An objective diagnosis of gout and calcium pyrophosphate deposition disease with machine learning of Raman spectra acquired in a point-of-care setting
Authors' conclusion
We here demonstrate that potentially complex Raman spectra from clinical patient samples can be successfully classified by a machine learning approach, resulting in an objective diagnosis independent of the opinion of the medical examiner.
Study context
- Population
- Synovial fluid samples from patients with various rheumatic diseases, three centres246 training, 200 validation samples
- Participants
- 446 participants
- Comparison
- Machine-learning classification of Raman spectra vs trained observers' classification
- Outcome
- Accuracy of classification of CPP and MSU crystals in synovial fluid Raman spectra
Methodological detailsmachine learning (principal component analysis followed by support vector machine), point-of-care setting, three centres
- Machine learning (principal component analysis followed by support vector machine)
- Point-of-care setting
- Three centres
Performance
As reported in the abstract
Accuracy of automated classification of Raman spectra (validation set)
Accuracy for classification of CPP (2023 ACR/EULAR CPPD classification criteria), 95% CI 92.3, 98.3
96.0%
Accuracy for classification of monosodium urate (MSU) (2015 ACR/EULAR gout classification criteria), 95% CI 87.9, 95.7
92.5%
Overall accuracy for classification of pathological crystals, 95% CI 82.7, 92.2
88.0%
Further findings
Overall accuracy for classification of pathological crystals
- Overall accuracy
- 88.0% (95% CI: 82.7, 92.2)
- Model able to discriminate between pathological crystals, artifacts and other particles such as microplastics
Original publication
An objective diagnosis of gout and calcium pyrophosphate deposition disease with machine learning of Raman spectra acquired in a point-of-care setting. Rheumatology (Oxford). Published 2 September 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →