Establishment and analysis of a novel diagnostic model for systemic juvenile idiopathic arthritis based on machine learning
Authors' conclusion
We successfully developed a new SJIA diagnostic model that can be used for a novel aid in the identification of SJIA.
Study context
- Population
- Peripheral blood mononuclear cell (PBMC) gene expression samples from Gene Expression Omnibus (GEO): training set merged from GSE11,907-GPL96, GSE8650-GPL96 and GSE13,501; validation set GSE7753training dataset
- Participants
- 125 vs 92
- Comparison
- sJIA vs healthy samples
- Systemic juvenile idiopathic arthritis (SJIA)n = 125
- Healthyn = 92reference
- Outcome
- RF diagnostic model based on 4 key genes (ALDH1A1, CEACAM1, YBX3, SLC6A8) distinguishing sJIA from healthy samples
Methodological detailsmachine learning, Lasso, RF-RFE and random forest (RF) classifier feature selection, 10-fold cross-validation with 5 repetitions, external validation dataset
- Machine learning
- Lasso, RF-RFE and RF classifier feature selection
- 10-fold cross-validation with 5 repetitions
- External validation dataset
Performance
As reported in the abstract
Mean AUC, 5-fold cross-validation
greater than 0.95
AUC, validation dataset
0.990
Original publication
Establishment and analysis of a novel diagnostic model for systemic juvenile idiopathic arthritis based on machine learning. Pediatr Rheumatol Online J. Published 19 January 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →