A5100701917- Radiomics and Machine Learning in Medical Imaging
- Advanced Neural Network Applications
- COVID-19 diagnosis using AI
- Image Enhancement Techniques
- Advanced Image and Video Retrieval Techniques
- Advanced X-ray and CT Imaging
- Topic Modeling
- Image and Signal Denoising Methods
- Privacy-Preserving Technologies in Data
- Multimodal Machine Learning Applications
- AI in cancer detection
- Advanced Optical Sensing Technologies
- Domain Adaptation and Few-Shot Learning
- Lung Cancer Diagnosis and Treatment
- Machine Learning and Data Classification
- Experimental Behavioral Economics Studies
- IoT and Edge/Fog Computing
- Brain Tumor Detection and Classification
- Hepatocellular Carcinoma Treatment and Prognosis
- Cell Image Analysis Techniques
- Advanced Vision and Imaging
- Caveolin-1 and cellular processes
- Colorectal Cancer Screening and Detection
- Speech and Audio Processing
- Advanced Database Systems and Queries
University of Indianapolis
2022-2025
University College London
2023-2025
Tianjin First Center Hospital
2024-2025
Tianjin Medical University
2024-2025
Nankai University
2024-2025
Indiana University – Purdue University Indianapolis
2021-2025
Beijing University of Posts and Telecommunications
2021-2024
Institute of Semiconductors
2019-2024
Chinese Academy of Sciences
2019-2024
Hainan University
2024
Purpose: Accurate segmentation of lung and infection in COVID-19 CT scans plays an important role the quantitative management patients. Most existing studies are based on large private annotated datasets that impractical to obtain from a single institution, especially when radiologists busy fighting coronavirus disease. Furthermore, it is hard compare current methods as they developed different datasets, trained settings, evaluated with metrics. Methods: To promote development data-efficient...
Circular dorsal ruffles (CDRs) are large and rounded membrane that function as precursors of macropinocytosis. We recently reported CDRs form in Hep3B hepatocellular carcinoma (HCC) cells, but not Huh7 HepG2 HCC cells or LO2 suggesting an unknown molecular mechanism implicates malignancy through macropinocytosis uptake excessive extracellular nutrients. In this study, we investigated the cellular role by focusing on GTPase-activating protein ARAP1. ARAP1 knock-out (KO) were generated....
BACKGROUND Proteins play a central role in regulating biological functions, and various pathways regulate their synthesis secretion. Endoplasmic reticulum-associated protein degradation (ERAD) is crucial for monitoring processing unfolded or misfolded proteins actively growing tumor cells. However, the of multiple ERAD complexes liver cancer remains unclear. AIM To elucidate effects SEL1L -mediated on Huh7 explore underlying mechanisms vivo vitro . METHODS cells were treated with inhibitor...
<title>Abstract</title> At-home rehabilitation for post-stroke patients presents significant challenges, as continuous, personalized care is often limited outside clinical settings. Additionally, the absence of comprehensive solutions addressing diverse needs in home environments complicates recovery efforts. Here, we introduce a smart platform that integrates wearable sensors, ambient monitoring, and large language model (LLM)-powered assistance to provide seamless health monitoring...
Research on screen content images (SCIs) attracts more attention as they are highly applied to image- and video-centric applications mobile other devices. It is important develop an efficient image-quality assessment (IQA) method for SCIs because IQA can guide optimize various image-processing methods improve user experience. In this paper, we propose a no-reference objective model including segmentation the analysis of local global perceptual feature representations. Since human visual...
Deep learning approaches have demonstrated success in no-reference image quality assessment tasks. However, due to the specific properties of screen content images (SCIs), deep neural networks for SCI are not as optimal those designed depicting natural scenes. In order tackle this discrepancy, a "naturalization" module composed an upsampling layer and convolutional is proposed transform SCIs characteristics more similar that images. addition, new model architecture along with data...
This paper is concerned with the mitigation of backscatter effects in a single gated image. A range-intensity-profile prior dehazing method proposed to estimate scene depth and finely remove water at different depths for underwater range-gated imaging. It based on that target intensity distributed range profiles images. The transmission depth-noise map are then calculated from depth. high-quality image restored by subtracting dividing transmission. simulation experimental results show works...
Earlier-stage evaluations of a new AI architecture/system need affordable benchmarks. Only using few component benchmarks like MLPerf alone in the other stages may lead to misleading conclusions. Moreover, learning dynamics are not well understood, and benchmarks' shelf-life is short. This paper proposes balanced benchmarking methodology. We use real-world cover factors space that impacts most considerable extent. After performing an exhaustive survey on Internet service domains, we identify...
Objective This study aimed to compare the health economic value of a non-invasive prenatal testing (NIPT) strategy against second-trimester triple screening (STS) for detection Down syndrome based on real-world data from China. Design A decision-analytical model was developed cost-effectiveness five strategies societal perspective. Cost and probability input were obtained surveys published sources. Setting Participants Women with singleton pregnancy. Interventions The were: (A) maternal age...
As one of the typical settings Federated Learning (FL), cross-silo FL allows organizations to jointly train an optimal Machine (ML) model. In this case, some may try obtain global model without contributing their local training, lowering social welfare. paper, we interactions among in as a public goods game for first time and theoretically prove that there exists dilemma where maximum welfare is not achieved Nash equilibrium. To over-come dilemma, employ Multi-player Multi-action...
With the further development of knowledge graphs, many weighted graphs (WKGs) have been published and greatly promote various applications. However, current deterministic graph embedding algorithms cannot encode well. This paper gives a promising framework WeExt that can extend models to enable them learn embeddings. In addtion, we introduce link prediction evaluate models' performance on completing WKGs. Finally, give concrete implementation based two translational distance semantic...