The validation of threshold decision ruls and calculator for APhiG algoritm for clarification of prostate cancer staging before treatment

IF 0.1 Q4 ONCOLOGY
N. Sergeeva, T. Skachkova, N. Marshutina, K. M. Nushko, I. Shevchuk, M. Nazirov, B. Alekseev, S. Pirogov, E. Yurkov, V. Gitis, A. Kaprin
{"title":"The validation of threshold decision ruls and calculator for APhiG algoritm for clarification of prostate cancer staging before treatment","authors":"N. Sergeeva, T. Skachkova, N. Marshutina, K. M. Nushko, I. Shevchuk, M. Nazirov, B. Alekseev, S. Pirogov, E. Yurkov, V. Gitis, A. Kaprin","doi":"10.17650/1726-9776-2020-16-1-43-53","DOIUrl":null,"url":null,"abstract":"<jats:p><jats:bold><jats:italic>Background.</jats:italic></jats:bold><jats:bold><jats:italic> </jats:italic></jats:bold><jats:italic>W</jats:italic><jats:italic>e</jats:italic><jats:italic> </jats:italic><jats:italic>have</jats:italic><jats:italic> </jats:italic><jats:italic>previously</jats:italic><jats:italic> </jats:italic><jats:italic>described</jats:italic><jats:italic> </jats:italic><jats:italic>an</jats:italic><jats:italic> </jats:italic><jats:italic>algorithm</jats:italic><jats:italic> </jats:italic><jats:italic>APhiG</jats:italic><jats:italic> </jats:italic><jats:italic>(Age</jats:italic><jats:italic> </jats:italic><jats:italic>of</jats:italic><jats:italic> </jats:italic><jats:italic>patients,</jats:italic><jats:italic> </jats:italic><jats:italic>Prostate</jats:italic><jats:italic> </jats:italic><jats:italic>health</jats:italic><jats:italic> </jats:italic><jats:italic>index</jats:italic><jats:italic> </jats:italic><jats:italic>and</jats:italic><jats:italic> </jats:italic><jats:italic>Gleason</jats:italic><jats:italic> </jats:italic><jats:italic>score),</jats:italic><jats:italic> </jats:italic><jats:italic>for</jats:italic><jats:italic> </jats:italic><jats:italic>staging</jats:italic><jats:italic> </jats:italic><jats:italic>of</jats:italic><jats:italic> </jats:italic><jats:italic>prostate</jats:italic><jats:italic> </jats:italic><jats:italic>cancer</jats:italic><jats:italic> </jats:italic><jats:italic>before</jats:italic><jats:italic> </jats:italic><jats:italic>treatment.</jats:italic><jats:italic> </jats:italic><jats:italic>The</jats:italic><jats:italic> </jats:italic><jats:italic>algorithm</jats:italic><jats:italic> </jats:italic><jats:italic>was</jats:italic><jats:italic> </jats:italic><jats:italic>developed</jats:italic><jats:italic> </jats:italic><jats:italic>by</jats:italic><jats:italic> </jats:italic><jats:italic>logistic</jats:italic><jats:italic> </jats:italic><jats:italic>regression</jats:italic><jats:italic> </jats:italic><jats:italic>on</jats:italic><jats:italic> </jats:italic><jats:italic>a training dataset and validated on a validation dataset (VD). <jats:bold>Objective. </jats:bold>Validation of threshold decision rules and a program for APhiG calculation on the VD.</jats:italic></jats:p><jats:p><jats:bold><jats:italic>Materials and methods. </jats:italic></jats:bold><jats:italic>R</jats:italic><jats:italic>OC</jats:italic><jats:italic> </jats:italic><jats:italic>curve analysis on VD (83 cases).</jats:italic></jats:p><jats:p><jats:bold><jats:italic>Results</jats:italic></jats:bold><jats:bold><jats:italic> </jats:italic></jats:bold><jats:bold><jats:italic>and</jats:italic></jats:bold><jats:bold><jats:italic> </jats:italic></jats:bold><jats:bold><jats:italic>conclusion.</jats:italic></jats:bold><jats:bold><jats:italic> </jats:italic></jats:bold><jats:italic>It</jats:italic><jats:italic> </jats:italic><jats:italic>was</jats:italic><jats:italic> </jats:italic><jats:italic>shown</jats:italic><jats:italic> </jats:italic><jats:italic>that</jats:italic><jats:italic> </jats:italic><jats:italic>sensitivity</jats:italic><jats:italic>,</jats:italic><jats:italic> </jats:italic><jats:italic>specificity</jats:italic><jats:italic>,</jats:italic><jats:italic> </jats:italic><jats:italic>positive</jats:italic><jats:italic> </jats:italic><jats:italic>and</jats:italic><jats:italic> </jats:italic><jats:italic>negative</jats:italic><jats:italic> </jats:italic><jats:italic>predictive</jats:italic><jats:italic> </jats:italic><jats:italic>value,</jats:italic><jats:italic> </jats:italic><jats:italic>diagnostic</jats:italic><jats:italic> </jats:italic><jats:italic>accuracy</jats:italic><jats:italic> </jats:italic><jats:italic>threshold</jats:italic><jats:italic> </jats:italic><jats:italic>decision</jats:italic><jats:italic> </jats:italic><jats:italic>rules and area under the curve (AUC) for APhiG in the VD (n = 83) not significantly different from those indicators in the training dataset (n = 337), which was the basis for the algorithm APhiG development.</jats:italic></jats:p>","PeriodicalId":42924,"journal":{"name":"Onkourologiya","volume":"16 1","pages":"43-53"},"PeriodicalIF":0.1000,"publicationDate":"2020-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Onkourologiya","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.17650/1726-9776-2020-16-1-43-53","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"ONCOLOGY","Score":null,"Total":0}
引用次数: 1

Abstract

Background. We have previously described an algorithm APhiG (Age of patients, Prostate health index and Gleason score), for staging of prostate cancer before treatment. The algorithm was developed by logistic regression on a training dataset and validated on a validation dataset (VD). Objective. Validation of threshold decision rules and a program for APhiG calculation on the VD.Materials and methods. ROC curve analysis on VD (83 cases).Results and conclusion. It was shown that sensitivity, specificity, positive and negative predictive value, diagnostic accuracy threshold decision rules and area under the curve (AUC) for APhiG in the VD (n = 83) not significantly different from those indicators in the training dataset (n = 337), which was the basis for the algorithm APhiG development.
阈值决策规则及APhiG算法用于明确前列腺癌治疗前分期的计算器验证
背景。我们之前描述了一种算法APhiG(患者年龄、前列腺健康指数和Gleason评分),用于治疗前前列腺癌的分期。该算法在训练数据集上进行了逻辑回归,并在验证数据集(VD)上进行了验证。目标。阈值判定规则的验证及在VD上的APhiG计算程序。材料和方法。83例VD的ROC曲线分析。结果与结论。结果表明,VD (n = 83)中APhiG的敏感性、特异性、阳性预测值和阴性预测值、诊断准确率阈值决策规则和曲线下面积(AUC)与训练数据集(n = 337)中的指标无显著差异,这是APhiG算法开发的基础。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Onkourologiya
Onkourologiya ONCOLOGY-
CiteScore
0.40
自引率
0.00%
发文量
59
审稿时长
10 weeks
期刊介绍: The main objective of the journal "Cancer urology" is publishing up-to-date information about scientific clinical researches, diagnostics, treatment of oncologic urological diseases. The aim of the edition is to inform the experts on oncologic urology about achievements in this area, to build understanding of the necessary integrated interdisciplinary approach in therapy, alongside with urologists, combining efforts of doctors of various specialties (cardiologists, pediatricians, chemotherapeutists et al.), to contribute to raising the effectiveness of oncologic patients’ treatment.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信