Construction and Validation of a Blood Routine-Based Prediction Model for Myelodysplastic Syndrome Shunts in Anemia Patients.

Background: This study aimed to construct and verify a prediction model of nomograms for blood routine examination in anemia patients and to provide shunt diagnosis and treatment of myelodysplastic syndrome (MDS) and non-MDS in anemia patients.

Methods: A total of 296 were divided into a train set and a test set at a ratio of 7:3. Least absolute shrinkage and selection operator (Lasso) regression was used to screen risk factors. Multivariate logistic regression analysis was used to construct a prediction model for MDS in anemia patients. The bootstrap resampling method was used to verify the model internally. The area under the curve of the receiver operating characteristic (ROC) curve and clinical decision curve were used to evaluate the prediction ability and clinical applicability of the nomogram prediction model.

Results: Among 207 anemia patients in the train set, the logistic regression analysis revealed that elevated MCV, increased RDW, and decreased PLT were all independent influencing factors of MDS. The nomogram prediction model constructed via the bootstrap method showed that the calibration curve fit well, indicating good model differentiation. The DCA curve and ROC curve showed that the net return rate of the clinical prediction model was high at any risk threshold and was verified by the validation set.

Conclusions: The constructed anemia shunt MDS prediction model is simple and accurate and has certain value for early clinical screening of MDS patients with anemia and accurate formulation of subsequent diagnostic and treatment measures.