Functional Connectivity-Based Prediction of Autism on Site Harmonized ABIDE Dataset.
Functional Connectivity-Based Prediction of Autism on Site Harmonized ABIDE Dataset.
复制标题
DOI：
10.1109/tbme.2021.3080259
复制
发表时间：
2021-12
期刊：
影响因子：
--
通讯作者：
Deshpande G
中科院分区：
文献类型：
--
作者：
Ingalhalikar M;Shinde S;Karmarkar A;Rajan A;Rangaprakash D;Deshpande G
从多站点公开可用的神经成像数据存储库中获得的较大样本量通过减轻维度灾难使基于机器学习的精神障碍诊断分类更加可行。然而，由于多部位数据是事后聚合的，即，它们是从具有不同采集参数的不同扫描仪采集的，因此非神经部位间变异性可能掩盖至少部分起源于神经的组间差异。因此，在基于机器学习的诊断分类的背景下通过较大样本量获得的优势可能无法实现。我们使用ComBat技术协调多站点神经成像数据来解决这个问题，ComBat技术基于经验贝叶斯公式来消除数据分布中的站点间差异，以提高诊断分类准确性。具体来说，我们证明了这一点，使用ABIDE（自闭症脑成像数据交换）多站点数据进行分类的个人与自闭症健康对照组使用静息状态fMRI为基础的功能连接数据。我们的研究结果表明，更高的分类精度跨多个分类模型可以获得（特别是基于人工神经网络的模型）从多站点数据后协调与ComBat技术相比，没有协调，优于早期的结果，从现有的研究使用ABIDE。此外，我们的网络消融分析促进了对自闭症谱系障碍病理学的重要见解，并且网络中的连通性对于与自闭症中的言语沟通障碍共变的分类很重要。使用ComBat的多站点数据协调改进了基于神经成像的精神障碍诊断分类。ComBat有可能使基于AI的临床决策支持系统在精神病学中更加可行。
The larger sample sizes available from multi-site publicly available neuroimaging data repositories makes machine-learning based diagnostic classification of mental disorders more feasible by alleviating the curse of dimensionality. However, since multi-site data are aggregated post-hoc, i.e. they were acquired from different scanners with different acquisition parameters, non-neural inter-site variability may mask inter-group differences that are at least in part neural in origin. Hence, the advantages gained by the larger sample size in the context of machine-learning based diagnostic classification may not be realized. We address this issue using harmonization of multi-site neuroimaging data using the ComBat technique, which is based on an empirical Bayes formulation to remove inter-site differences in data distributions, to improve diagnostic classification accuracy. Specifically, we demonstrate this using ABIDE (Autism Brain Imaging Data Exchange) multisite data for classifying individuals with Autism from healthy controls using resting state fMRI-based functional connectivity data. Our results show that higher classification accuracies across multiple classification models can be obtained (especially for models based on artificial neural networks) from multi-site data post harmonization with the ComBat technique as compared to without harmonization, outperforming earlier results from existing studies using ABIDE. Furthermore, our network ablation analysis facilitated important insights into autism spectrum disorder pathology and the connectivity in networks shown to be important for classification covaried with verbal communication impairments in Autism. Multi-site data harmonization using ComBat improves neuroimaging-based diagnostic classification of mental disorders. ComBat has the potential to make AI-based clinical decision-support systems more feasible in psychiatry.