Related Experiment Video
Updated: Jan 17, 2026
MicroRNA Based Liquid Biopsy: The Experience of the Plasma miRNA Signature Classifier MSC for Lung Cancer Screening
Published on: October 26, 2017
MALDI-TOF MS and Machine Learning Explanations for the Detection of SARS-CoV‑2 Infection in Human Plasma:
Meritxell Deulofeu1, Esteban García-Cuesta2, Eladia María Peña-Méndez3
1Research Group of Clinical Anatomy, Embryology and Neuroscience (NEOMA), Department of Medical Sciences, University of Girona, Girona 17003, Catalonia, Spain.
Matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS) combined with machine learning effectively identifies SARS-CoV-2 infection and stratifies patient severity. This approach aids in diagnosing COVID-19 and evaluating patient outcomes.
Area of Science:
- Clinical proteomics and metabolomics
- Infectious disease diagnostics
- Machine learning in healthcare
Background:
- The COVID-19 pandemic presented significant challenges to hospital operations and necessitated advanced diagnostic and prognostic tools.
- Accurate risk assessment and patient stratification are crucial for managing severe COVID-19 cases and optimizing treatment strategies.
Purpose of the Study:
- To develop and validate a MALDI-TOF MS-based method for identifying SARS-CoV-2 infection using plasma samples.
- To classify patients based on disease severity (COVID-19, ICU, Post-ICU) using machine learning and spectral data.
- To identify potential protein biomarkers indicative of SARS-CoV-2 infection and disease progression.
Main Methods:
- Plasma samples were analyzed using MALDI-TOF MS to generate spectral fingerprints.
- Machine learning algorithms, including SHAP for model interpretability, were employed to analyze spectral data.
- A prospective study design was used to evaluate the discriminatory power of identified spectral features and potential biomarkers.
Main Results:
- The developed MALDI-TOF MS and ML model demonstrated significant discriminatory power between healthy controls and COVID-19 patients, as well as ICU and post-ICU patients.
- Distinct spectral signatures were identified, enabling the separation and identification of different patient groups based on infection status and severity.
- Differences in spectral profiles were observed between patients requiring ICU admission and those who did not, highlighting prognostic potential.
Conclusions:
- MALDI-TOF MS coupled with advanced machine learning algorithms offers a powerful tool for diagnosing COVID-19 and stratifying patients by disease severity.
- The study successfully identified potential biomarkers through spectral analysis, contributing to a deeper understanding of SARS-CoV-2 pathophysiology.
- This integrated approach provides a valuable method for clinical decision-making and patient management during infectious disease outbreaks.
