A5050560717- Functional Brain Connectivity Studies
- Brain Tumor Detection and Classification
- Advanced MRI Techniques and Applications
- Advanced Neuroimaging Techniques and Applications
- EEG and Brain-Computer Interfaces
- Medical Image Segmentation Techniques
- Domain Adaptation and Few-Shot Learning
- Neural dynamics and brain function
- Radiomics and Machine Learning in Medical Imaging
- Image Retrieval and Classification Techniques
- Dementia and Cognitive Impairment Research
- Machine Learning in Healthcare
- AI in cancer detection
- COVID-19 diagnosis using AI
- Neonatal and fetal brain pathology
- Medical Imaging Techniques and Applications
- Advanced Image and Video Retrieval Techniques
- Mental Health Research Topics
- Face and Expression Recognition
- Metal and Thin Film Mechanics
- Medical Imaging and Analysis
- Advancements in Battery Materials
- Neurological Disease Mechanisms and Treatments
- Minerals Flotation and Separation Techniques
- Advanced Image Fusion Techniques
University of North Carolina at Chapel Hill
2016-2025
Sichuan University
2023-2025
West China Hospital of Sichuan University
2025
State Key Laboratory of Biotherapy
2023-2025
Shandong Normal University
2024
Imaging Center
2015-2024
Liaoning University
2019-2024
Wuhan University
2012-2024
Beihang University
2022-2024
Xi’an University
2018-2024
Structural magnetic resonance imaging (sMRI) has been widely used for computer-aided diagnosis of neurodegenerative disorders, e.g., Alzheimer's disease (AD), due to its sensitivity morphological changes caused by brain atrophy. Recently, a few deep learning methods (e.g., convolutional neural networks, CNNs) have proposed learn task-oriented features from sMRI AD diagnosis, and achieved superior performance than the conventional learning-based using hand-crafted features. However, these...
As shown in the literature, methods based on multiple templates usually achieve better performance, compared with those using only a single template for processing medical images. However, most existing multi-template simply average or concatenate sets of features extracted from different templates, which potentially ignores important structural information contained data. Accordingly, this paper, we propose novel relationship induced learning method automatic diagnosis Alzheimer's disease...
The fusion of complementary information contained in multi-modality data [e.g., magnetic resonance imaging (MRI), positron emission tomography (PET), and genetic data] has advanced the progress automated Alzheimer's disease (AD) diagnosis. However, based AD diagnostic models are often hindered by missing data, i.e., not all subjects have complete data. One simple solution used many previous studies is to discard samples with modalities. this significantly reduces number training samples,...
One of the major challenges in anatomical landmark detection, based on deep neural networks, is limited availability medical imaging data for network learning. To address this problem, we present a two-stage task-oriented learning method to detect large-scale landmarks simultaneously real time, using training data. Specifically, our consists two convolutional networks (CNN), with each focusing one specific task. alleviate problem data, first stage, propose CNN regression model millions image...
Autism spectrum disorder (ASD) is a neurodevelopmental that characterized by wide range of symptoms. Identifying biomarkers for accurate diagnosis crucial early intervention ASD. While multi-site data increase sample size and statistical power, they suffer from inter-site heterogeneity. To address this issue, we propose adaption framework via low-rank representation decomposition (maLRR) ASD identification based on functional MRI (fMRI). The main idea to determine common the multiple sites,...
Brain connectivity alterations associated with mental disorders have been widely reported in both functional MRI (fMRI) and diffusion (dMRI). However, extracting useful information from the vast amount of afforded by brain networks remains a great challenge. Capturing network topology, graph convolutional (GCNs) demonstrated to be superior learning representations tailored for identifying specific disorders. Existing construction techniques generally rely on parcellation define...
Machine learning methods have successfully been used to predict the conversion of mild cognitive impairment (MCI) Alzheimer's disease (AD), by classifying MCI converters (MCI-C) from nonconverters (MCI-NC). However, most existing construct classifiers using data one particular target domain (e.g., MCI), and ignore in other related domains AD normal control (NC)) that may provide valuable information improve prediction performance. To address is limitation, we develop a novel transfer method...
Software defect prediction (SDP), which classifies software modules into defect-prone and not-defect-prone categories, provides an effective way to maintain high quality systems. Most existing SDP models attempt attain lower classification error rates other than misclassification costs. However, in many real-world applications, misclassifying as ones usually leads higher costs ones. In this paper, we first propose a new two-stage cost-sensitive learning (TSCS) method for SDP, by utilizing...
Early identification of dementia at the stage mild cognitive impairment (MCI) is crucial for timely diagnosis and intervention Alzheimer's disease (AD). Although several pioneering studies have been devoted to automated AD based on resting-state functional magnetic resonance imaging (rs-fMRI), their performance somewhat limited due non-effective mining spatial-temporal dependency. Besides, few these existing approaches consider explicit detection modeling discriminative brain regions (i.e.,...
Multitemplate-based brain morphometric pattern analysis using magnetic resonance imaging has been recently proposed for automatic diagnosis of Alzheimer's disease (AD) and its prodromal stage (i.e., mild cognitive impairment or MCI). In such methods, multiview morphological patterns generated from multiple templates are used as feature representation images. However, existing multitemplate-based methods often simply assume that each class is represented by a specific type data distribution...
Magnetic resonance fingerprinting (MRF) is a quantitative imaging technique that can simultaneously measure multiple important tissue properties of human body. Although MRF has demonstrated improved scan efficiency as compared to conventional techniques, further acceleration still desired for translation into routine clinical practice. The purpose this paper accelerate acquisition by developing new quantification method allows accurate with fewer sampling data. Most the existing approaches...
Abstract Multi‐atlas based methods have been recently used for classification of Alzheimer's disease (AD) and its prodromal stage, that is, mild cognitive impairment (MCI). Compared with traditional single‐atlas methods, multiatlas adopt multiple predefined atlases thus are less biased by a certain atlas. However, most existing simply average or concatenate the features from atlases, which may ignore potentially important diagnosis information related to anatomical differences among...