文献速递神经MR

基于深度迁移学习的大脑MRI全自动阿尔茨海默病检测与分类

Deep transfer learning-based fully automated detection and classification of Alzheimer's disease on brain MRI.
PubMed 文献 · Br J Radiol · 2022-06-09

本研究利用ADNI数据集305例受试者的3D T1加权脑MRI,采用ResNet101、Xception和InceptionV3三种预训练卷积神经网络进行迁移学习,对阿尔茨海默病进行二分类和四分类,并在OASIS(n=30)和AIBL(n=60)外部测试集验证。InceptionV3-TL在ADNI测试集二分类和多分类准确率均为93.75%,AUC分别为92.0%和96.0%,在外部测试集也表现良好,提示深度迁移学习可作为辅助工具。

要点

  • InceptionV3-TL在ADNI测试集全图二分类和多分类准确率均为93.75%,AUC分别为92.0%和96.0%。
  • 以灰质分割图为输入时,二分类和多分类性能均高于白质和脑脊液分割图。
  • 在OASIS和AIBL外部测试集上,InceptionV3-TL二分类准确率/AUC分别为93.33%/93.0%和93.33%/95.0%。

核心结论

  • 深度迁移学习可用于脑MRI的AD自动检测与分类,在内部和外部测试数据中均表现出较高准确性和稳健性。

研究设计与样本

  • 研究采用ADNI数据集305例受试者的3D T1加权脑MRI,使用ResNet101、Xception和InceptionV3三种预训练CNN进行迁移学习,并与从头训练比较;外部测试集来自OASIS(n=30)和AIBL(n=60)。

关键影像表现/征象

  • Unet网络将MRI扫描分割为灰质、白质和脑脊液特征成分。
  • 以灰质分割作为输入时,二分类和多分类AD分类性能最高,优于白质和脑脊液分割。

诊断标准与关键数值

  • ADNI测试集全图二分类:InceptionV3-TL准确率93.75%,AUC 92.0%。
  • ADNI测试集全图多分类:InceptionV3-TL准确率93.75%,AUC 96.0%。
  • OASIS测试集全图二分类:InceptionV3-TL准确率93.33%,AUC 93.0%。
  • AIBL测试集全图二分类:InceptionV3-TL准确率93.33%,AUC 95.0%;多分类准确率90.0%,AUC 93.0%。

临床应用与阅片要点

  • 本研究提示深度迁移学习模型可能作为辅助工具,帮助临床医生形成客观意见和正确诊断。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

OBJECTIVES: To employ different automated convolutional neural network (CNN)-based transfer learning (TL) methods for both binary and multiclass classification of Alzheimer's disease (AD) using brain MRI.

METHODS: Herein, we applied three popular pre-trained CNN models (ResNet101, Xception, and InceptionV3) using a fine-tuned approach of TL on 3D T1-weighted brain MRI from a subset of ADNI dataset (n = 305 subjects). To evaluate power of TL, the aforementioned networks were also trained from scratch for performance comparison. Initially, Unet network segmentedthe MRI scans into characteristic components of gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF). The proposed networks were trained and tested over the pre-processed and augmented segmented and whole images for both binary (NC/AD + progressive mild cognitive impairment (pMCI)+stable MCI (sMCI)) and 4-class (AD/pMCI/sMCI/NC) classification. Also, two independent test sets from the OASIS (n = 30) and AIBL (n = 60) datasets were used to externally assess the performance of the proposed algorithms.

RESULTS: The proposed TL-based CNN models achieved better performance compared to the training CNN models from scratch. On the ADNI test set, InceptionV3-TL achieved the highest accuracy of 93.75% and AUC of 92.0% for binary classification, as well as the highest accuracy of 93.75% and AUC of 96.0% for multiclass classification of AD on the whole images. On the OASIS test set, InceptionV3-TL outperformed two other models by achieving 93.33% accuracy with 93.0% AUC in binary classification of AD on the whole images. On the AIBL test set, InceptionV3-TL also outperformed two other models in both binary and multiclass classification tasks on the whole MR images and achieved accuracy/AUC of 93.33%/95.0% and 90.0%/93.0%, respectively. The GM segment as input provided the highest performance in both binary and multiclass classification of AD, as compared to the WM and CSF segments.

CONCLUSION: This study demonstrates the potential of applying deep TL approach for automated detection and classification of AD using brain MRI with high accuracy and robustness across internal and external test data, suggesting that these models can possibly be used as a supportive tool to assist clinicians in creating objective opinion and correct diagnosis.

ADVANCES IN KNOWLEDGE: We used CNN-based TL approaches and the augmentation techniques to overcome the insufficient data problem. Our study provides evidence that deep TL algorithms can be used for both binary and multiclass classification of AD with high accuracy.

来源:Br J Radiol,查看原文。版权归原作者所有。

引用

  1. Ghaffari, Tavakoli, Pirzad Jahromi, et al. Deep transfer learning-based fully automated detection and classification of Alzheimer's disease on brain MRI. Br J Radiol. 2022. PMID: 35616643. doi:10.1259/bjr.20211253

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。