Cargando…

Machine-learning-based analysis of the sensitivity and specificity on lipid-lowering effect of one-month-administered statins

Few predictive studies have been reported on the efficacy of atorvastatin in reducing lipoprotein cholesterol to be qualified after 1-month course of treatment in different individuals. A total of 14,180 community-based residents aged ≥ 65 received health checkup, 1013 of whom had low-density lipopr...

Descripción completa

Detalles Bibliográficos
Autores principales: Liu, Huiqin, Jiao, Ronghong, Wang, Lingling, Feng, Fei, Zhao, Xiaohui, Yang, Juan
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Lippincott Williams & Wilkins 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9981436/
https://www.ncbi.nlm.nih.gov/pubmed/36862920
http://dx.doi.org/10.1097/MD.0000000000033139
_version_ 1784900102484131840
author Liu, Huiqin
Jiao, Ronghong
Wang, Lingling
Feng, Fei
Zhao, Xiaohui
Yang, Juan
author_facet Liu, Huiqin
Jiao, Ronghong
Wang, Lingling
Feng, Fei
Zhao, Xiaohui
Yang, Juan
author_sort Liu, Huiqin
collection PubMed
description Few predictive studies have been reported on the efficacy of atorvastatin in reducing lipoprotein cholesterol to be qualified after 1-month course of treatment in different individuals. A total of 14,180 community-based residents aged ≥ 65 received health checkup, 1013 of whom had low-density lipoprotein (LDL) higher than 2.6mmol/L so that they were put on 1-month course of treatment with atorvastatin. At its completion, lipoprotein cholesterol was measured again. With < 2.6 mmol/L considered as the treatment standard, 411 individuals were judged as the qualified group, and 602, and as the unqualified group. The basic sociodemographic features covered 57 items. The data were randomly divided into train sets and test ones. The recursive random-forest algorithm was applied to predicting the patients response to atorvastatin, the recursive feature elimination method, to screening all the physical indicators. The overall accuracy, sensitivity and specificity were calculated, respectively, and so were the receiver operator characteristic curve and the area under the curve of the test set. In the prediction model on the efficacy of 1-month treatment of statins for LDL, the sensitivity, 86.86%; and the specificity, 94.83%. In the prediction model on the efficacy of the same treatment for triglyceride, the sensitivity, 71.21%; and the specificity, 73.46%. As to the prediction of total cholesterol, the sensitivity, 94.38%; and the specificity, 96.55%. And in the case of high-density lipoprotein (HDL), the sensitivity, 84.86%; and the specificity, 100%. recursive feature elimination analysis showed that total cholesterol was the most important feature of atorvastatin efficacy of reducing LDL; that HDL was the most important one of its efficacies of reducing triglycerides; that LDL was the most important one of its efficacies of reducing total cholesterol; and that triglyceride was the most important one of its efficacies of reducing HDL. Random-forest can help predict whether atorvastatin efficacy of reducing lipoprotein cholesterol to be qualified after 1-month course of treatment in different individuals.
format Online
Article
Text
id pubmed-9981436
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher Lippincott Williams & Wilkins
record_format MEDLINE/PubMed
spelling pubmed-99814362023-03-04 Machine-learning-based analysis of the sensitivity and specificity on lipid-lowering effect of one-month-administered statins Liu, Huiqin Jiao, Ronghong Wang, Lingling Feng, Fei Zhao, Xiaohui Yang, Juan Medicine (Baltimore) 4200 Few predictive studies have been reported on the efficacy of atorvastatin in reducing lipoprotein cholesterol to be qualified after 1-month course of treatment in different individuals. A total of 14,180 community-based residents aged ≥ 65 received health checkup, 1013 of whom had low-density lipoprotein (LDL) higher than 2.6mmol/L so that they were put on 1-month course of treatment with atorvastatin. At its completion, lipoprotein cholesterol was measured again. With < 2.6 mmol/L considered as the treatment standard, 411 individuals were judged as the qualified group, and 602, and as the unqualified group. The basic sociodemographic features covered 57 items. The data were randomly divided into train sets and test ones. The recursive random-forest algorithm was applied to predicting the patients response to atorvastatin, the recursive feature elimination method, to screening all the physical indicators. The overall accuracy, sensitivity and specificity were calculated, respectively, and so were the receiver operator characteristic curve and the area under the curve of the test set. In the prediction model on the efficacy of 1-month treatment of statins for LDL, the sensitivity, 86.86%; and the specificity, 94.83%. In the prediction model on the efficacy of the same treatment for triglyceride, the sensitivity, 71.21%; and the specificity, 73.46%. As to the prediction of total cholesterol, the sensitivity, 94.38%; and the specificity, 96.55%. And in the case of high-density lipoprotein (HDL), the sensitivity, 84.86%; and the specificity, 100%. recursive feature elimination analysis showed that total cholesterol was the most important feature of atorvastatin efficacy of reducing LDL; that HDL was the most important one of its efficacies of reducing triglycerides; that LDL was the most important one of its efficacies of reducing total cholesterol; and that triglyceride was the most important one of its efficacies of reducing HDL. Random-forest can help predict whether atorvastatin efficacy of reducing lipoprotein cholesterol to be qualified after 1-month course of treatment in different individuals. Lippincott Williams & Wilkins 2023-03-03 /pmc/articles/PMC9981436/ /pubmed/36862920 http://dx.doi.org/10.1097/MD.0000000000033139 Text en Copyright © 2023 the Author(s). Published by Wolters Kluwer Health, Inc. https://creativecommons.org/licenses/by-nc/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial License 4.0 (CCBY-NC) (https://creativecommons.org/licenses/by-nc/4.0/) , where it is permissible to download, share, remix, transform, and buildup the work provided it is properly cited. The work cannot be used commercially without permission from the journal.
spellingShingle 4200
Liu, Huiqin
Jiao, Ronghong
Wang, Lingling
Feng, Fei
Zhao, Xiaohui
Yang, Juan
Machine-learning-based analysis of the sensitivity and specificity on lipid-lowering effect of one-month-administered statins
title Machine-learning-based analysis of the sensitivity and specificity on lipid-lowering effect of one-month-administered statins
title_full Machine-learning-based analysis of the sensitivity and specificity on lipid-lowering effect of one-month-administered statins
title_fullStr Machine-learning-based analysis of the sensitivity and specificity on lipid-lowering effect of one-month-administered statins
title_full_unstemmed Machine-learning-based analysis of the sensitivity and specificity on lipid-lowering effect of one-month-administered statins
title_short Machine-learning-based analysis of the sensitivity and specificity on lipid-lowering effect of one-month-administered statins
title_sort machine-learning-based analysis of the sensitivity and specificity on lipid-lowering effect of one-month-administered statins
topic 4200
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9981436/
https://www.ncbi.nlm.nih.gov/pubmed/36862920
http://dx.doi.org/10.1097/MD.0000000000033139
work_keys_str_mv AT liuhuiqin machinelearningbasedanalysisofthesensitivityandspecificityonlipidloweringeffectofonemonthadministeredstatins
AT jiaoronghong machinelearningbasedanalysisofthesensitivityandspecificityonlipidloweringeffectofonemonthadministeredstatins
AT wanglingling machinelearningbasedanalysisofthesensitivityandspecificityonlipidloweringeffectofonemonthadministeredstatins
AT fengfei machinelearningbasedanalysisofthesensitivityandspecificityonlipidloweringeffectofonemonthadministeredstatins
AT zhaoxiaohui machinelearningbasedanalysisofthesensitivityandspecificityonlipidloweringeffectofonemonthadministeredstatins
AT yangjuan machinelearningbasedanalysisofthesensitivityandspecificityonlipidloweringeffectofonemonthadministeredstatins