文献速递神经儿科影像MR

基于全脑MR影像组学构建儿童孤独症谱系障碍预测模型:一项机器学习研究

Constructing a Predictive Model for Children with Autism Spectrum Disorder Based on Whole-Brain MR Radiomics: A Machine Learning Study.
PubMed 文献 · AJNR Am J Neuroradiol · 2026-01-05

本研究利用孤独症脑影像数据交换数据库223例(120例孤独症谱系障碍)及乔治城大学与迈阿密大学独立外部测试集87例(38例孤独症谱系障碍),提取全脑MR结构像白质、灰质与脑脊液影像组学特征构建影像组学标签,结合言语智商构建机器学习模型。决策树联合模型表现最佳,训练集、测试集、外部测试集AUC分别为0.87、0.84、0.83,提示加入言语智商数据与决策树算法可提升孤独症谱系障碍识别效能。

要点

  • 全脑MR结构像影像组学标签可识别孤独症谱系障碍,外部测试集AUC为0.74(95%CI 0.64-0.83)。
  • 言语智商是多因素logistic回归筛出的孤独症谱系障碍预测因子。
  • 决策树联合模型最佳,外部测试集敏感度0.86、特异度0.66。

核心结论

  • 基于全脑MRI影像特征可构建识别孤独症谱系障碍的影像组学标签,加入言语智商数据与决策树算法模型可提升识别效能,可为临床实践提供适应性策略。

研究设计与样本

  • 研究纳入孤独症脑影像数据交换数据库223例受试者(其中120例确诊孤独症谱系障碍),按7:3随机分为训练集与测试集。
  • 独立外部测试集来自乔治城大学与迈阿密大学孤独症谱系障碍数据集,共87例受试者(其中38例孤独症谱系障碍)。
  • 从每例受试者全脑MR结构图像提取白质、灰质与脑脊液定量影像组学特征,基于训练集数据降维构建影像组学标签。
  • 采用多因素logistic回归从临床特征中筛选孤独症谱系障碍独立预测因子,联合影像组学标签构建多种机器学习模型。
  • 采用受试者工作特征曲线评估训练集、测试集与外部测试集数据,并以模型截断值0.4285将数据集分为低风险与高风险亚组比较孤独症谱系障碍实际人数。

诊断标准与关键数值

  • 影像组学标签在训练集、测试集、外部测试集的AUC分别为0.78(95%CI 0.71-0.85)、0.75(95%CI 0.67-0.83)、0.74(95%CI 0.64-0.83)。
  • 多因素logistic回归显示言语智商为孤独症谱系障碍的预测因子。
  • 决策树算法联合影像组学标签构建的联合模型表现最佳,训练集、测试集、外部测试集AUC分别为0.87(95%CI 0.81-0.92)、0.84(95%CI 0.76-0.91)、0.83(95%CI 0.74-0.91)。
  • 联合模型训练集、测试集、外部测试集敏感度分别为0.89(95%CI 0.82-0.95)、0.84(95%CI 0.73-0.93)、0.86(95%CI 0.72-0.96),特异度分别为0.70(95%CI 0.60-0.79)、0.63(95%CI 0.52-0.74)、0.66(95%CI 0.49-0.80)。
  • 按模型截断值0.4285分出的低风险与高风险亚组,孤独症谱系障碍实际人数在训练集(χ2=21.325;P<.05)、测试集(χ2=5.379;P<.05)、外部测试集(χ2=21.52;P<.05)差异均有统计学意义。

临床应用与阅片要点

  • 原文结论认为影像组学标签可基于全脑MRI影像特征识别孤独症谱系障碍,联合言语智商数据与决策树算法模型可提升识别效能,可为临床实践提供适应性策略。

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

BACKGROUND AND PURPOSE: Current autism spectrum disorder (ASD) diagnosis depends on subjective behavioral assessments causing delays and variability. Structural MRI shows brain abnormalities but traditional analysis lacks sensitivity. This study aims to construct a prediction model based on whole-brain imaging radiomics analysis by using machine learning for identifying children with ASD.

MATERIALS AND METHODS: This study used imaging and clinical data from 223 subjects in the Autism Brain Imaging Data Exchange database, including 120 patients diagnosed with ASD. These data were randomly divided into training and test sets in a 7:3 ratio. An independent external test set comprising 87 participants (38 with ASD) from the Georgetown University and University of Miami ASD data set was also utilized. Then, quantitative radiomics features of white matter, gray matter, and CSF were extracted from the whole-brain MR structural images of each subject, and feature dimensionality reduction was performed based on the training set data to construct radiomics signature. Afterward, multivariate logistic regression was used to screen independent predictors of ASD from clinical features and then, combined with radiomics signature, multiple machine learning models were constructed to predict ASD. Finally, the optimal model was selected, and the receiver operating characteristic curve was used to evaluate the training, test, and external test sets data. Simultaneously, the model divides the data set into low-risk and high-risk subgroups, comparing the actual number of individuals with ASD between the 2 subgroups to evaluate the clinical efficacy of the model.

RESULTS: The areas under the curve (AUCs) of radiomics markers in the training set, test set, and external test set were 0.78 (95% CI, 0.71-0.85), 0.75 (95% CI, 0.67-0.83), and 0.74 (95% CI, 0.64-0.83), respectively. Multivariate logistic regression showed that the Verbal Intelligence Quotient (VIQ) was a predictor of ASD. The joint model constructed by the decision tree algorithm with radiomics markers performed best, with AUC of 0.87 (95% CI, 0.81-0.92), 0.84 (95% CI, 0.76-0.91), and 0.83 (95% CI, 0.74-0.91), a sensitivity of 0.89 (95% CI, 0.82-0.95), 0.84 (95% CI, 0.73-0.93), and 0.86 (95% CI, 0.72-0.96), and the specificity of 0.70 (95% CI, 0.60-0.79), 0.63 (95% CI, 0.52-0.74), and 0.66 (95% CI, 0.49-0.80), respectively. The low-risk subgroup and high-risk subgroup classified according to the cutoff value of 0.4285 of the model showed statistically significant differences in the actual number of patients with ASD in the training set (χ2 = 21.325; P < .05), the test set (χ2 = 5.379; P < .05), and the external test set (χ2 = 21.52; P < .05).

CONCLUSIONS: Radiomics signature for identifying ASD can be constructed based on whole-brain MRI imaging features. The performance of identifying ASD can be improved by adding VIQ data and the decision tree algorithm model, which can provide an adaptive strategy for clinical practice.

来源:AJNR Am J Neuroradiol,查看原文。版权归原作者所有。

引用

  1. Chen X, Peng J, Zhang Z, et al. Constructing a Predictive Model for Children with Autism Spectrum Disorder Based on Whole-Brain MR Radiomics: A Machine Learning Study. AJNR Am J Neuroradiol. 2026. PMID: 40721282. doi:10.3174/ajnr.A8939

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。