A5101725247- AI in cancer detection
- Radiomics and Machine Learning in Medical Imaging
- Muscle Physiology and Disorders
- Advanced Image and Video Retrieval Techniques
- Cell Image Analysis Techniques
- Image Retrieval and Classification Techniques
- Medical Image Segmentation Techniques
- Digital Imaging for Blood Diseases
- Image Processing Techniques and Applications
- Medical Imaging Techniques and Applications
- Face and Expression Recognition
- Generative Adversarial Networks and Image Synthesis
- Cardiomyopathy and Myosin Studies
- Advanced Image Processing Techniques
- Multimodal Machine Learning Applications
- Advanced Neural Network Applications
- Neurogenetic and Muscular Disorders Research
- COVID-19 diagnosis using AI
- Visual Attention and Saliency Detection
- Nutrition and Health in Aging
- Pancreatic and Hepatic Oncology Research
- Advanced X-ray and CT Imaging
- Advanced MRI Techniques and Applications
- Topic Modeling
- Domain Adaptation and Few-Shot Learning
Northwest University
2025
Dalian Medical University
2025
First Affiliated Hospital of Dalian Medical University
2025
Peking University Shenzhen Hospital
2023-2024
RELX Group (United States)
2023
Istituto Nazionale di Fisica Nucleare, Sezione di Milano
2023
Faculty of 1000 (United States)
2023
University of Leicester
2022
State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation
2022
Modular Genetics (United States)
2022
Synthesized medical images have several important applications, e.g., as an intermedium in cross-modality image registration and supplementary training samples to boost the generalization capability of a classifier. Especially, synthesized computed tomography (CT) data can provide X-ray attenuation map for radiation therapy planning. In this work, we propose generic synthesis approach with following targets: 1) synthesizing realistic looking 3D using unpaired data, 2) ensuring consistent...
Computer-aided image analysis of histopathology specimens could potentially provide support for early detection and improved characterization diseases such as brain tumor, pancreatic neuroendocrine tumor (NET), breast cancer. Automated nucleus segmentation is a prerequisite various quantitative analyses including automatic morphological feature computation. However, it remains to be challenging problem due the complex nature images. In this paper, we propose learning-based framework robust...
The inability to interpret the model prediction in semantically and visually meaningful ways is a well-known shortcoming of most existing computer-aided diagnosis methods. In this paper, we propose MDNet establish direct multimodal mapping between medical images diagnostic reports that can read images, generate reports, retrieve by symptom descriptions, visualize attention, provide justifications network process. includes an image language model. proposed enhance multi-scale feature...
This paper presents a novel method to deal with the challenging task of generating photographic images conditioned on semantic image descriptions. Our introduces accompanying hierarchical-nested adversarial objectives inside network hierarchies, which regularize mid-level representations and assist generator training capture complex statistics. We present an extensile single-stream architecture better adapt jointed discriminators push generated up high resolutions. adopt multi-purpose loss...
X-linked myotubular myopathy (XLMTM) results from MTM1 gene mutations and myotubularin deficiency. Most XLMTM patients develop severe muscle weakness leading to respiratory failure death, typically within 2 years of age. Our objective was evaluate the efficacy safety systemic therapy in p.N155K canine model by performing a dose escalation study. A recombinant adeno-associated virus serotype 8 (rAAV8) vector expressing (cMTM1) under muscle-specific desmin promoter (rAAV8-cMTM1) administered...
The goal of MRI reconstruction is to restore a high fidelity image from partially observed measurements. This partial view naturally induces uncertainty that can only be reduced by acquiring additional In this paper, we present novel method for that, at inference time, dynamically selects the measurements take and iteratively refines prediction in order best reduce error and, thus, its uncertainty. We validate our on large scale knee dataset, as well ImageNet. Results show (1) system...
Although attention mechanisms have been widely used in deep learning for many tasks, they are rarely utilized to solve multiple instance (MIL) problems, where only a general category label is given instances contained one bag. Additionally, previous MIL methods firstly utilize the mechanism learn weights and then employ fully connected layer predict bag label, so that prediction largely determined by effectiveness of learned weights. To alleviate this issue, paper, we propose novel loss...
Objective: Nucleus recognition is a critical yet challenging step in histopathology image analysis, for example, Ki67 immunohistochemistry stained images. Although many automated methods have been proposed, most use multi-stage processing pipeline to categorize nuclei, leading cumbersome, low-throughput, and error-prone assessments. To address this issue, we propose novel deep fully convolutional network single-stage nucleus recognition. Methods: Instead of conducting direct pixel-wise...
Non-small cell lung cancer (NSCLC), the most common type of cancer, is one serious diseases causing death for both men and women. Computer-aided diagnosis survival prediction NSCLC, great importance in providing assistance to personalize therapy planning patients. In this paper we have proposed an integrated framework NSCLC computer-aided analysis using novel image markers. The entire biomedical imaging informatics consists detection, segmentation, classification, discovery markers,...
Nebulin is a giant filamentous protein that coextensive with the actin filaments of skeletal muscle sarcomere. mutations are main cause nemaline myopathy (NEM), typical adult patients having low expression nebulin, yet roles nebulin in remain poorly understood. To establish nebulin's functional muscle, we studied novel conditional KO (Neb cKO) mouse model which deletion was driven by creatine kinase (MCK) promotor. Neb cKO mice born high levels their muscles, but within weeks after birth...
Computer-aided diagnosis of medical images requires thorough analysis image details. For example, examining all cells enables fine-grained categorization histopathological images. Traditional computational methods may have efficiency issues when performing such detailed analysis. In this paper, we propose a robust and scalable solution to achieve this. Specifically, segmentation method is developed delineate region-of-interests (e.g., cells) accurately, using hierarchical voting repulsive...