Opaque Ontology: Neuroimaging Classification of ICD-10 Diagnostic Groups in the UK Biobank.
Opaque Ontology: Neuroimaging Classification of ICD-10 Diagnostic Groups in the UK Biobank.
复制标题
不透明本体论：英国生物银行 ICD-10 诊断组的神经影像分类。
DOI：
10.1101/2024.04.15.589555
复制
发表时间：
2024
期刊：
影响因子：
--
通讯作者：
Bijsterbosch,Janine
中科院分区：
文献类型：
--
作者：
Easley,Ty;Luo,Xiaoke;Hannon,Kayla;Lenzini,Petra;Bijsterbosch,Janine
背景使用机器学习来对诊断病例与基于诊断本体（诸如国际疾病分类第十修订版（ICD-10））从神经成像特征定义的对照进行分类现在在广泛的诊断领域中是常见的。然而，这种分类的跨诊断比较缺乏。这样的transdiagnosis比较是很重要的，以建立分类模型的特异性，设置基准，并评估诊断ontologs.ResultsWe的价值调查病例对照分类准确性在17个不同的ICD-10诊断组从第五章（精神和行为障碍）和第六章（神经系统疾病）使用的数据从英国生物银行。使用神经成像（结构或功能性脑磁共振成像特征集）或社会人口特征训练分类模型。随机森林分类模型采用严格的洗牌分裂，估计稳定性以及病例对照分类的准确性。诊断分类准确性以来自相同特征集的年龄分类（最老与最年轻）和其他分类器类型（k-最近邻和线性支持向量机）为基准。与所有特征集的年龄分类准确率高相反，很少有ICD-10诊断组被分类为显著高于机会（即，基于结构神经影像学特征的脱髓鞘疾病和基于社会人口统计学和功能神经影像学特征的抑郁症）。因此，我们建议谨慎使用ICD-10诊断组作为基于脑的疾病预测研究的目标标签。
BackgroundThe use of machine learning to classify diagnostic cases versus controls defined based on diagnostic ontologies such as the International Classification of Diseases, Tenth Revision (ICD-10) from neuroimaging features is now commonplace across a wide range of diagnostic fields. However, transdiagnostic comparisons of such classifications are lacking. Such transdiagnostic comparisons are important to establish the specificity of classification models, set benchmarks, and assess the value of diagnostic ontologies.ResultsWe investigated case-control classification accuracy in 17 different ICD-10 diagnostic groups from Chapter V (mental and behavioral disorders) and Chapter VI (diseases of the nervous system) using data from the UK Biobank. Classification models were trained using either neuroimaging (structural or functional brain magnetic resonance imaging feature sets) or sociodemographic features. Random forest classification models were adopted using rigorous shuffle-splits to estimate stability as well as accuracy of case-control classifications. Diagnostic classification accuracies were benchmarked against age classification (oldest vs. youngest) from the same feature sets and against additional classifier types (k-nearest neighbors and linear support vector machine). In contrast to age classification accuracy, which was high for all feature sets, few ICD-10 diagnostic groups were classified significantly above chance (namely, demyelinating diseases based on structural neuroimaging features and depression based on sociodemographic and functional neuroimaging features).ConclusionThese findings highlight challenges with the current disease classification system, leading us to recommend caution with the use of ICD-10 diagnostic groups as target labels in brain-based disease prediction studies.