Background: Although convolutional neural networks (CNN) achieve high diagnostic accuracy for detecting Alzheimer's disease (AD) dementia based on magnetic resonance imaging (MRI) scans, they are not yet applied in clinical routine. One important reason for this is a lack of model comprehensibility. Recently developed visualization methods for deriving CNN relevance maps may help to fill this gap. We investigated whether models with higher accuracy also rely more on discriminative brain regions predefined by prior knowledge. Methods: We trained a CNN for the detection of AD in N=663 T1-weighted MRI scans of patients with dementia and amnestic mild cognitive impairment (MCI) and verified the accuracy of the models via cross-validation and in three independent samples including N=1655 cases. We evaluated the association of relevance scores and hippocampus volume to validate the clinical utility of this approach. To improve model comprehensibility, we implemented an interactive visualization of 3D CNN relevance maps. Results: Across three independent datasets, group separation showed high accuracy for AD dementia vs. controls (AUC$\geq$0.92) and moderate accuracy for MCI vs. controls (AUC$\approx$0.75). Relevance maps indicated that hippocampal atrophy was considered as the most informative factor for AD detection, with additional contributions from atrophy in other cortical and subcortical regions. Relevance scores within the hippocampus were highly correlated with hippocampal volumes (Pearson's r$\approx$-0.86, p<0.001). Conclusion: The relevance maps highlighted atrophy in regions that we had hypothesized a priori. This strengthens the comprehensibility of the CNN models, which were trained in a purely data-driven manner based on the scans and diagnosis labels.


翻译:虽然 convoilal 神经神经网络(CNN) 在磁共振成像(MRI)扫描的基础上,在测出阿尔茨海默氏病(AD)痴呆症(AD)的诊断精确度很高,但还没有在临床常规中应用,其重要原因之一是缺乏模型的可理解性。最近开发的用于绘制CNN相关地图的视觉化方法可能有助于填补这一差距。我们调查了精确度较高的模型是否还更多地依赖先前知识所预先定义的具有歧视性的大脑区域。方法:我们培训了一台用于检测患有老年痴呆症(AD)的AD(AD) 0.63 T1加权MRI扫描病人(MCI),并核实了模型的准确性,包括N=1655案例。我们评估了相关评分和海马氏体运动量的关联性方法,为了提高模型的可理解性,我们实施了3DCNN相关地图的互动直观化。结果:在三个独立的数据集中,对患有痴呆症(AD dementia ralalalalalalalal dal) ad realtial dealate a deal dealate, macial dealtiques a a a a a lax lax lax a a lax a lax a a latix a lax a lax a lax a lax a dalticalticalticaltical demod.

0
下载
关闭预览

相关内容

机器学习系统设计系统评估标准
商业数据分析,39页ppt
专知会员服务
159+阅读 · 2020年6月2日
机器学习入门的经验与建议
专知会员服务
92+阅读 · 2019年10月10日
最新BERT相关论文清单,BERT-related Papers
专知会员服务
52+阅读 · 2019年9月29日
disentangled-representation-papers
CreateAMind
26+阅读 · 2018年9月12日
已删除
将门创投
3+阅读 · 2018年3月13日
VIP会员
相关VIP内容
商业数据分析,39页ppt
专知会员服务
159+阅读 · 2020年6月2日
机器学习入门的经验与建议
专知会员服务
92+阅读 · 2019年10月10日
最新BERT相关论文清单,BERT-related Papers
专知会员服务
52+阅读 · 2019年9月29日
相关资讯
disentangled-representation-papers
CreateAMind
26+阅读 · 2018年9月12日
已删除
将门创投
3+阅读 · 2018年3月13日
Top
微信扫码咨询专知VIP会员