A practical Alzheimer’s disease classifier via brain imaging-based deep learning on 85,721 samples
文献类型:期刊论文
作者 | Lu,Bin12,13; Li,Hui-Xian12,13; Chang,Zhi-Kai12,13; Li,Le2; Chen,Ning-Xuan12,13; Zhu,Zhi-Chen12,13; Zhou,Hui-Xia11,12,13; Li,Xue-Ying1,10,13; Wang,Yu-Wei12,13; Cui,Shi-Xian1,10,13 |
刊名 | Journal of Big Data |
出版日期 | 2022-10-13 |
卷号 | 9期号:1 |
关键词 | Alzheimer’s disease Convolutional neural network Magnetic resonance brain imaging Sex differences Transfer learning |
DOI | 10.1186/s40537-022-00650-y |
通讯作者 | Yan,Chao-Gan(yancg@psych.ac.cn) |
英文摘要 | AbstractBeyond detecting brain lesions or tumors, comparatively little success has been attained in identifying brain disorders such as Alzheimer’s disease (AD), based on magnetic resonance imaging (MRI). Many machine learning algorithms to detect AD have been trained using limited training data, meaning they often generalize poorly when applied to scans from previously unseen scanners/populations. Therefore, we built a practical brain MRI-based AD diagnostic classifier using deep learning/transfer learning on a dataset of unprecedented size and diversity. A retrospective MRI dataset pooled from more than 217 sites/scanners constituted one of the largest brain MRI samples to date (85,721 scans from 50,876 participants) between January 2017 and August 2021. Next, a state-of-the-art deep convolutional neural network, Inception-ResNet-V2, was built as a sex classifier with high generalization capability. The sex classifier achieved 94.9% accuracy and served as a base model in transfer learning for the objective diagnosis of AD. After transfer learning, the model fine-tuned for AD classification achieved 90.9% accuracy in leave-sites-out cross-validation on the Alzheimer’s Disease Neuroimaging Initiative (ADNI, 6,857 samples) dataset and 94.5%/93.6%/91.1% accuracy for direct tests on three unseen independent datasets (AIBL, 669 samples / MIRIAD, 644 samples / OASIS, 1,123 samples). When this AD classifier was tested on brain images from unseen mild cognitive impairment (MCI) patients, MCI patients who converted to AD were 3 times more likely to be predicted as AD than MCI patients who did not convert (65.2% vs. 20.6%). Predicted scores from the AD classifier showed significant correlations with illness severity. In sum, the proposed AD classifier offers a medical-grade marker that has potential to be integrated into AD diagnostic practice. |
语种 | 英语 |
出版者 | Springer International Publishing |
WOS记录号 | BMC:10.1186/S40537-022-00650-Y |
源URL | [http://ir.psych.ac.cn/handle/311026/43676] |
专题 | 心理研究所_中国科学院行为科学重点实验室 |
通讯作者 | Yan,Chao-Gan |
作者单位 | 1.University of Chinese Academy of Science; Sino-Danish College 2.Beijing Language and Culture University; Center for Cognitive Science of Language 3.Chinese Academy of Sciences; Magnetic Resonance Imaging Research Center, Institute of Psychology 4.Institute of Psychology, Chinese Academy of Sciences; International Big-Data Center for Depression Research 5.Nathan Kline Institute for Psychiatric Research 6.NYU Grossman School of Medicine; Department of Child and Adolescent Psychiatry 7.University of Southern California; Imaging Genetics Center, Mark & Mary Stevens Institute for Neuroimaging & Informatics, Keck School of Medicine 8.Zhejiang University; Department of Radiology, The First Affiliated Hospital, College of Medicine 9.Fudan University; Department of Neurosurgery, Huashan Hospital 10.Sino-Danish Center for Education and Research |
推荐引用方式 GB/T 7714 | Lu,Bin,Li,Hui-Xian,Chang,Zhi-Kai,et al. A practical Alzheimer’s disease classifier via brain imaging-based deep learning on 85,721 samples[J]. Journal of Big Data,2022,9(1). |
APA | Lu,Bin.,Li,Hui-Xian.,Chang,Zhi-Kai.,Li,Le.,Chen,Ning-Xuan.,...&Yan,Chao-Gan.(2022).A practical Alzheimer’s disease classifier via brain imaging-based deep learning on 85,721 samples.Journal of Big Data,9(1). |
MLA | Lu,Bin,et al."A practical Alzheimer’s disease classifier via brain imaging-based deep learning on 85,721 samples".Journal of Big Data 9.1(2022). |
入库方式: OAI收割
来源:心理研究所
浏览0
下载0
收藏0
其他版本
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。