AUTHOR=Shao Bing , Qu Youyang , Zhang Wei , Zhan Haihe , Li Zerong , Han Xingyu , Ma Mengchao , Du Zhimin 

TITLE=Machine Learning-Based Prediction Method for Tremors Induced by Tacrolimus in the Treatment of Nephrotic Syndrome

JOURNAL=Frontiers in Pharmacology

VOLUME=Volume 13 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/pharmacology/articles/10.3389/fphar.2022.708610

DOI=10.3389/fphar.2022.708610

ISSN=1663-9812

ABSTRACT=Tremors have been reported even with low dose of tacrolimus in patients of nephrotic syndrome and are responsible for hampering the day to day working of young active patients of Nephrotic syndrome.This study proposes a neural network model based on seven variables to predicting development of tremors following Tacrolimus. The sensitivity and specificity of this algorithm is high.. A total of 252 patients were included in this study out of which 39 (15.5%) experienced tremors. Other 181 patients (including 32 who experienced tremors) were randomly assigned into a training dataset, and the remaining were assigned into an external validation set. We used a recursive feature elimination algorithm to train the training dataset in turn through 10-fold cross-validation. The classification performance of the classifier was then used as the evaluation criterion for these subsets to find the optimal features subset. A neural network was used as a classification algorithm to accurately predict tremors using the optimal features subset. This model was subsequently tested in the validation dataset. The optimal features subset contained seven variables (creatinine, D-dimer, total protein, calcium ion, platelet distribution width, serum kalium, and fibrinogen), and the highest accuracy obtained was 0.8288. The neural network model based on these seven variables obtained an area under the curve (AUC) value of 0.9726, accuracy of 0.9345, sensitivity of 0.9712, and specificity of 0.7586 in the training set. Meanwhile, the external validation achieved an accuracy of 0.8214, sensitivity of 0.8378, and specificity of 0.7000 in the validation dataset. This model was capable of predicting tremors caused by tacrolimus with an excellent degree of accuracy, which can be beneficial in the treatment of nephrotic syndrome patients.