Discriminating fingerprints of chronic neuropathic pain following spinal cord injury using artificial neural networks and mass spectrometry analysis of female mice serum
Language English Country Great Britain, England Media print-electronic
Document type Journal Article
PubMed
39455011
DOI
10.1016/j.neuint.2024.105890
PII: S0197-0186(24)00217-1
Knihovny.cz E-resources
- Keywords
- Artificial intelligence, Artificial neural networks, Central neuropathic pain, MALDI-TOF MS, Spectral profiles, mass spectrometry, spinal cord injury,
- MeSH
- Biomarkers blood MeSH
- Chronic Pain blood diagnosis etiology MeSH
- Mice, Inbred C57BL MeSH
- Mice MeSH
- Neuralgia * blood diagnosis etiology MeSH
- Neural Networks, Computer * MeSH
- Spinal Cord Injuries * complications blood MeSH
- Spectrometry, Mass, Matrix-Assisted Laser Desorption-Ionization * methods MeSH
- Animals MeSH
- Check Tag
- Mice MeSH
- Female MeSH
- Animals MeSH
- Publication type
- Journal Article MeSH
- Names of Substances
- Biomarkers MeSH
Spinal cord injury (SCI) often leads to central neuropathic pain, a condition associated with significant morbidity and is challenging in terms of the clinical management. Despite extensive efforts, identifying effective biomarkers for neuropathic pain remains elusive. Here we propose a novel approach combining matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) with artificial neural networks (ANNs) to discriminate between mass spectral profiles associated with chronic neuropathic pain induced by SCI in female mice. Functional evaluations revealed persistent chronic neuropathic pain following mild SCI as well as minor locomotor disruptions, confirming the value of collecting serum samples. Mass spectra analysis revealed distinct profiles between chronic SCI and sham controls. On applying ANNs, 100% success was achieved in distinguishing between the two groups through the intensities of m/z peaks. Additionally, the ANNs also successfully discriminated between chronic and acute SCI phases. When reflexive pain response data was integrated with mass spectra, there was no improvement in the classification. These findings offer insights into neuropathic pain pathophysiology and underscore the potential of MALDI-TOF MS coupled with ANNs as a diagnostic tool for chronic neuropathic pain, potentially guiding attempts to discover biomarkers and develop treatments.
References provided by Crossref.org
