From the Department of Radiology and Research Institute of Radiology (C.H.S., W.H.S., S.J.K.).
From the Department of Radiology and Research Institute of Radiology (C.H.S., W.H.S., S.J.K.)
AJNR Am J Neuroradiol. 2020 Dec;41(12):2227-2234. doi: 10.3174/ajnr.A6848. Epub 2020 Nov 5.
Limited evidence has suggested that a deep learning automatic brain segmentation and classification method, based on T1-weighted brain MR images, can predict Alzheimer disease. Our aim was to develop and validate a deep learning-based automatic brain segmentation and classification algorithm for the diagnosis of Alzheimer disease using 3D T1-weighted brain MR images.
A deep learning-based algorithm was developed using a dataset of T1-weighted brain MR images in consecutive patients with Alzheimer disease and mild cognitive impairment. We developed a 2-step algorithm using a convolutional neural network to perform brain parcellation followed by 3 classifier techniques including XGBoost for disease prediction. All classification experiments were performed using 5-fold cross-validation. The diagnostic performance of the XGBoost method was compared with logistic regression and a linear Support Vector Machine by calculating their areas under the curve for differentiating Alzheimer disease from mild cognitive impairment and mild cognitive impairment from healthy controls.
In a total of 4 datasets, 1099, 212, 711, and 705 eligible patients were included. Compared with the linear Support Vector Machine and logistic regression, XGBoost significantly improved the prediction of Alzheimer disease (< .001). In terms of differentiating Alzheimer disease from mild cognitive impairment, the 3 algorithms resulted in areas under the curve of 0.758-0.825. XGBoost had a sensitivity of 68% and a specificity of 70%. In terms of differentiating mild cognitive impairment from the healthy control group, the 3 algorithms resulted in areas under the curve of 0.668-0.870. XGBoost had a sensitivity of 79% and a specificity of 80%.
The deep learning-based automatic brain segmentation and classification algorithm allowed an accurate diagnosis of Alzheimer disease using T1-weighted brain MR images. The widespread availability of T1-weighted brain MR imaging suggests that this algorithm is a promising and widely applicable method for predicting Alzheimer disease.
有有限的证据表明，基于 T1 加权脑磁共振成像的深度学习自动脑分割和分类方法可预测阿尔茨海默病。我们的目的是开发和验证一种基于深度学习的自动脑分割和分类算法，用于使用 3D T1 加权脑磁共振成像诊断阿尔茨海默病。
我们使用连续阿尔茨海默病和轻度认知障碍患者的 T1 加权脑磁共振成像数据集开发了一种基于深度学习的算法。我们使用卷积神经网络开发了一个 2 步算法，用于进行脑分割，然后使用 3 种分类器技术（包括 XGBoost）进行疾病预测。所有分类实验均采用 5 折交叉验证进行。通过计算区分阿尔茨海默病与轻度认知障碍、轻度认知障碍与健康对照的曲线下面积，比较 XGBoost 方法与逻辑回归和线性支持向量机的诊断性能。
在总共 4 个数据集中，纳入了 1099、212、711 和 705 例符合条件的患者。与线性支持向量机和逻辑回归相比，XGBoost 显著提高了阿尔茨海默病的预测能力（<.001）。在区分阿尔茨海默病与轻度认知障碍方面，这 3 种算法的曲线下面积为 0.758-0.825。XGBoost 的灵敏度为 68%，特异性为 70%。在区分轻度认知障碍与健康对照组方面，这 3 种算法的曲线下面积为 0.668-0.870。XGBoost 的灵敏度为 79%，特异性为 80%。
基于深度学习的自动脑分割和分类算法可使用 T1 加权脑磁共振成像准确诊断阿尔茨海默病。T1 加权脑磁共振成像广泛应用，提示该算法是一种有前途且广泛适用的预测阿尔茨海默病的方法。
