify pathogens and their associated antibiotic resistance [4]. Identification of pathogens through their VOC profiles by mass spectrometry (MS) has seen extensive research [1,2,10,11]. These studies have analyzed VOCs emissions in exhaled breath, blood, and urine, and have been successful in demonstrating the diagnostic potential of bacterial VOC profiling [12–14]. The work proposed in this study builds on this premise and presents a method for identifying a pathogen based on its VOC profile

Objectives

Methods

Results

Conclusion