DeepEZ: A Graph Convolutional Network for Automated Epileptogenic Zone Localization From Resting-State fMRI Connectivity.

OBJECTIVE: Epileptogenic zone (EZ) localization is a crucial step during diagnostic work up and therapeutic planning in medication refractory epilepsy. In this paper, we present the first deep learning approach to localize the EZ based on resting-state fMRI (rs-fMRI) data. METHODS: Our network, called DeepEZ, uses a cascade of graph convolutions that emphasize signal propagation along expected anatomical pathways. We also integrate domain-specific information, such as an asymmetry term on the predicted EZ and a learned subject-specific bias to mitigate environmental confounds. RESULTS: We validate DeepEZ on rs-fMRI collected from 14 patients with focal epilepsy at the University of Wisconsin Madison. Using cross validation, we demonstrate that DeepEZ achieves consistently high EZ localization performance (Accuracy: 0.88 ± 0.03; AUC: 0.73 ± 0.03) that far outstripped any of the baseline methods. This performance is notable given the variability in EZ locations and scanner type across the cohort. CONCLUSION: Our results highlight the promise of using DeepEZ as an accurate and noninvasive therapeutic planning tool for medication refractory epilepsy. SIGNIFICANCE: While prior work in EZ localization focused on identifying localized aberrant signatures, there is growing evidence that epileptic seizures affect inter-regional connectivity in the brain. DeepEZ allows clinicians to harness this information from noninvasive imaging that can easily be integrated into the existing clinical workflow.

Automated eloquent cortex localization in brain tumor patients using multi-task graph neural networks.

Localizing the eloquent cortex is a crucial part of presurgical planning. While invasive mapping is the gold standard, there is increasing interest in using noninvasive fMRI to shorten and improve the process. However, many surgical patients cannot adequately perform task-based fMRI protocols. Resting-state fMRI has emerged as an alternative modality, but automated eloquent cortex localization remains an open challenge. In this paper, we develop a novel deep learning architecture to simultaneously identify language and primary motor cortex from rs-fMRI connectivity. Our approach uses the representational power of convolutional neural networks alongside the generalization power of multi-task learning to find a shared representation between the eloquent subnetworks. We validate our method on data from the publicly available Human Connectome Project and on a brain tumor dataset acquired at the Johns Hopkins Hospital. We compare our method against feature-based machine learning approaches and a fully-connected deep learning model that does not account for the shared network organization of the data. Our model achieves significantly better performance than competing baselines. We also assess the generalizability and robustness of our method. Our results clearly demonstrate the advantages of our graph convolution architecture combined with multi-task learning and highlight the promise of using rs-fMRI as a presurgical mapping tool.

A Multi-Scale Spatial and Temporal Attention Network on Dynamic Connectivity to Localize The Eloquent Cortex in Brain Tumor Patients.

We present a deep neural network architecture that combines multi-scale spatial attention with temporal attention to simultaneously localize the language and motor areas of the eloquent cortex from dynamic functional connectivity data. Our multi-scale spatial attention operates on graph-based features extracted from the connectivity matrices, thus honing in on the inter-regional interactions that collectively define the eloquent cortex. At the same time, our temporal attention model selects the intervals during which these interactions are most pronounced. The final stage of our model employs multi-task learning to differentiate between the eloquent subsystems. Our training strategy enables us to handle missing eloquent class labels by freezing the weights in those branches while updating the rest of the network weights. We evaluate our method on resting-state fMRI data from one synthetic dataset and one in-house brain tumor dataset while using task fMRI activations as ground-truth labels for the eloquent cortex. Our model achieves higher localization accuracies than conventional deep learning approaches. It also produces interpretable spatial and temporal attention features which can provide further insights for presurgical planning. Thus, our model shows translational promise for improving the safety of brain tumor resections.