Adversarial confound regression and uncertainty measurements to classify heterogeneous clinical MRI in Mass General Brigham

Abstract

In this work, we introduce a novel deep learning architecture, MUCRAN (Multi-Confound Regression Adversarial Network), to train a deep learning model on clinical brain MRI while regressing demographic and technical confounding factors. We trained MUCRAN using 17,076 clinical T1 Axial brain MRIs collected from Massachusetts General Hospital before 2019 and demonstrated that MUCRAN could successfully regress major confounding factors in the vast clinical dataset. We also applied a method for quantifying uncertainty across an ensemble of these models to automatically exclude out-of-distribution data in AD detection. By combining MUCRAN and the uncertainty quantification method, we showed consistent and significant increases in the AD detection accuracy for newly collected MGH data (post-2019; 84.6% with MUCRAN vs. 72.5% without MUCRAN) and for data from other hospitals (90.3% from Brigham and Women’s Hospital and 81.0% from other hospitals). MUCRAN offers a generalizable approach for deep-learning-based disease detection in heterogenous clinical data.

A confounder controlled machine learning approach: Group analysis and classification of schizophrenia and Alzheimer’s disease using resting-state functional network connectivity

“…This might be because of other confounder variables, such as the race of subjects that are not collected for PLOS ONE our datasets. Furthermore, as an interesting future direction, one can apply more advanced approaches (e.g., generative adversarial networks (GANs) as suggested by [30,31]) to control confounders simultaneously to predict the target values (i.e., diagnostic). Finally, the scope of future work on this topic can extend to using other brain features, such as dynamic FNC, ICAestimated time courses, and even voxel-level MRI images to identify similarities and differences between AD and SZ.…”