AUTHOR=Lindtner Richard Andreas , Kampik Lukas , Noack Larissa , Putzer David , Arora Rohit , Coraça-Huber Débora Cristina , Schirmer Michael , Degenhart Gerald , Lackner Michaela , Popp Jürgen , Badzoka Jovan , Kappacher Christoph , Huck Christian Wolfgang , Pallua Johannes Dominikus 

TITLE=Effective detection of staphylococcal infections in human bone tissue using combined raman microscopy and micro-computed tomography

JOURNAL=Frontiers in Bioengineering and Biotechnology

VOLUME=Volume 13 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/bioengineering-and-biotechnology/articles/10.3389/fbioe.2025.1617285

DOI=10.3389/fbioe.2025.1617285

ISSN=2296-4185

ABSTRACT=Bone infections caused by Staphylococcus aureus and Staphylococcus epidermidis are serious complications in orthopedic surgery. These infections commonly occur in joint replacements, fracture management, and bone grafting procedures. Rapid and accurate pathogen-specific diagnostic methods are urgently needed to support early clinical decisions. Current culture-based methods are slow and delay effective treatment. This study evaluated the diagnostic value of combining Raman microscopy with high-resolution micro-computed tomography (micro-CT). Human bone samples, either uninfected or inoculated with S. aureus or S. epidermidis, were analyzed. Raman spectroscopy detected distinct spectral changes in inoculated bones, including reduced intensity of phosphate (v1PO43−), Amide III, and CH2 deformation bands. A single principal component explained 96%–98% of the variance in these infection-related markers. Specifically, the v1PO43− and CH2 deformation bands effectively differentiated between S. aureus and S. epidermidis infections, capturing 99%–100% variance. Micro-CT analysis showed significant structural changes in inoculated bones. Trabecular volume, number, and spacing were particularly affected. Among these, VOX-BV/TV and Mean1 best differentiated between S. aureus and S. epidermidis infections (both p < 0.0001). Support vector machine (SVM) classification repeated stratified k-folg cross-validation accurately detected inoculation status. Combining Raman and micro-CT features yielded moderately improved classification performance in pathogen-specific discrimination. These findings demonstrate that combining molecular (Raman spectroscopy) and structural (micro-CT) methods allows rapid, non-destructive diagnosis of bone infections. This multimodal approach may improve diagnostic precision, supports timely clinical decisions, and ultimately improves patient outcomes in orthopedic and trauma surgery.