Abstract:Brain state classification by applying deep learning techniques on neuroimaging data has become a recent topic of research. However, unlike domains where the data is low dimensional or there are large number of available training samples, neuroimaging data is high dimensional and has few training samples. To tackle these issues, we present a sparse feedforward deep neural architecture for encoding and decoding the structural connectome of the human brain. We use a sparsely connected element-wise multiplication as the first hidden layer and a fixed transform layer as the output layer. The number of trainable parameters and the training time is significantly reduced compared to feedforward networks. We demonstrate superior performance of this architecture in encoding the structural connectome implicated in Alzheimer's disease (AD) and Parkinson's disease (PD) from DTI brain scans. For decoding, we propose recursive feature elimination (RFE) algorithm based on DeepLIFT, layer-wise relevance propagation (LRP), and Integrated Gradients (IG) algorithms to remove irrelevant features and thereby identify key biomarkers associated with AD and PD. We show that the proposed architecture reduces 45.1% and 47.1% of the trainable parameters compared to a feedforward DNN with an increase in accuracy by 2.6 % and 3.1% for cognitively normal (CN) vs AD and CN vs PD classification, respectively. We also show that the proposed RFE method leads to a further increase in accuracy by 2.1% and 4% for CN vs AD and CN vs PD classification, while removing approximately 90% to 95% irrelevant features. Furthermore, we argue that the biomarkers (i.e., key brain regions and connections) identified are consistent with previous literature. We show that relevancy score-based methods can yield high discriminative power and are suitable for brain decoding. We also show that the proposed approach led to a reduction in the number of trainable network parameters, an increase in classification accuracy, and a detection of brain connections and regions that were consistent with earlier studies.

Semantic segmentation of microscopic neuroanatomical data by combining topological priors with encoder–decoder deep networks

Deep Learning With Anatomical Priors: Imitating Enhanced Autoencoders In Latent Space For Improved Pelvic Bone Segmentation In MRI

Large Scale Image Segmentation with Structured Loss Based Deep Learning for Connectome Reconstruction

Sparse Deep Neural Network for Encoding and Decoding the Structural Connectome

Bridging the Semantic Latent Space Between Brain and Machine: Similarity is All You Need

Deep residual contextual and subpixel convolution network for automated neuronal structure segmentation in micro-connectomics

Boosting Multilabel Semantic Segmentation for Somata and Vessels in Mouse Brain

A 3D semantic segmentation network for accurate neuronal soma segmentation

Deep Neural Network-Based Semantic Segmentation of Microvascular Decompression Images

Deep vessel segmentation with joint multi-prior encoding

Large-scale Exploration of Neuronal Morphologies Using Deep Learning and Augmented Reality.

DVNet: A Memory-Efficient Three-Dimensional CNN for Large-Scale Neurovascular Reconstruction

TransSea: Hybrid CNN–Transformer With Semantic Awareness for 3-D Brain Tumor Segmentation

TransSea: Hybrid CNN-Transformer with Semantic Awareness for 3D Brain Tumor Segmentation

Enhancing Neural Segmentation on Low-Quality SEM Images Using Self-Attention Based Fusion Module with Auxiliary Deblurring Network

Deep ensembles in bioimage segmentation

Neurovascular Segmentation in sOCT with Deep Learning and Synthetic Training Data

Automated deep learning segmentation of high-resolution 7 T postmortem MRI for quantitative analysis of structure-pathology correlations in neurodegenerative diseases

DeepMRSeg: A convolutional deep neural network for anatomy and abnormality segmentation on MR images

A deep learning-based strategy for producing dense 3D segmentations from sparsely annotated 2D images

Automated deep learning segmentation of high-resolution 7 Tesla postmortem MRI for quantitative analysis of structure-pathology correlations in neurodegenerative diseases