A5100378764- Radiomics and Machine Learning in Medical Imaging
- COVID-19 diagnosis using AI
- Stock Market Forecasting Methods
- Energy Load and Power Forecasting
- Immune Cell Function and Interaction
- Bayesian Modeling and Causal Inference
- Grey System Theory Applications
- Cancer Immunotherapy and Biomarkers
- CAR-T cell therapy research
- Multi-Criteria Decision Making
- Smart Parking Systems Research
- Toxin Mechanisms and Immunotoxins
- Pharmacological Effects of Natural Compounds
- Systemic Lupus Erythematosus Research
- Mesenchymal stem cell research
- Lipoproteins and Cardiovascular Health
- Seismology and Earthquake Studies
- Artificial Intelligence in Healthcare and Education
- Data Stream Mining Techniques
- Statistical Methods and Inference
- Pharmaceutical Economics and Policy
- Advanced Radiotherapy Techniques
- Bone and Dental Protein Studies
- Forecasting Techniques and Applications
- Water Quality Monitoring and Analysis
University of Technology Sydney
2025
Chongqing Medical University
2023-2024
Shandong University of Science and Technology
2022
Beijing Jiaotong University
2022
Guangzhou University of Chinese Medicine
2022
Peng Cheng Laboratory
2021
Liaoning University of Traditional Chinese Medicine
2021
General Hospital of Shenyang Military Region
2018
Rutgers, The State University of New Jersey
2009
The state-of-the-art online learning models generally conduct a single gradient descent when new sample arrives and thus suffer from suboptimal model weights. To this end, we introduce an broad system framework with closed-form solutions for each update. Different employing existing incremental algorithms tasks, which tend to incur degraded accuracy expensive update overhead, design effective weight estimation algorithm efficient updating strategy remedy the above two deficiencies,...
Recent advances in Reinforcement Learning from Human Feedback (RLHF) have shown that KL-regularization plays a pivotal role improving the efficiency of RL fine-tuning for large language models (LLMs). Despite its empirical advantage, theoretical difference between KL-regularized and standard remains largely under-explored. While there is recent line work on analysis objective decision making \citep{xiong2024iterative, xie2024exploratory,zhao2024sharp}, these analyses either reduce to...
Abstract Hotspot driver mutations presented by human leukocyte antigens might be recognized anti-tumor T cells. Based on their advantages of tumor-specificity and immunogenicity, neoantigens derived from hotspot mutations, such as PIK3CA H1047L , may serve emerging targets for cancer immunotherapies. NetMHCpan V4.1 was utilized predicting neoepitopes mutation. Using in vitro stimulation, antigen-specific cells targeting the HLA-A*11:01-restricted mutation were isolated healthy donor-derived...
This paper develops a theory for group Lasso using concept called strong sparsity. Our result shows that is superior to standard strongly group-sparse signals. provides convincing theoretical justification sparse regularization when the underlying structure consistent with data. Moreover, predicts some limitations of formulation are confirmed by simulation studies.
Chest radiographs clearly present the characteristics of lung lesions in patients with new coronary pneumonia, thus they can be leveraged to build a pneumonia detection model provide doctors favorable auxiliary diagnosis results. This paper proposes COVID-19 localization and identification approach based on yolov5 EfficientNet. Due inherent reasons such as computational complexity network structure, features single are usually limited representation, EfficientNet provides competitive feature...
Objective To evaluate the clinical value of 125I seeds implantation (RSI) for treatment lymph nodes metastases (LNM) in patients with 131I refractory differentiated thyroid carcinoma (RAIR-DTC). Methods A total 42 RAIR-DTC LNM (14 males, 28 females, median age 49 years) who underwent RSI guided by CT from January 2015 to June 2016 were retrospectively analyzed. All and their serum thyroglobulin (Tg) levels measured 2, 4 6 months post-treatment. The size Tg before after compared,...
Reinforcement learning from human feedback (RLHF) has emerged as the primary method for aligning large language models (LLMs) with preferences. The RLHF process typically starts by training a reward model (RM) using preference data. Conventional RMs are trained on pairwise responses to same user request, relative ratings indicating which response humans prefer. RM serves proxy However, due black-box nature of RMs, their outputs lack interpretability, cannot intuitively understand why an...
Forecasting regional economic activity is a progressively significant element of research. Regional prediction can directly assist local, national, and subnational policymakers. forecast be employed for defining macroeconomic forces, such as stock market cyclicality national labor movement. The recent advances machine learning (ML) models to solve the time series problem. Since parameters involved in ML model considerably influence performance, parameter tuning process also becomes...
Abstract Hotspot driver mutations presented by human leukocyte antigens (HLAs) can be recognized antitumor T cells. Based on their advantages of tumor-specificity and immunogenicity, neoantigens derived from hotspot mutations, such as PIK3CA H1047L may serve emerging targets for cancer immunotherapies. NetMHC V4.1 were utilized predicting neoepitopes mutation. Using in vitro stimulation, antigen specific cells targeting the HLA-A*11:01-restricted mutation isolated healthy donor-derived...
This research focuses on the direct and indirect systemic risk spillovers among East Asian, European, U.S. stock markets under COVID-19 pandemic. Based GARCH-Copula-CoVaR model, we construct spillover matrix of further explore path through R-vine. The empirical results first show in that Hong Kong exhibited largest change value after pandemic erupted, implying market is more sensitive to extreme events. Compared quiet period, Russia’s output increased significantly pandemic, environment...
Despite tremendous efforts, it is very challenging to generate a robust model assist in the accurate quantification assessment of COVID-19 on chest CT images. Due nature blurred boundaries, supervised segmentation methods usually suffer from annotation biases. To support unbiased lesion localisation and minimise labeling costs, we propose data-driven framework by only image-level labels. The can explicitly separate potential lesions original images, with help generative adversarial network...
At present, the prediction of stock market is one most popular and valuable research fields in financial field. More more scholars are engaged forecast, exploring law development, new science technology constantly applied to price forecast. In this paper, we proposed a closing model based on XGBoost Grid SearchCV algorithms. Experimental results show that our idea represents better performance than other machine learning methods. Specifically, RMSE value 1.39%, 2.43% 8.33% lower SVM...
C-HN, TZ and Y-FL conducted the experiments.YY helped in processing medicine