利用广义加性模型(GA2M)和传统方法建立卵巢癌初级保健风险评估算法。

IF 2.2 4区 医学 Q1 MEDICINE, GENERAL & INTERNAL
Current Medical Research and Opinion Pub Date : 2025-06-01 Epub Date: 2025-07-25 DOI:10.1080/03007995.2025.2534467
Francesco Lapi, Ettore Marconi, Lorenzo Nuti, Iacopo Cricelli, Marco Gorini, Stefania Marcoli, Alessandro Rossi, Claudio Cricelli
{"title":"利用广义加性模型(GA2M)和传统方法建立卵巢癌初级保健风险评估算法。","authors":"Francesco Lapi, Ettore Marconi, Lorenzo Nuti, Iacopo Cricelli, Marco Gorini, Stefania Marcoli, Alessandro Rossi, Claudio Cricelli","doi":"10.1080/03007995.2025.2534467","DOIUrl":null,"url":null,"abstract":"<p><strong>Objectives: </strong>We aimed to evaluate models designed to support the exploration and early detection of potential Ovarian Cancer (OC) using either Machine Learning (ML) techniques or traditional methodologies, using primary care data. This evaluation aimed to facilitate appropriate and timely referrals to specialists.</p><p><strong>Methods: </strong>The Health Search database, containing healthcare records of 1 million adults, was used to predict OC cases among patients aged 18+ without prior OC diagnosis from 1 January 2002, to June 2021. GA<sup>2</sup>M, logistic regression, and Gradient Boosting Machines (GBM) were trained using 20 determinants, with prediction performance assessed via Area Under Curve (AUC) and Average Precision (AP). A traditional cohort design with nested case-control analysis developed a prediction algorithm using the same determinants, evaluating explained variation, AUC, and slope calibration.</p><p><strong>Results: </strong>Comparing the predictive performances of the three models, the AUC and AP for GA<sup>2</sup>M showed the highest values, which were equal to 69.1 and 0.7%, respectively. The GA<sup>2</sup>M-based algorithm outperformed the algorithm obtained with the traditional approach, which showed an overestimation of risk, as confirmed by a calibration slope of 1.75 along with an AUC of 55%.</p><p><strong>Conclusions: </strong>The GA<sup>2</sup>M-based algorithm outperformed the light GBM, logistic regression, and traditional models in predicting OC. This suggests that ML techniques are preferable for algorithms involving complex predictor interplays in assessing rare event risks, such as OC. The GA<sup>2</sup>M-based algorithm is reliable for OC prediction in primary care compared to traditional methods, indicating its potential for use in decision support systems for general practitioners.</p>","PeriodicalId":10814,"journal":{"name":"Current Medical Research and Opinion","volume":" ","pages":"939-948"},"PeriodicalIF":2.2000,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"To create an algorithm assessing the risk of ovarian cancer in primary care using generalized additive<sup>2</sup> model (GA<sup>2</sup>M) and traditional methods.\",\"authors\":\"Francesco Lapi, Ettore Marconi, Lorenzo Nuti, Iacopo Cricelli, Marco Gorini, Stefania Marcoli, Alessandro Rossi, Claudio Cricelli\",\"doi\":\"10.1080/03007995.2025.2534467\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Objectives: </strong>We aimed to evaluate models designed to support the exploration and early detection of potential Ovarian Cancer (OC) using either Machine Learning (ML) techniques or traditional methodologies, using primary care data. This evaluation aimed to facilitate appropriate and timely referrals to specialists.</p><p><strong>Methods: </strong>The Health Search database, containing healthcare records of 1 million adults, was used to predict OC cases among patients aged 18+ without prior OC diagnosis from 1 January 2002, to June 2021. GA<sup>2</sup>M, logistic regression, and Gradient Boosting Machines (GBM) were trained using 20 determinants, with prediction performance assessed via Area Under Curve (AUC) and Average Precision (AP). A traditional cohort design with nested case-control analysis developed a prediction algorithm using the same determinants, evaluating explained variation, AUC, and slope calibration.</p><p><strong>Results: </strong>Comparing the predictive performances of the three models, the AUC and AP for GA<sup>2</sup>M showed the highest values, which were equal to 69.1 and 0.7%, respectively. The GA<sup>2</sup>M-based algorithm outperformed the algorithm obtained with the traditional approach, which showed an overestimation of risk, as confirmed by a calibration slope of 1.75 along with an AUC of 55%.</p><p><strong>Conclusions: </strong>The GA<sup>2</sup>M-based algorithm outperformed the light GBM, logistic regression, and traditional models in predicting OC. This suggests that ML techniques are preferable for algorithms involving complex predictor interplays in assessing rare event risks, such as OC. The GA<sup>2</sup>M-based algorithm is reliable for OC prediction in primary care compared to traditional methods, indicating its potential for use in decision support systems for general practitioners.</p>\",\"PeriodicalId\":10814,\"journal\":{\"name\":\"Current Medical Research and Opinion\",\"volume\":\" \",\"pages\":\"939-948\"},\"PeriodicalIF\":2.2000,\"publicationDate\":\"2025-06-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Current Medical Research and Opinion\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1080/03007995.2025.2534467\",\"RegionNum\":4,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2025/7/25 0:00:00\",\"PubModel\":\"Epub\",\"JCR\":\"Q1\",\"JCRName\":\"MEDICINE, GENERAL & INTERNAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Current Medical Research and Opinion","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1080/03007995.2025.2534467","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/7/25 0:00:00","PubModel":"Epub","JCR":"Q1","JCRName":"MEDICINE, GENERAL & INTERNAL","Score":null,"Total":0}
引用次数: 0

摘要

目的:我们的目的是评估旨在支持探索和早期发现潜在卵巢癌(OC)的模型,使用机器学习(ML)技术或传统方法,使用初级保健数据。这项评估旨在促进适当和及时的转介给专家。方法:从2002年1月1日至2021年6月,使用Health Search数据库(包含100万成年人的医疗记录)预测未确诊的18岁以上OC患者的OC病例。使用20个决定因素训练GA2M、逻辑回归和梯度增强机(GBM),并通过曲线下面积(AUC)和平均精度(AP)评估预测性能。传统的队列设计与嵌套病例对照分析开发了一种预测算法,使用相同的决定因素,评估解释变异、AUC和斜率校准。结果:比较三种模型的预测性能,GA2M的AUC和AP值最高,分别为69.1%和0.7%。基于ga2m的算法优于传统方法获得的算法,后者显示出对风险的高估,正如校准斜率为1.75和AUC为55%所证实的那样。结论:基于ga2m的算法在预测OC方面优于轻GBM、逻辑回归和传统模型。这表明在评估罕见事件风险(如OC)时,ML技术更适合涉及复杂预测因子相互作用的算法。与传统方法相比,基于ga2m的算法在初级保健中对OC的预测是可靠的,这表明它在全科医生决策支持系统中的应用潜力。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
To create an algorithm assessing the risk of ovarian cancer in primary care using generalized additive2 model (GA2M) and traditional methods.

Objectives: We aimed to evaluate models designed to support the exploration and early detection of potential Ovarian Cancer (OC) using either Machine Learning (ML) techniques or traditional methodologies, using primary care data. This evaluation aimed to facilitate appropriate and timely referrals to specialists.

Methods: The Health Search database, containing healthcare records of 1 million adults, was used to predict OC cases among patients aged 18+ without prior OC diagnosis from 1 January 2002, to June 2021. GA2M, logistic regression, and Gradient Boosting Machines (GBM) were trained using 20 determinants, with prediction performance assessed via Area Under Curve (AUC) and Average Precision (AP). A traditional cohort design with nested case-control analysis developed a prediction algorithm using the same determinants, evaluating explained variation, AUC, and slope calibration.

Results: Comparing the predictive performances of the three models, the AUC and AP for GA2M showed the highest values, which were equal to 69.1 and 0.7%, respectively. The GA2M-based algorithm outperformed the algorithm obtained with the traditional approach, which showed an overestimation of risk, as confirmed by a calibration slope of 1.75 along with an AUC of 55%.

Conclusions: The GA2M-based algorithm outperformed the light GBM, logistic regression, and traditional models in predicting OC. This suggests that ML techniques are preferable for algorithms involving complex predictor interplays in assessing rare event risks, such as OC. The GA2M-based algorithm is reliable for OC prediction in primary care compared to traditional methods, indicating its potential for use in decision support systems for general practitioners.

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
Current Medical Research and Opinion
Current Medical Research and Opinion 医学-医学:内科
CiteScore
4.40
自引率
4.30%
发文量
247
审稿时长
3-8 weeks
期刊介绍: Current Medical Research and Opinion is a MEDLINE-indexed, peer-reviewed, international journal for the rapid publication of original research on new and existing drugs and therapies, Phase II-IV studies, and post-marketing investigations. Equivalence, safety and efficacy/effectiveness studies are especially encouraged. Preclinical, Phase I, pharmacoeconomic, outcomes and quality of life studies may also be considered if there is clear clinical relevance
×
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