Frontiers in immunology · 2026
Research on a rapid detection method for systemic lupus erythematosus based on Raman spectroscopy.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Rao Jingbao, Lin Yu, Zhong Xiaozhen, Huang Zhimin, Xiong Dan
Аннотация
BACKGROUND: Systemic lupus erythematosus (SLE) is a multisystem autoimmune disease. Clinical diagnosis is hampered by subjective assessment tools, the low sensitivity and frequent cross-reactivity of traditional biomarkers, and by a limited ability to reflect disease status in real time; moreover, SLE presents phenotypes similar to those of other autoimmune conditions such as RA. There is an urgent need for non-invasive, efficient technologies to improve the accuracy of early diagnosis.
METHODS: This study established a rapid method for detecting systemic lupus erythematosus (SLE) using Raman spectroscopy. Raman spectra of peripheral blood mononuclear cells (PBMCs) were obtained from 28 SLE patients and 69 controls, the latter group comprising individuals with rheumatoid arthritis, Sjögren's syndrome, dermatomyositis, and healthy subjects. After preprocessing the raw spectra, the data were input to several classification models. Finally, a multilayer perceptron (MLP) integrated the feature outputs from each classifier to build an ensemble learner that produced the classification and prediction results.
RESULTS: The LDA results indicated that SLE samples were highly dispersed, reflecting disease heterogeneity, whereas disease control samples formed a tight cluster, indicating molecular homogeneity. In PBMCs from SLE patients, Raman peak intensities for monosaccharides and phenylalanine were higher than in the healthy control group, while the reverse was true for guanine and nucleic acids. Some metabolite peak intensities were higher in the disease comparator group. The integrated model showed the best performance, achieving an accuracy of 93% and an AUC of 0.98; it outperformed single models and effectively discriminated between SLE and control populations.
CONCLUSION: This proof-of-concept study demonstrates the feasibility of using PBMC Raman spectroscopy with ensemble learning to discriminate SLE from non-SLE populations (including healthy controls and other autoimmune diseases). The observed spectral differences provide descriptive molecular fingerprints that may guide future hypothesis-driven investigations.
METHODS: This study established a rapid method for detecting systemic lupus erythematosus (SLE) using Raman spectroscopy. Raman spectra of peripheral blood mononuclear cells (PBMCs) were obtained from 28 SLE patients and 69 controls, the latter group comprising individuals with rheumatoid arthritis, Sjögren's syndrome, dermatomyositis, and healthy subjects. After preprocessing the raw spectra, the data were input to several classification models. Finally, a multilayer perceptron (MLP) integrated the feature outputs from each classifier to build an ensemble learner that produced the classification and prediction results.
RESULTS: The LDA results indicated that SLE samples were highly dispersed, reflecting disease heterogeneity, whereas disease control samples formed a tight cluster, indicating molecular homogeneity. In PBMCs from SLE patients, Raman peak intensities for monosaccharides and phenylalanine were higher than in the healthy control group, while the reverse was true for guanine and nucleic acids. Some metabolite peak intensities were higher in the disease comparator group. The integrated model showed the best performance, achieving an accuracy of 93% and an AUC of 0.98; it outperformed single models and effectively discriminated between SLE and control populations.
CONCLUSION: This proof-of-concept study demonstrates the feasibility of using PBMC Raman spectroscopy with ensemble learning to discriminate SLE from non-SLE populations (including healthy controls and other autoimmune diseases). The observed spectral differences provide descriptive molecular fingerprints that may guide future hypothesis-driven investigations.