Early diagnosis remains a significant challenge for many neurological disorders, especially for rare disorders where studying large cohorts is not possible. A novel solution that investigators have undertaken is combining advanced machine learning algorithms with resting-state functional Magnetic Resonance Imaging to unveil hidden pathological brain connectome patterns to uncover diagnostic and prognostic biomarkers. Recently, state-of-the-art deep learning techniques are outperforming traditional machine learning methods and are hailed as a milestone for artificial intelligence. However, whole brain classification that combines brain connectome with deep learning has been hindered by insufficient training samples. Inspired by the transfer learning strategy employed in computer vision, we exploited previously collected resting-state functional MRI data for healthy subjects from existing databases and transferred this knowledge for new disease classification tasks. We developed a deep transfer learning neural network (DTL-NN) framework for enhancing the classification of whole brain functional connectivity patterns. Briefly, we trained a stacked sparse autoencoder (SSAE) prototype to learn healthy functional connectivity patterns in an offline learning environment. Then, the SSAE prototype was transferred to a DTL-NN model for a new classification task. To test the validity of our framework, we collected resting-state functional MRI data from the Autism Brain Imaging Data Exchange (ABIDE) repository. Using autism spectrum disorder (ASD) classification as a target task, we compared the performance of our DTL-NN approach with a traditional deep neural network and support vector machine models across four ABIDE data sites that enrolled at least 60 subjects. As compared to traditional models, our DTL-NN approach achieved an improved performance in accuracy, sensitivity, specificity and area under receiver operating characteristic curve. These findings suggest that DTL-NN approaches could enhance disease classification for neurological conditions, where accumulating large neuroimaging datasets has been challenging.

Autism spectrum disorder (ASD) is a developmental disorder, affecting about 1% of the global population. Currently, the only clinical method for diagnosing ASD are standardized ASD tests which require prolonged diagnostic time and increased medical costs. Our objective was to explore the predictive power of personal characteristic data (PCD) from a large well-characterized dataset to improve upon prior diagnostic models of ASD. We extracted six personal characteristics (age, sex, handedness, and three individual measures of IQ) from 851 subjects in the Autism Brain Imaging Data Exchange (ABIDE) database. ABIDE is an international collaborative project that collected data from a large number of ASD patients and typical non-ASD controls from 17 research and clinical institutes. We employed this publicly available database to test nine supervised machine learning models. We implemented a cross-validation strategy to train and test those machine learning models for classification between typical non-ASD controls and ASD patients. We assessed classification performance using accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). Of the nine models we tested using six personal characteristics, the neural network model performed the best with a mean AUC (SD) of 0.646 (0.005), followed by k-nearest neighbor with a mean AUC (SD) of 0.641 (0.004). This study established an optimal ASD classification performance with PCD as features. With additional discriminative features (e.g., neuroimaging), machine learning models may ultimately enable automated clinical diagnosis of autism.

Investigation of the brain's functional connectome can improve our understanding of how an individual brain's organizational changes influence cognitive function and could result in improved individual risk stratification. Brain connectome studies in adults and older children have shown that abnormal network properties may be useful as discriminative features and have exploited machine learning models for early diagnosis in a variety of neurological conditions. However, analogous studies in neonates are rare and with limited significant findings. In this paper, we propose an artificial neural network (ANN) framework for early prediction of cognitive deficits in very preterm infants based on functional connectome data from resting state fMRI. Specifically, we conducted feature selection via stacked sparse autoencoder and outcome prediction via support vector machine (SVM). The proposed ANN model was unsupervised learned using brain connectome data from 884 subjects in autism brain imaging data exchange database and SVM was cross-validated on 28 very preterm infants (born at 23-31 weeks of gestation and without brain injury; scanned at term-equivalent postmenstrual age). Using 90 regions of interests, we found that the ANN model applied to functional connectome data from very premature infants can predict cognitive outcome at 2 years of corrected age with an accuracy of 70.6% and area under receiver operating characteristic curve of 0.76. We also noted that several frontal lobe and somatosensory regions, significantly contributed to prediction of cognitive deficits 2 years later. Our work can be considered as a proof of concept for utilizing ANN models on functional connectome data to capture the individual variability inherent in the developing brains of preterm infants. The full potential of ANN will be realized and more robust conclusions drawn when applied to much larger neuroimaging datasets, as we plan to do.

The hypergraph structure has been utilized to characterize the brain functional connectome (FC) by capturing the high order relationships among multiple brain regions of interest (ROIs) compared with a simple graph. Accordingly, hypergraph neural network (HGNN) models have emerged and provided efficient tools for hypergraph embedding learning. However, most existing HGNN models can only be applied to pre-constructed hypergraphs with a static structure during model training, which might not be a sufficient representation of the complex brain networks. In this study, we propose a dynamic weighted hypergraph convolutional network (dwHGCN) framework to consider a dynamic hypergraph with learnable hyperedge weights. Specifically, we generate hyperedges based on sparse representation and calculate the hyper similarity as node features. The hypergraph and node features are fed into a neural network model, where the hyperedge weights are updated adaptively during training. The dwHGCN facilitates the learning of brain FC features by assigning larger weights to hyperedges with higher discriminative power. The weighting strategy also improves the interpretability of the model by identifying the highly active interactions among ROIs shared by a common hyperedge. We validate the performance of the proposed model on two classification tasks with three paradigms functional magnetic resonance imaging (fMRI) data from Philadelphia Neurodevelopmental Cohort. Experimental results demonstrate the superiority of our proposed method over existing hypergraph neural networks. We believe our model can be applied to other applications in neuroimaging for its strength in representation learning and interpretation.

Survivors following very premature birth (i.e., ≤ 32 weeks gestational age) remain at high risk for neurodevelopmental impairments. Recent advances in deep learning techniques have made it possible to aid the early diagnosis and prognosis of neurodevelopmental deficits. Deep learning models typically require training on large datasets, and unfortunately, large neuroimaging datasets with clinical outcome annotations are typically limited, especially in neonates. Transfer learning represents an important step to solve the fundamental problem of insufficient training data in deep learning. In this work, we developed a multi-task, multi-stage deep transfer learning framework using the fusion of brain connectome and clinical data for early joint prediction of multiple abnormal neurodevelopmental (cognitive, language and motor) outcomes at 2 years corrected age in very preterm infants. The proposed framework maximizes the value of both available annotated and non-annotated data in model training by performing both supervised and unsupervised learning. We first pre-trained a deep neural network prototype in a supervised fashion using 884 older children and adult subjects, and then re-trained this prototype using 291 neonatal subjects without supervision. Finally, we fine-tuned and validated the pre-trained model using 33 preterm infants. Our proposed model identified very preterm infants at high-risk for cognitive, language, and motor deficits at 2 years corrected age with an area under the receiver operating characteristic curve of 0.86, 0.66 and 0.84, respectively. Employing such a deep learning model, once externally validated, may facilitate risk stratification at term-equivalent age for early identification of long-term neurodevelopmental deficits and targeted early interventions to improve clinical outcomes in very preterm infants.

Advances in artificial intelligence (AI), especially deep learning (DL), have facilitated magnetic resonance imaging (MRI) data analysis, enabling AI-assisted medical image diagnoses and prognoses. However, most of the DL models are considered as “black boxes”. There is an unmet need to demystify DL models so domain experts can trust these high-performance DL models. This has resulted in a sub-domain of AI research called explainable artificial intelligence (XAI). In the last decade, many experts have dedicated their efforts to developing novel XAI methods that are competent at visualizing and explaining the logic behind data-driven DL models. However, XAI techniques are still in their infancy for medical MRI image analysis. This study aims to outline the XAI applications that are able to interpret DL models for MRI data analysis. We first introduce several common MRI data modalities. Then, a brief history of DL models is discussed. Next, we highlight XAI frameworks and elaborate on the principles of multiple popular XAI methods. Moreover, studies on XAI applications in MRI image analysis are reviewed across the tissues/organs of the human body. A quantitative analysis is conducted to reveal the insights of MRI researchers on these XAI techniques. Finally, evaluations of XAI methods are discussed. This survey presents recent advances in the XAI domain for explaining the DL models that have been utilized in MRI applications.