Deep attentive spatio-temporal feature learning for automatic resting-state fMRI denoising

• Published in: NeuroImage (Orlando, Fla.) Vol. 254; p. 119127
• Main Authors: Heo, Keun-Soo, Shin, Dong-Hee, Hung, Sheng-Che, Lin, Weili, Zhang, Han, Shen, Dinggang, Kam, Tae-Eui
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.07.2022; Elsevier Limited; Elsevier
• Subjects: Blood pressure; Brain mapping; Convolutional neural network; Deep learning; Denoising; Functional magnetic resonance imaging; Independent component analysis; Infants; Machine learning; Neuroimaging; Noise; Physiology; Resting-state fMRI; Temporal variations; Time series; Deep learning; Resting-state fMRI; Convolutional neural network; Independent component analysis; Denoising
• ISSN: 1053-8119; 1095-9572; 1095-9572
• DOI: 10.1016/j.neuroimage.2022.119127

•We propose a unified deep attentive CNN framework for automatic rs-fMRI denoising.•It simultaneously learns spatio-temporal features of noise in a data-driven manner.•We provide visual explanations to depict how the CNNs work for noise detection.•It achieves high performance on various datasets including infant cohorts.•It can be integrated into any pipelines by accelerating speed (<1s per component). Resting-state functional magnetic resonance imaging (rs-fMRI) is a non-invasive functional neuroimaging modality that has been widely used to investigate functional connectomes in the brain. Since noise and artifacts generated by non-neuronal physiological activities are predominant in raw rs-fMRI data, effective noise removal is one of the most important preprocessing steps prior to any subsequent analysis. For rs-fMRI denoising, a common trend is to decompose rs-fMRI data into multiple components and then regress out noise-related components. Therefore, various machine learning techniques have been used in such analyses with predefined procedures and manually engineered features. However, the lack of a universal definition of a noise-related source or artifact complicates manual feature engineering. Manual feature selection can result in the failure to capture unknown types of noise. Furthermore, the possibility that the hand-crafted features will only work for the broader population (e.g., healthy adults) but not for “outliers” (e.g., infants or subjects that belong to a disease cohort) is quite high. In practice, we have limited knowledge of which features should be extracted; thus, multi-classifier assembly must be implemented to improve performance, although this process is quite time-consuming. However, in real rs-fMRI applications, fast and accurate automatic identification of noise-related components on different datasets is critical. To solve this problem, we propose a novel, automatic, and end-to-end deep learning framework dedicated to noise-related component identification via a faster and more effective multi-layer feature extraction strategy that learns deeply embedded spatio-temporal features of the components. In this study, we achieved remarkable performance on various rs-fMRI datasets, including multiple adult rs-fMRI datasets from different rs-fMRI studies and an infant rs-fMRI dataset, which is quite heterogeneous and differs from that of adults. Our proposed framework also dramatically increases the noise detection speed owing to its inherent ability for deep learning (< 1s for single-component classification). It can be easily integrated into any preprocessing pipeline, even those that do not use standard procedures but depend on alternative toolboxes.