Bacterial Identification and Susceptibility Testing

This study improves ML-based C. difficile typing from MALDI-TOF MS spectra by using data augmentation to reduce performance variability across different experimental conditions.

The paper introduces a large mass spectra dataset for antimicrobial resistance prediction, evaluates its performance, and analyzes the impact of data heterogeneity on machine learning models.

This paper uses spectral flow cytometry to detect antibiotic-resistant bacteria by analyzing autofluorescence signatures, showing differences between resistant and susceptible strains.

The paper improves antimicrobial resistance prediction using MALDI-TOF MS by incorporating phylogenetic structure through hierarchical clustering, increasing average precision for specific resistance predictions.

This study evaluates the FilmArray BCID panel's accuracy in detecting pathogens in blood cultures with activated carbon powder, finding high consistency with clinical methods.

This study uses mass spectrometry and machine learning to predict antibiotic susceptibility of Staphylococcus aureus, enabling faster and more accurate treatment decisions.

This study evaluates NGS-based methods for detecting healthcare-acquired pathogens in simulated skin transmission events using metagenomic and metatranscriptomic sequencing.

Bac-PULCE enables rapid bacterial strain and AMR profiling using CRISPR-based enrichment and long-read sequencing with minimal infrastructure.

This study evaluates metagenomic next-generation sequencing (mNGS) of cerebrospinal fluid for diagnosing central nervous system infections, finding it detects pathogens in 36.6% of cases with some false positives and negatives.

This study evaluates the MinION sequencer's ability to identify and differentiate upper respiratory disease pathogens using targeted and whole genome sequencing, achieving strain-level accuracy despite sequencing errors.

This paper develops a rapid workflow for detecting and characterizing bacteremia pathogens using metagenomic sequencing and targeted PCR, achieving sensitivities of 1x10^6 and 1x10^5 cells/mL for untargeted and targeted detection, respectively.

The paper presents a fast antibiotic susceptibility testing method that determines bacterial growth rate responses to antibiotics in under 25 minutes using microfluidic imaging.