使用机器学习预测冠状病毒患者的阿尔茨海默氏症。

IF 1.4 4区医学 Q3 Medicine Iranian Journal of Public Health Pub Date : 2023-10-01 DOI:10.18502/ijph.v52i10.13856

Shahriar Mohammadi, Soraya Zarei, Hossain Jabbari

{"title":"使用机器学习预测冠状病毒患者的阿尔茨海默氏症。","authors":"Shahriar Mohammadi, Soraya Zarei, Hossain Jabbari","doi":"10.18502/ijph.v52i10.13856","DOIUrl":null,"url":null,"abstract":"Background: One of the negative effects of the COVID-19 illness, which has affected people all across the world, is Alzheimer's disease. Oblivion after COVID-19 has created a variety of issues for many people. Predicting this issue in COVID-19 patients can considerably lessen the severity of the problem.Methods: Alzheimer's disease was predicted in Iranian persons with COVID-19 in using three algorithms: Nave Bayes, Random Forest, and KNN. Data collected by private questioner from hospitals of Tehran Province, Iran, during Oct 2020 to Sep 2021. For ML models, performance is quantified using measures such as Precision, Recall, Accuracy, and F1-score.Results: The Nave Bayes, Random Forest algorithm has a prediction accuracy of higher than 80%. The predicted accuracy of the random forest algorithm was higher than the other two algorithms.Conclusion: The Random Forest algorithm outperformed the other two algorithms in predicting Alzheimer's disease in persons using COVID-19. The findings of this study could help persons with COVID-19 avoid Alzheimer's problems.","PeriodicalId":14685,"journal":{"name":"Iranian Journal of Public Health","volume":"52 10","pages":"2179-2185"},"PeriodicalIF":1.4000,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10612562/pdf/","citationCount":"0","resultStr":"{\"title\":\"Prediction of Alzheimer's in People with Coronavirus Using Machine Learning.\",\"authors\":\"Shahriar Mohammadi, Soraya Zarei, Hossain Jabbari\",\"doi\":\"10.18502/ijph.v52i10.13856\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Background: One of the negative effects of the COVID-19 illness, which has affected people all across the world, is Alzheimer's disease. Oblivion after COVID-19 has created a variety of issues for many people. Predicting this issue in COVID-19 patients can considerably lessen the severity of the problem.Methods: Alzheimer's disease was predicted in Iranian persons with COVID-19 in using three algorithms: Nave Bayes, Random Forest, and KNN. Data collected by private questioner from hospitals of Tehran Province, Iran, during Oct 2020 to Sep 2021. For ML models, performance is quantified using measures such as Precision, Recall, Accuracy, and F1-score.Results: The Nave Bayes, Random Forest algorithm has a prediction accuracy of higher than 80%. The predicted accuracy of the random forest algorithm was higher than the other two algorithms.Conclusion: The Random Forest algorithm outperformed the other two algorithms in predicting Alzheimer's disease in persons using COVID-19. The findings of this study could help persons with COVID-19 avoid Alzheimer's problems.\",\"PeriodicalId\":14685,\"journal\":{\"name\":\"Iranian Journal of Public Health\",\"volume\":\"52 10\",\"pages\":\"2179-2185\"},\"PeriodicalIF\":1.4000,\"publicationDate\":\"2023-10-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10612562/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Iranian Journal of Public Health\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.18502/ijph.v52i10.13856\",\"RegionNum\":4,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"Medicine\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Iranian Journal of Public Health","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.18502/ijph.v52i10.13856","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"Medicine","Score":null,"Total":0}

引用次数: 0

摘要

背景：新冠肺炎疾病的负面影响之一是阿尔茨海默病，该疾病已影响到世界各地的人们。新冠肺炎后的遗忘给许多人带来了各种各样的问题。在新冠肺炎患者中预测这一问题可以大大减轻问题的严重性。方法：使用Nave Bayes、Random Forest和KNN三种算法预测伊朗新冠肺炎患者的阿尔茨海默病。私人提问者在2020年10月至2021年9月期间从伊朗德黑兰省医院收集的数据。对于ML模型，使用Precision、Recall、Accuracy和F1-score等指标来量化性能。结果：Nave Bayes，Random Forest算法的预测准确率高于80%。随机森林算法的预测精度高于其他两种算法。结论：随机森林算法在预测新冠肺炎患者阿尔茨海默病方面优于其他两种算法。这项研究的发现可以帮助新冠肺炎患者避免阿尔茨海默病问题。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

摘要图片

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

Prediction of Alzheimer's in People with Coronavirus Using Machine Learning.

Background: One of the negative effects of the COVID-19 illness, which has affected people all across the world, is Alzheimer's disease. Oblivion after COVID-19 has created a variety of issues for many people. Predicting this issue in COVID-19 patients can considerably lessen the severity of the problem.

Methods: Alzheimer's disease was predicted in Iranian persons with COVID-19 in using three algorithms: Nave Bayes, Random Forest, and KNN. Data collected by private questioner from hospitals of Tehran Province, Iran, during Oct 2020 to Sep 2021. For ML models, performance is quantified using measures such as Precision, Recall, Accuracy, and F1-score.

Results: The Nave Bayes, Random Forest algorithm has a prediction accuracy of higher than 80%. The predicted accuracy of the random forest algorithm was higher than the other two algorithms.

Conclusion: The Random Forest algorithm outperformed the other two algorithms in predicting Alzheimer's disease in persons using COVID-19. The findings of this study could help persons with COVID-19 avoid Alzheimer's problems.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Iranian Journal of Public Health 医学-公共卫生、环境卫生与职业卫生

CiteScore

2.20

自引率

7.10%

发文量

审稿时长

2 months

期刊介绍： Iranian Journal of Public Health has been continuously published since 1971, as the only Journal in all health domains, with wide distribution (including WHO in Geneva and Cairo) in two languages (English and Persian). From 2001 issue, the Journal is published only in English language. During the last 41 years more than 2000 scientific research papers, results of health activities, surveys and services, have been published in this Journal. To meet the increasing demand of respected researchers, as of January 2012, the Journal is published monthly. I wish this will assist to promote the level of global knowledge. The main topics that the Journal would welcome are: Bioethics, Disaster and Health, Entomology, Epidemiology, Health and Environment, Health Economics, Health Services, Immunology, Medical Genetics, Mental Health, Microbiology, Nutrition and Food Safety, Occupational Health, Oral Health. We would be very delighted to receive your Original papers, Review Articles, Short communications, Case reports and Scientific Letters to the Editor on the above mentioned research areas.