Dimensionality reduction in 3D causal deep learning for neuroimage generation: an evaluation study.

IF 1.7 Q3 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING
Journal of Medical Imaging Pub Date : 2025-03-01 Epub Date: 2025-04-22 DOI:10.1117/1.JMI.12.2.024506
Erik Y Ohara, Vibujithan Vigneshwaran, Raissa Souza, Finn G Vamosi, Matthias Wilms, Nils D Forkert
{"title":"Dimensionality reduction in 3D causal deep learning for neuroimage generation: an evaluation study.","authors":"Erik Y Ohara, Vibujithan Vigneshwaran, Raissa Souza, Finn G Vamosi, Matthias Wilms, Nils D Forkert","doi":"10.1117/1.JMI.12.2.024506","DOIUrl":null,"url":null,"abstract":"<p><strong>Purpose: </strong>Causal deep learning (DL) using normalizing flows allows the generation of true counterfactual images, which is relevant for many medical applications such as explainability of decisions, image harmonization, and in-silico studies. However, such models are computationally expensive when applied directly to high-resolution 3D images and, therefore, require image dimensionality reduction (DR) to efficiently process the data. The goal of this work was to compare how different DR methods affect counterfactual neuroimage generation.</p><p><strong>Approach: </strong>Five DR techniques [2D principal component analysis (PCA), 2.5D PCA, 3D PCA, autoencoder, and Vector Quantised-Variational AutoEncoder] were applied to 23,692 3D brain images to create low-dimensional representations for causal DL model training. Convolutional neural networks were used to quantitatively evaluate age and sex changes on the counterfactual neuroimages. Age alterations were measured using the mean absolute error (MAE), whereas sex changes were assessed via classification accuracy.</p><p><strong>Results: </strong>The 2.5D PCA technique achieved the lowest MAE of 4.16 when changing the age variable of an original image. When sex was changed, the autoencoder embedding led to the highest classification accuracy of 97.84% while also significantly impacting the age variable predictions, increasing the MAE to 5.24 years. Overall, 3D PCA provided the best balance, with an age prediction MAE of 4.57 years while maintaining 94.01% sex classification accuracy when altering the age variable and 94.73% sex classification accuracy and the lowest age prediction MAE (3.84 years) when altering the sex variable.</p><p><strong>Conclusions: </strong>3D PCA appears to be the best-suited DR method for causal neuroimage analysis.</p>","PeriodicalId":47707,"journal":{"name":"Journal of Medical Imaging","volume":"12 2","pages":"024506"},"PeriodicalIF":1.7000,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12014944/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Medical Imaging","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1117/1.JMI.12.2.024506","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/4/22 0:00:00","PubModel":"Epub","JCR":"Q3","JCRName":"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING","Score":null,"Total":0}
引用次数: 0

Abstract

Purpose: Causal deep learning (DL) using normalizing flows allows the generation of true counterfactual images, which is relevant for many medical applications such as explainability of decisions, image harmonization, and in-silico studies. However, such models are computationally expensive when applied directly to high-resolution 3D images and, therefore, require image dimensionality reduction (DR) to efficiently process the data. The goal of this work was to compare how different DR methods affect counterfactual neuroimage generation.

Approach: Five DR techniques [2D principal component analysis (PCA), 2.5D PCA, 3D PCA, autoencoder, and Vector Quantised-Variational AutoEncoder] were applied to 23,692 3D brain images to create low-dimensional representations for causal DL model training. Convolutional neural networks were used to quantitatively evaluate age and sex changes on the counterfactual neuroimages. Age alterations were measured using the mean absolute error (MAE), whereas sex changes were assessed via classification accuracy.

Results: The 2.5D PCA technique achieved the lowest MAE of 4.16 when changing the age variable of an original image. When sex was changed, the autoencoder embedding led to the highest classification accuracy of 97.84% while also significantly impacting the age variable predictions, increasing the MAE to 5.24 years. Overall, 3D PCA provided the best balance, with an age prediction MAE of 4.57 years while maintaining 94.01% sex classification accuracy when altering the age variable and 94.73% sex classification accuracy and the lowest age prediction MAE (3.84 years) when altering the sex variable.

Conclusions: 3D PCA appears to be the best-suited DR method for causal neuroimage analysis.

神经图像生成的三维因果深度学习的降维:一项评估研究。
目的:使用归一化流的因果深度学习(DL)允许生成真正的反事实图像,这与许多医学应用相关,例如决策的可解释性,图像协调和计算机研究。然而,当直接应用于高分辨率3D图像时,这种模型的计算成本很高,因此需要图像降维(DR)来有效地处理数据。这项工作的目的是比较不同的DR方法如何影响反事实神经图像的生成。方法:将五种DR技术(2D主成分分析(PCA)、2.5D主成分分析、3D主成分分析、自编码器和矢量量化变分自编码器)应用于23,692张3D脑图像,为因果DL模型训练创建低维表示。使用卷积神经网络定量评价反事实神经图像上的年龄和性别变化。使用平均绝对误差(MAE)测量年龄变化,而通过分类准确性评估性别变化。结果:2.5D PCA技术在改变原始图像的年龄变量时,MAE最低,为4.16。当性别改变时,自编码器嵌入的分类准确率最高,达到97.84%,同时也显著影响年龄变量的预测,将MAE提高到5.24岁。总体而言,3D PCA提供了最好的平衡,改变年龄变量时,年龄预测MAE为4.57岁,性别分类准确率为94.01%;改变性别变量时,性别分类准确率为94.73%,年龄预测MAE最低(3.84岁)。结论:三维PCA似乎是最适合用于因果神经图像分析的DR方法。
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
Journal of Medical Imaging
Journal of Medical Imaging RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING-
CiteScore
4.10
自引率
4.20%
发文量
0
期刊介绍: JMI covers fundamental and translational research, as well as applications, focused on medical imaging, which continue to yield physical and biomedical advancements in the early detection, diagnostics, and therapy of disease as well as in the understanding of normal. The scope of JMI includes: Imaging physics, Tomographic reconstruction algorithms (such as those in CT and MRI), Image processing and deep learning, Computer-aided diagnosis and quantitative image analysis, Visualization and modeling, Picture archiving and communications systems (PACS), Image perception and observer performance, Technology assessment, Ultrasonic imaging, Image-guided procedures, Digital pathology, Biomedical applications of biomedical imaging. JMI allows for the peer-reviewed communication and archiving of scientific developments, translational and clinical applications, reviews, and recommendations for the field.
×
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