Lin Andong, Chen Yini, Chen Yi, Ye Zhinan, Luo Weili, Chen Ying, Zhang Yaping, Wang Wenjie
Department of Neurology, Municipal Hospital Affiliated to Taizhou University, Taizhou, China.
Department of Radiology, The First Affiliated Hospital of Dalian Medical University, Dalian, Liaoning Province, China.
Front Aging Neurosci. 2024 Oct 4;16:1460293. doi: 10.3389/fnagi.2024.1460293. eCollection 2024.
Mild Cognitive Impairment (MCI) is a recognized precursor to Alzheimer's Disease (AD), presenting a significant risk of progression. Early detection and intervention in MCI can potentially slow disease advancement, offering substantial clinical benefits. This study employed radiomics and machine learning methodologies to distinguish between MCI and Normal Cognition (NC) groups.
The study included 172 MCI patients and 183 healthy controls from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database, all of whom had 3D-T1 weighted MRI structural images. The cerebellar gray and white matter were segmented automatically using volBrain software, and radiomic features were extracted and screened through Pyradiomics. The screened features were then input into various machine learning models, including Random Forest (RF), Logistic Regression (LR), eXtreme Gradient Boosting (XGBoost), Support Vector Machines (SVM), K Nearest Neighbors (KNN), Extra Trees, Light Gradient Boosting Machine (LightGBM), and Multilayer Perceptron (MLP). Each model was optimized for penalty parameters through 5-fold cross-validation to construct radiomic models. The DeLong test was used to evaluate the performance of different models.
The LightGBM model, which utilizes a combination of cerebellar gray and white matter features (comprising eight gray matter and eight white matter features), emerges as the most effective model for radiomics feature analysis. The model demonstrates an Area Under the Curve (AUC) of 0.863 for the training set and 0.776 for the test set.
Radiomic features based on the cerebellar gray and white matter, combined with machine learning, can objectively diagnose MCI, which provides significant clinical value for assisted diagnosis.
轻度认知障碍（MCI）是公认的阿尔茨海默病（AD）前驱症状，具有显著的进展风险。对MCI进行早期检测和干预可能会减缓疾病进展，带来重大临床益处。本研究采用放射组学和机器学习方法区分MCI组和正常认知（NC）组。
该研究纳入了来自阿尔茨海默病神经影像倡议（ADNI）数据库的172例MCI患者和183例健康对照者，所有参与者均有3D-T1加权MRI结构图像。使用volBrain软件自动分割小脑灰质和白质，并通过Pyradiomics提取和筛选放射组学特征。然后将筛选后的特征输入到各种机器学习模型中，包括随机森林（RF）、逻辑回归（LR）、极端梯度提升（XGBoost）、支持向量机（SVM）、K近邻（KNN）、极端随机树、轻量级梯度提升机（LightGBM）和多层感知器（MLP）。通过5折交叉验证对每个模型的惩罚参数进行优化，以构建放射组学模型。使用德龙检验评估不同模型的性能。
利用小脑灰质和白质特征组合（包括八个灰质特征和八个白质特征）的LightGBM模型，成为放射组学特征分析中最有效的模型。该模型在训练集上的曲线下面积（AUC）为0.863，在测试集上为0.776。
基于小脑灰质和白质的放射组学特征，结合机器学习，能够客观诊断MCI，为辅助诊断提供了重要的临床价值。
