M
Mingxia Liu
Researcher at University of North Carolina at Chapel Hill
Publications - 174
Citations - 5776
Mingxia Liu is an academic researcher from University of North Carolina at Chapel Hill. The author has contributed to research in topics: Computer science & Feature selection. The author has an hindex of 32, co-authored 145 publications receiving 3174 citations. Previous affiliations of Mingxia Liu include Nanjing University of Aeronautics and Astronautics.
Papers
More filters
Journal ArticleDOI
Hierarchical Fully Convolutional Network for Joint Atrophy Localization and Alzheimer's Disease Diagnosis Using Structural MRI
TL;DR: A hierarchical fully convolutional network (H-FCN) is proposed to automatically identify discriminative local patches and regions in the whole brain sMRI, upon which multi-scale feature representations are then jointly learned and fused to construct hierarchical classification models for AD diagnosis.
Journal ArticleDOI
Landmark-based deep multi-instance learning for brain disease diagnosis
TL;DR: This paper adopts a data‐driven learning approach to discover disease‐related anatomical landmarks in the brain MR images, along with their nearby image patches, and learns an end‐to‐end MR image classifier for capturing both the local structural information conveyed by image patches located by landmarks and the global structural information derived from all detected landmarks.
Journal ArticleDOI
Joint Classification and Regression via Deep Multi-Task Multi-Channel Learning for Alzheimer's Disease Diagnosis
TL;DR: This work identifies the discriminative anatomical landmarks from MR images in a data-driven manner, and proposes a deep multi-task multi-channel convolutional neural network for joint classification and regression, using MRI data and demographic information of subjects.
Journal ArticleDOI
Domain Adaptation for Medical Image Analysis: A Survey
Hao Guan,Mingxia Liu +1 more
TL;DR: In this article, a survey of domain adaptation methods for medical image analysis is presented, and the authors categorize the existing methods into shallow and deep models, and each of them is further divided into supervised, semi-supervised and unsupervised methods.
Journal ArticleDOI
Detecting Anatomical Landmarks From Limited Medical Imaging Data Using Two-Stage Task-Oriented Deep Neural Networks
TL;DR: A two-stage task-oriented deep learning method to detect large-scale anatomical landmarks simultaneously in real time, using limited training data, consisting of two deep convolutional neural networks, with each focusing on one specific task.