Stability of leukocyte research parameters over time on the Sysmex XN: How to quantify the changes in cell morphology

Construction of the prediction model for multiple myeloma based on machine learning

INTRODUCTION

The global burden of multiple myeloma (MM) is increasing every year. Here, we have developed machine learning models to provide a reference for the early detection of MM.

METHODS

A total of 465 patients and 150 healthy controls were enrolled in this retrospective study. Based on the variable screening strategy of least absolute shrinkage and selection operator (LASSO), three prediction models, logistic regression (LR), support vector machine (SVM), and random forest (RF), were established combining complete blood count (CBC) and cell population data (CPD) parameters in the training set (210 cases), and were verified in the validation set (90 cases) and test set (165 cases). The performance of each model was analyzed using receiver operating characteristic (ROC) curve, calibration curves, and decision curve analysis (DCA). Accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and area under the ROC curve (AUC) were applied to evaluate the models. Delong test was used to compare the AUC of the models.

RESULTS

Six parameters including RBC (1012/L), RDW-CV (%), IG (%), NE-WZ, LY-WX, and LY-WZ were screened out by LASSO to construct the model. Among the three models, the AUC of RF model in the training set, validation set, and test set were 0.956, 0.892, and 0.875, which were higher than those of LR model (0.901, 0.849, and 0.858) and SVM model (0.929, 0.868, and 0.846). Delong test showed that there were significant differences among the models in the training set, no significant differences in the validation set, and significant differences only between SVM and RF models in the test set. The calibration curve and DCA showed that the three models had good validity and feasibility, and the RF model performed best.

CONCLUSION

The proposed RF model may be a useful auxiliary tool for rapid screening of MM patients.

Reviewing the value of leukocytes cell population data (CPD) in the management of sepsis

Sepsis is a medical emergency that describes the body's systemic immune response to an infection and can lead to end-stage organic dysfunction and death. Despite the advances in understanding the pathophysiology of this syndrome and therapies, sepsis remains one of the leading causes of morbidity and mortality in critically ill patients. Early diagnosis and rapid intervention are essential to improve outcomes, which inspired the concept "golden hour," during which the correction of shock and organic dysfunction can improve the patients' outcomes. But the initial presentation of sepsis is often nonspecific and its severity is difficult to assess. Anomalies in temperature, heart and respiratory rates and leukocyte counts are manifestations of systemic inflammatory response syndrome (SIRS). Diagnosis, management and follow-up of patients with sepsis remains a challenge, and diverse biomarkers have been proposed for the timely diagnosis and prognosis of septic patients: lactic acid, procalcitonin (PCT), C-reactive protein, immature granulocytes. The host's initial response to infection is a humoral, cellular and neuroendocrine reaction to infection, and leukocytes interact with endothelial cells. The new generation of hematological analyzers incorporates technological innovations allowing to expand the information derived from the complete blood count: new leukocyte derived parameters are emerging as potentially useful markers in different clinical situations. Additional research parameters cell population data (CPD), characterizing different leukocyte populations have become available, and preliminary observations suggest their utility in the diagnosis of sepsis. This review emphasizes the value of CPD, reported by modern cellular counters for early recognition of sepsis, and therefore the potential improvement in patient outcomes.