Longitudinal interpretability of deep learning based breast cancer risk prediction.

IF 3.3 3区 医学 Q2 ENGINEERING, BIOMEDICAL
Zan Klanecek, Yao-Kuan Wang, Tobias Wagner, Lesley Cockmartin, Nicholas Marshall, Brayden Schott, Ali Deatsch, Andrej Studen, Katja Jarm, Mateja Krajc, Miloš Vrhovec, Hilde Bosmans, Robert Jeraj
{"title":"Longitudinal interpretability of deep learning based breast cancer risk prediction.","authors":"Zan Klanecek, Yao-Kuan Wang, Tobias Wagner, Lesley Cockmartin, Nicholas Marshall, Brayden Schott, Ali Deatsch, Andrej Studen, Katja Jarm, Mateja Krajc, Miloš Vrhovec, Hilde Bosmans, Robert Jeraj","doi":"10.1088/1361-6560/ad9db3","DOIUrl":null,"url":null,"abstract":"<p><p><i>Objective.</i>Deep-learning-based models have achieved state-of-the-art breast cancer risk (BCR) prediction performance. However, these models are highly complex, and the underlying mechanisms of BCR prediction are not fully understood. Key questions include whether these models can detect breast morphologic changes that lead to cancer. These findings would boost confidence in utilizing BCR models in practice and provide clinicians with new perspectives. In this work, we aimed to determine when oncogenic processes in the breast provide sufficient signal for the models to detect these changes.<i>Approach.</i>In total, 1210 screening mammograms were collected for patients screened at different times before the cancer was screen-detected and 2400 mammograms for patients with at least ten years of follow-up. MIRAI, a BCR risk prediction model, was used to estimate the BCR. Attribution heterogeneity was defined as the relative difference between the attributions obtained from the right and left breasts using one of the eight interpretability techniques. Model reliance on the side of the breast with cancer was quantified with AUC. The Mann-Whitney U test was used to check for significant differences in median absolute Attribution Heterogeneity between cancer patients and healthy individuals.<i>Results.</i>All tested attribution methods showed a similar longitudinal trend, where the model reliance on the side of the breast with cancer was the highest for the 0-1 years-to-cancer interval (AUC = 0.85-0.95), dropped for the 1-3 years-to-cancer interval (AUC = 0.64-0.71), and remained above the threshold for random performance for the 3-5 years-to-cancer interval (AUC = 0.51-0.58). For all eight attribution methods, the median values of absolute attribution heterogeneity were significantly larger for patients diagnosed with cancer at one point (<i>p</i>< 0.01).<i>Significance.</i>Interpretability of BCR prediction has revealed that long-term predictions (beyond three years) are most likely based on typical breast characteristics, such as breast density; for mid-term predictions (one to three years), the model appears to detect early signs of tumor development, while for short-term predictions (up to a year), the BCR model essentially functions as a breast cancer detection model.</p>","PeriodicalId":20185,"journal":{"name":"Physics in medicine and biology","volume":" ","pages":""},"PeriodicalIF":3.3000,"publicationDate":"2024-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Physics in medicine and biology","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1088/1361-6560/ad9db3","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, BIOMEDICAL","Score":null,"Total":0}
引用次数: 0

Abstract

Objective.Deep-learning-based models have achieved state-of-the-art breast cancer risk (BCR) prediction performance. However, these models are highly complex, and the underlying mechanisms of BCR prediction are not fully understood. Key questions include whether these models can detect breast morphologic changes that lead to cancer. These findings would boost confidence in utilizing BCR models in practice and provide clinicians with new perspectives. In this work, we aimed to determine when oncogenic processes in the breast provide sufficient signal for the models to detect these changes.Approach.In total, 1210 screening mammograms were collected for patients screened at different times before the cancer was screen-detected and 2400 mammograms for patients with at least ten years of follow-up. MIRAI, a BCR risk prediction model, was used to estimate the BCR. Attribution heterogeneity was defined as the relative difference between the attributions obtained from the right and left breasts using one of the eight interpretability techniques. Model reliance on the side of the breast with cancer was quantified with AUC. The Mann-Whitney U test was used to check for significant differences in median absolute Attribution Heterogeneity between cancer patients and healthy individuals.Results.All tested attribution methods showed a similar longitudinal trend, where the model reliance on the side of the breast with cancer was the highest for the 0-1 years-to-cancer interval (AUC = 0.85-0.95), dropped for the 1-3 years-to-cancer interval (AUC = 0.64-0.71), and remained above the threshold for random performance for the 3-5 years-to-cancer interval (AUC = 0.51-0.58). For all eight attribution methods, the median values of absolute attribution heterogeneity were significantly larger for patients diagnosed with cancer at one point (p< 0.01).Significance.Interpretability of BCR prediction has revealed that long-term predictions (beyond three years) are most likely based on typical breast characteristics, such as breast density; for mid-term predictions (one to three years), the model appears to detect early signs of tumor development, while for short-term predictions (up to a year), the BCR model essentially functions as a breast cancer detection model.

基于深度学习的乳腺癌风险预测的纵向可解释性。
目的:基于深度学习的模型已经实现了最先进的乳腺癌风险(BCR)预测性能。然而,这些模型非常复杂,BCR预测的潜在机制尚不完全清楚。关键问题包括这些模型能否检测出导致癌症的乳腺形态学变化。这些发现将增强在实践中使用BCR模型的信心,并为临床医生提供新的视角。在这项工作中,我们的目的是确定乳腺中的致癌过程何时为模型提供足够的信号来检测这些变化。方法:总共收集了1210张筛查乳房x光片,用于筛查癌症前不同时间筛查的患者,并收集了2400张筛查乳房x光片,用于至少随访10年的患者。采用BCR风险预测模型MIRAI对BCR进行估计。归因异质性定义为使用八种可解释性技术中的一种所获得的左右乳房归因的相对差异。用AUC量化对乳腺癌一侧的模型依赖性。采用Mann-Whitney U检验检验癌症患者与健康个体的中位绝对归因异质性是否存在显著差异。所有测试的归因方法都显示出类似的纵向趋势,在0-1年至癌症区间(AUC=0.85-0.95),模型对乳腺癌一侧的依赖程度最高,在1-3年至癌症区间(AUC=0.64-0.71)下降,在3-5年至癌症区间(AUC=0.51-0.58)仍然高于随机表现的阈值。对于所有八种归因方法,绝对归因异质性的中位数值在诊断为癌症的患者中显著大于某一点(p
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
Physics in medicine and biology
Physics in medicine and biology 医学-工程:生物医学
CiteScore
6.50
自引率
14.30%
发文量
409
审稿时长
2 months
期刊介绍: The development and application of theoretical, computational and experimental physics to medicine, physiology and biology. Topics covered are: therapy physics (including ionizing and non-ionizing radiation); biomedical imaging (e.g. x-ray, magnetic resonance, ultrasound, optical and nuclear imaging); image-guided interventions; image reconstruction and analysis (including kinetic modelling); artificial intelligence in biomedical physics and analysis; nanoparticles in imaging and therapy; radiobiology; radiation protection and patient dose monitoring; radiation dosimetry
×
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:481959085
Book学术官方微信