Answers are grounded in this paper and cite their sources.
Objectives: To construct a prediction model for the disease activity of rheumatoid arthritis(RA) patients based on machine learning algorithms.Methodology: This was a retrospective study. The clinical data of 300 RA patients admitted to Affiliated Hospital of Hebei University from January 2023 to January 2025 were retrospectively selected. The patients were randomly divided into a modeling group (n = 210) and a validation group (n = 90) at a ratio of 7:3, with those in the modeling group further divided into a stable group and an active group according to the disease activity. Lasso regression was used to screen for influencing factors. Internal validation was conducted using Bootstrap, with ROC and calibration curves used to assess the discrimination and consistency of the nomogram model. Results: Differences were observed in ESR, CRP, anti-CCP antibody, ASO, RF, AKA, IgA, and IgG between the two groups(P< 0.05). Among the five machine learning algorithm models, the AUC of the Logistic regression model for predicting the disease activity of RA patients was 0.864, making this model optimal. Additionally, the Logistic analysis suggested ESR, CRP, anti-CCP antibody, ASO, RF, and AKA as risk factors for disease activity in RA patients(P< 0.05).Conclusion: ESR, CRP, anti-CCP antibody, ASO, RF, and AKA are influencing factors of disease activity in RA patients. The nomogram prediction model for disease activity in RA patients constructed based on machine learning algorithms is beneficial for clinicians to identify high-risk patients in early stages.
No takes yet. Share an insight, caveat, or question.
Qi et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: