Deep Learning-Based Prediction of PET Amyloid Status Using MRI.

Donghoon Kim, Jon André Ottesen, Ashwin Kumar, Brandon C Ho, Elsa Bismuth, Christina B Young, Elizabeth Mormino, Greg Zaharchuk
{"title":"Deep Learning-Based Prediction of PET Amyloid Status Using MRI.","authors":"Donghoon Kim, Jon André Ottesen, Ashwin Kumar, Brandon C Ho, Elsa Bismuth, Christina B Young, Elizabeth Mormino, Greg Zaharchuk","doi":"10.3174/ajnr.A8899","DOIUrl":null,"url":null,"abstract":"<p><strong>Background and purpose: </strong>Identifying amyloid-beta (Aβ)-positive patients is essential for Alzheimer's disease (AD) clinical trials and disease-modifying treatments but currently requires PET or cerebrospinal fluid sampling. Previous MRI-based deep learning models, using only T1-weighted (T1w) images, have shown moderate performance.</p><p><strong>Materials and methods: </strong>Multi-contrast MRI and PET-based quantitative Aβ deposition were retrospectively obtained from three public datasets: ADNI, OASIS3, and A4. Aβ positivity was defined using each dataset's recommended centiloid threshold. Two EfficientNet models were trained to predict amyloid positivity: one using only T1w images and another incorporating both T1w and T2-FLAIR. Model performance was assessed using an internal held-out test set, evaluating AUC, accuracy, sensitivity, and specificity. External validation was conducted using an independent cohort from Stanford Alzheimer's Disease Research Center. DeLong's and McNemar's tests were used to compare AUC and accuracy, respectively.</p><p><strong>Results: </strong>A total of 4,056 exams (mean [SD] age: 71.6 [6.3] years; 55% female; 55% amyloid-positive) were used for network development, and 149 exams were used for external testing (mean [SD] age: 72.1 [9.6] years; 58% female; 56% amyloid-positive). The multi-contrast model outperformed the single-modality model in the internal held-out test set (AUC: 0.67, 95% CI: 0.65-0.70, <i>P</i> < 0.001; accuracy: 0.63, 95% CI: 0.62-0.65, <i>P</i> < 0.001) compared to the T1w-only model (AUC: 0.61; accuracy: 0.59). Among cognitive subgroups, the highest performance (AUC: 0.71) was observed in mild cognitive impairment. The multi-contrast model also demonstrated consistent performance in the external test set (AUC: 0.65, 95% CI: 0.60-0.71, <i>P</i> = 0.014; accuracy: 0.62, 95% CI: 0.58- 0.65, <i>P</i> < 0.001).</p><p><strong>Conclusions: </strong>The use of multi-contrast MRI, specifically incorporating T2-FLAIR in addition to T1w images, significantly improved the predictive accuracy of PET-determined amyloid status from MRI scans using a deep learning approach.</p><p><strong>Abbreviations: </strong>Aβ= amyloid-beta; AD= Alzheimer's disease; AUC= area under the receiver operating characteristic curve; CN= cognitively normal; MCI= mild cognitive impairment; T1w = T1-wegithed; T2-FLAIR = T2-weighted fluid attenuated inversion recovery; FBP=<sup>18</sup>F-florbetapir; FBB=<sup>18</sup>F-florbetaben; SUVR= standard uptake value ratio.</p>","PeriodicalId":93863,"journal":{"name":"AJNR. American journal of neuroradiology","volume":" ","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2025-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"AJNR. American journal of neuroradiology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.3174/ajnr.A8899","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Background and purpose: Identifying amyloid-beta (Aβ)-positive patients is essential for Alzheimer's disease (AD) clinical trials and disease-modifying treatments but currently requires PET or cerebrospinal fluid sampling. Previous MRI-based deep learning models, using only T1-weighted (T1w) images, have shown moderate performance.

Materials and methods: Multi-contrast MRI and PET-based quantitative Aβ deposition were retrospectively obtained from three public datasets: ADNI, OASIS3, and A4. Aβ positivity was defined using each dataset's recommended centiloid threshold. Two EfficientNet models were trained to predict amyloid positivity: one using only T1w images and another incorporating both T1w and T2-FLAIR. Model performance was assessed using an internal held-out test set, evaluating AUC, accuracy, sensitivity, and specificity. External validation was conducted using an independent cohort from Stanford Alzheimer's Disease Research Center. DeLong's and McNemar's tests were used to compare AUC and accuracy, respectively.

Results: A total of 4,056 exams (mean [SD] age: 71.6 [6.3] years; 55% female; 55% amyloid-positive) were used for network development, and 149 exams were used for external testing (mean [SD] age: 72.1 [9.6] years; 58% female; 56% amyloid-positive). The multi-contrast model outperformed the single-modality model in the internal held-out test set (AUC: 0.67, 95% CI: 0.65-0.70, P < 0.001; accuracy: 0.63, 95% CI: 0.62-0.65, P < 0.001) compared to the T1w-only model (AUC: 0.61; accuracy: 0.59). Among cognitive subgroups, the highest performance (AUC: 0.71) was observed in mild cognitive impairment. The multi-contrast model also demonstrated consistent performance in the external test set (AUC: 0.65, 95% CI: 0.60-0.71, P = 0.014; accuracy: 0.62, 95% CI: 0.58- 0.65, P < 0.001).

Conclusions: The use of multi-contrast MRI, specifically incorporating T2-FLAIR in addition to T1w images, significantly improved the predictive accuracy of PET-determined amyloid status from MRI scans using a deep learning approach.

Abbreviations: Aβ= amyloid-beta; AD= Alzheimer's disease; AUC= area under the receiver operating characteristic curve; CN= cognitively normal; MCI= mild cognitive impairment; T1w = T1-wegithed; T2-FLAIR = T2-weighted fluid attenuated inversion recovery; FBP=18F-florbetapir; FBB=18F-florbetaben; SUVR= standard uptake value ratio.

基于深度学习的PET淀粉样蛋白状态MRI预测。
背景和目的:识别β淀粉样蛋白(Aβ)阳性患者对于阿尔茨海默病(AD)临床试验和疾病改善治疗至关重要,但目前需要PET或脑脊液取样。以前基于mri的深度学习模型,仅使用t1加权(T1w)图像,表现一般。材料和方法:回顾性地从三个公共数据集(ADNI, OASIS3和A4)中获得了基于多层对比MRI和pet的定量Aβ沉积。使用每个数据集推荐的centiloid阈值定义Aβ阳性。两个EfficientNet模型被训练来预测淀粉样蛋白阳性:一个只使用T1w图像,另一个同时使用T1w和T2-FLAIR图像。使用内部测试集评估模型性能,评估AUC、准确性、灵敏度和特异性。外部验证使用来自斯坦福阿尔茨海默病研究中心的独立队列进行。DeLong’s和McNemar’s试验分别用于比较AUC和准确度。结果:共检查4056例(平均[SD]年龄:71.6[6.3]岁;55%的女性;55%淀粉样蛋白阳性)用于网络发展,149例检查用于外部测试(平均[SD]年龄:72.1[9.6]岁;58%的女性;amyloid-positive 56%)。多重对比模型在内撑检验集中优于单模态模型(AUC: 0.67, 95% CI: 0.65 ~ 0.70, P < 0.001;准确度:0.63,95% CI: 0.62-0.65, P < 0.001),与仅t1w模型相比(AUC: 0.61;准确性:0.59)。在认知亚组中,轻度认知障碍表现最高(AUC: 0.71)。多重对比模型在外部测试集中也表现出一致的性能(AUC: 0.65, 95% CI: 0.60-0.71, P = 0.014;准确度:0.62,95% CI: 0.58 ~ 0.65, P < 0.001)。结论:使用多层对比MRI,特别是结合T2-FLAIR和T1w图像,使用深度学习方法显着提高了pet确定的MRI扫描淀粉样蛋白状态的预测准确性。缩写:Aβ=淀粉样蛋白;AD=阿尔茨海默病;AUC=接收机工作特性曲线下面积;CN=认知正常;MCI=轻度认知障碍;T1w = t1加权;T2-FLAIR = t2加权流体衰减反演采收率;出口押汇= 18 f-florbetapir;FBB = 18 f-florbetaben;SUVR=标准摄取值比。
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
自引率
0.00%
发文量
0
×
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