文献速递儿童青少年精神

预测儿童青少年注意缺陷多动障碍发病:全国性深度学习研究

Predicting childhood and adolescent attention-deficit/hyperactivity disorder onset: a nationwide deep learning approach.
PubMed 文献 · Mol Psychiatry · 2022-12-19

研究基于瑞典1995至1999年出生的238,696人的登记数据,比较多种机器学习方法预测儿童青少年注意缺陷多动障碍(ADHD)诊断的能力。深度神经网络(DNN)表现最佳,受试者工作特征曲线下面积为0.75(95%CI 0.74-0.76),平衡准确率为0.69。模型若经外部验证且具成本效益,或可提示临床医生对高危个体进行筛查。

要点

  • DNN预测儿童青少年ADHD的AUC为0.75,95%CI 0.74-0.76,平衡准确率0.69。
  • 在0.45概率阈值下,敏感度为71.66%,特异度为65.0%。
  • 贡献最大的前5个特征:父母有犯罪定罪、男性、亲属患ADHD、不及格学科数、言语或学习障碍。

核心结论

  • 仅使用瑞典登记数据训练的DNN模型预测儿童青少年ADHD取得了良好的判别能力。
  • 模型若在外部样本中重复验证并证实成本效益,可用于提示临床医生对应当筛查的个体进行筛查,辅助决策以减少误诊。

研究设计与样本

  • 分析瑞典1995至1999年出生并居住的238,696人的登记数据。
  • 使用逻辑回归、随机森林、梯度提升、XGBoost、惩罚逻辑回归、深度神经网络(DNN)及集成模型等多种机器学习技术评估登记数据辅助ADHD诊断的能力。

主要发现

  • 表现最佳的模型为DNN,受试者工作特征曲线下面积为0.75,95%CI(0.74-0.76),平衡准确率为0.69。
  • 在0.45概率阈值下,敏感度为71.66%,特异度为65.0%。
  • 所有模型在特征重要性上总体一致(τ > .5)。
  • 对分类贡献最大的前5个特征为:父母有犯罪定罪、男性、亲属患ADHD、不及格学科数、言语或学习障碍。

关键数值与推荐

  • DNN的AUC为0.75,95%CI(0.74-0.76),平衡准确率0.69。
  • 0.45概率阈值下敏感度71.66%、特异度65.0%。
  • 特征重要性总体一致性τ > .5。

临床怎么用

  • 模型若经外部验证并证实成本效益,可提示临床医生对应当筛查ADHD的个体进行筛查,并辅助临床决策以减少误诊。
  • 当前结果仍需在不同人群中验证并纳入新的预测因子。

局限性

  • 结果需在外部样本中重复验证,并在不同人群中进一步验证。
  • 需纳入新的预测因子并评估成本效益。

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)。

原文

Attention-deficit/hyperactivity disorder (ADHD) is a heterogeneous disorder with a high degree of psychiatric and physical comorbidity, which complicates its diagnosis in childhood and adolescence. We analyzed registry data from 238,696 persons born and living in Sweden between 1995 and 1999. Several machine learning techniques were used to assess the ability of registry data to inform the diagnosis of ADHD in childhood and adolescence: logistic regression, random Forest, gradient boosting, XGBoost, penalized logistic regression, deep neural network (DNN), and ensemble models. The best fitting model was the DNN, achieving an area under the receiver operating characteristic curve of 0.75, 95% CI (0.74-0.76) and balanced accuracy of 0.69. At the 0.45 probability threshold, sensitivity was 71.66% and specificity was 65.0%. There was an overall agreement in the feature importance among all models (τ > .5). The top 5 features contributing to classification were having a parent with criminal convictions, male sex, having a relative with ADHD, number of academic subjects failed, and speech/learning disabilities. A DNN model predicting childhood and adolescent ADHD trained exclusively on Swedish register data achieved good discrimination. If replicated and validated in an external sample, and proven to be cost-effective, this model could be used to alert clinicians to individuals who ought to be screened for ADHD and to aid clinicians' decision-making with the goal of decreasing misdiagnoses. Further research is needed to validate results in different populations and to incorporate new predictors.

来源:Mol Psychiatry,查看原文。版权归原作者所有。

引用

  1. Garcia-Argibay M, Zhang-James Y, Cortese S, et al. Predicting childhood and adolescent attention-deficit/hyperactivity disorder onset: a nationwide deep learning approach. Mol Psychiatry. 2022. PMID: 36536075. doi:10.1038/s41380-022-01918-8

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。