Skip to main content
eScholarship
Open Access Publications from the University of California

Scholarly Works (9 results)

  • Thesis
  • Peer Reviewed
UC Santa Barbara Electronic Theses and Dissertations (2019)

This research focuses on introducing novel machine learning algorithms for predicting the outcome of patients with brain disorders using MR images. We first introduce the challenges in medical image analysis and overview the magnetic resonance imaging, datasets, and existing tools. A brain lesion segmentation method is then presented. The main novelty of this method is a new feature fusion method that integrates location information and the state-of-the-art patch-based neural networks for lesion segmentation. The proposed feature fusion method improves the segmentation performance of the patch-based neural networks. Thereafter, we focus on predicting the overall survival of brain tumor patients. We introduce a novel feature called the tractographic feature to capture the potentially damaged regions due to the presence of the lesion. The tractographic feature is built from the lesion and average connectome information from a group of normal subjects. It takes into account different functional regions that are affected by the lesion, thus complementing the commonly used lesion volume features. The tractographic feature is tested on the Multi-modal Brain Tumor Segmentation (BraTS) 2018 dataset and achieves a better overall survival prediction performance than other features and the gold standard that uses patient age. The proposed tractographic feature is also used to predict the clinical outcome of stroke patients. On the publicly available stroke benchmark, Ischemic Stroke Lesion Segmentation (ISLES) 2017, our proposed tractographic feature achieves higher accuracy than the state-of-the-art feature descriptors. Finally, we focus on predicting the outcome of mild traumatic brain injury (TBI) patients. The tractographic feature cannot be built in this case since the mTBI patients do not have brain lesions in the traditional MR images or CT scans. Here, we present an unsupervised 3D feature clustering algorithm, consisting of 3D dictionary learning, convolutional network, k-means clustering algorithm with group constraint, to gather mild TBI patients into three groups using their structural and diffusion MR images. The proposed method won the 3rd place in Mild Traumatic Brain Injury Outcome Prediction (mTOP) 2016 challenge.

    Cover page: Integrating Brain Connectome and Lesion Data for Patient Outcome Prediction
    • Article
    • Peer Reviewed
    UC Irvine Previously Published Works (2019)
    We propose a novel, simple and effective method to integrate lesion prior and a 3D U-Net for improving brain tumor segmentation. First, we utilize the ground-truth brain tumor lesions from a group of patients to generate the heatmaps of different types of lesions. These heatmaps are used to create the volume-of-interest (VOI) map which contains prior information about brain tumor lesions. The VOI map is then integrated with the multimodal MR images and input to a 3D U-Net for segmentation. The proposed method is evaluated on a public benchmark dataset, and the experimental results show that the proposed feature fusion method achieves an improvement over the baseline methods. In addition, our proposed method also achieves a competitive performance compared to state-of-the-art methods.
      Cover page: Improving 3D U-Net for Brain Tumor Segmentation by Utilizing Lesion Prior
      • Article
      • Peer Reviewed
      UC Irvine Previously Published Works (2020)
      The volume of stroke lesion is the gold standard for predicting the clinical outcome of stroke patients. However, the presence of stroke lesion may cause neural disruptions to other brain regions, and these potentially damaged regions may affect the clinical outcome of stroke patients. In this paper, we introduce the tractographic feature to capture these potentially damaged regions and predict the modified Rankin Scale (mRS), which is a widely used outcome measure in stroke clinical trials. The tractographic feature is built from the stroke lesion and average connectome information from a group of normal subjects. The tractographic feature takes into account different functional regions that may be affected by the stroke, thus complementing the commonly used stroke volume features. The proposed tractographic feature is tested on a public stroke benchmark Ischemic Stroke Lesion Segmentation 2017 and achieves higher accuracy than the stroke volume and the state-of-the-art feature on predicting the mRS grades of stroke patients. Also, the tractographic feature yields a lower average absolute error than the commonly used stroke volume feature.
        Cover page: Predicting Clinical Outcome of Stroke Patients with Tractographic Feature.
        • Article
        • Peer Reviewed
        UC Irvine Previously Published Works (2019)
        In this work, we utilize T1-weighted MR images and StackNet to predict fluid intelligence in adolescents. Our framework includes feature extraction, feature normalization, feature denoising, feature selection, training a StackNet, and predicting fluid intelligence. The extracted feature is the distribution of different brain tissues in different brain parcellation regions. The proposed StackNet consists of three layers and 11 models. Each layer uses the predictions from all previous layers including the input layer. The proposed StackNet is tested on a public benchmark Adolescent Brain Cognitive Development Neurocognitive Prediction Challenge 2019 and achieves a mean squared error of 82.42 on the combined training and validation set with 10-fold cross-validation. In addition, the proposed StackNet also achieves a mean squared error of 94.25 on the testing data. The source code is available on GitHub.
          Cover page: Predicting Fluid Intelligence of Children Using T1-Weighted MR Images and a StackNet
          • Article
          • Peer Reviewed
          UC Santa Barbara Previously Published Works (2020)
          The manual brain tumor annotation process is time consuming and resource consuming, therefore, an automated and accurate brain tumor segmentation tool is greatly in demand. In this paper, we introduce a novel method to integrate location information with the state-of-the-art patch-based neural networks for brain tumor segmentation. This is motivated by the observation that lesions are not uniformly distributed across different brain parcellation regions and that a locality-sensitive segmentation is likely to obtain better segmentation accuracy. Toward this, we use an existing brain parcellation atlas in the Montreal Neurological Institute (MNI) space and map this atlas to the individual subject data. This mapped atlas in the subject data space is integrated with structural Magnetic Resonance (MR) imaging data, and patch-based neural networks, including 3D U-Net and DeepMedic, are trained to classify the different brain lesions. Multiple state-of-the-art neural networks are trained and integrated with XGBoost fusion in the proposed two-level ensemble method. The first level reduces the uncertainty of the same type of models with different seed initializations, and the second level leverages the advantages of different types of neural network models. The proposed location information fusion method improves the segmentation performance of state-of-the-art networks including 3D U-Net and DeepMedic. Our proposed ensemble also achieves better segmentation performance compared to the state-of-the-art networks in BraTS 2017 and rivals state-of-the-art networks in BraTS 2018. Detailed results are provided on the public multimodal brain tumor segmentation (BraTS) benchmarks.
            Cover page: Improving Patch-Based Convolutional Neural Networks for MRI Brain Tumor Segmentation by Leveraging Location Information
            • Article
            • Peer Reviewed
            UC Santa Barbara Previously Published Works (2019)
            We consider the problem of accurately identifying cell boundaries and labeling individual cells in confocal microscopy images, specifically, 3D image stacks of cells with tagged cell membranes. Precise identification of cell boundaries, their shapes, and quantifying inter-cellular space leads to a better understanding of cell morphogenesis. Towards this, we outline a cell segmentation method that uses a deep neural network architecture to extract a confidence map of cell boundaries, followed by a 3D watershed algorithm and a final refinement using a conditional random field. In addition to improving the accuracy of segmentation compared to other state-of-the-art methods, the proposed approach also generalizes well to different datasets without the need to retrain the network for each dataset. Detailed experimental results are provided, and the source code is available on GitHub 1.
              Cover page: Accurate 3D Cell Segmentation Using Deep Features and CRF Refinement
              • Article
              • Peer Reviewed
              UC Santa Barbara Previously Published Works (2020)
              [This corrects the article DOI: 10.3389/fnins.2019.01449.].
                Cover page: Corrigendum: Improving Patch-Based Convolutional Neural Networks for MRI Brain Tumor Segmentation by Leveraging Location Information
                • Article
                • Peer Reviewed
                UC Irvine Previously Published Works (2019)
                This paper introduces a novel methodology to integrate human brain connectomics and parcellation for brain tumor segmentation and survival prediction. For segmentation, we utilize an existing brain parcellation atlas in the MNI152 1 mm space and map this parcellation to each individual subject data. We use deep neural network architectures together with hard negative mining to achieve the final voxel level classification. For survival prediction, we present a new method for combining features from connectomics data, brain parcellation information, and the brain tumor mask. We leverage the average connectome information from the Human Connectome Project and map each subject brain volume onto this common connectome space. From this, we compute tractographic features that describe potential neural disruptions due to the brain tumor. These features are then used to predict the overall survival of the subjects. The main novelty in the proposed methods is the use of normalized brain parcellation data and tractography data from the human connectome project for analyzing MR images for segmentation and survival prediction. Experimental results are reported on the BraTS2018 dataset.
                  Cover page: Brain Tumor Segmentation and Tractographic Feature Extraction from Structural MR Images for Overall Survival Prediction
                  • Article
                  • Peer Reviewed
                  UC Santa Barbara Previously Published Works (2016)
                  We present an unsupervised three-dimensional feature clustering algorithm to gather the mTOP2016 challenge data into 3 groups. We use the brain MR-T1, diffusion tensor fractional anisotropy, and diffusion tensor mean diffusivity images provided by the mTOP2016 competition. A distance-based size constraint method for data clustering is used. The proposed approach achieves 0.267 adjusted rand index and 0.3556 homogeneity score within the 15 labeled subjects, corresponding to 10 correctly classified data items. Based on visual exploration of the data, we believe that a localized analysis of the lesion regions, using the computed tractography data, is a promising direction to pursue.
                    Cover page: Unsupervised 3-D Feature Learning for Mild Traumatic Brain Injury
                    Top