A5020517620- Imbalanced Data Classification Techniques
- Machine Fault Diagnosis Techniques
- Industrial Vision Systems and Defect Detection
- Metaheuristic Optimization Algorithms Research
- Advanced Multi-Objective Optimization Algorithms
- Anomaly Detection Techniques and Applications
- Cardiac Imaging and Diagnostics
- Robotic Path Planning Algorithms
- Advanced Computational Techniques and Applications
- Electricity Theft Detection Techniques
- Mobile Health and mHealth Applications
- Manufacturing Process and Optimization
- Face and Expression Recognition
- Advanced Neural Network Applications
- Machine Learning and Data Classification
- Thyroid Cancer Diagnosis and Treatment
- Advanced X-ray and CT Imaging
- Phytochemical and Pharmacological Studies
- Child Development and Digital Technology
- Medical Imaging Techniques and Applications
- Agriculture and Biological Studies
- Dental Radiography and Imaging
- Robot Manipulation and Learning
- Digital Mental Health Interventions
- Diverse Scientific Research in Ukraine
Guizhou University
2019-2025
Sichuan University
2021-2025
Shanghai Institute of Technology
2022-2024
Nanjing University
2022
Tsinghua University
2014
Guangzhou University
2010
Herzog Hospital
1993
Geriatric Research Education and Clinical Center
1993
Learning primitive (i.e., attribute and object) concepts from seen compositions is the primary challenge of Compositional Zero-Shot (CZSL). Existing CZSL solutions typically rely on oversimplified data assumptions, e.g., modeling each with a single centroid representation, ignoring natural diversities (resp. when coupled different objects attribute). In this work, we develop ClusPro, robust clustering-based prototype mining framework for that defines conceptual boundaries primitives through...
Abstract The prediction of tool remaining useful life (RUL) is crucial for ensuring the quality and reliability components in high-end equipment. However, practical applications, wear data typically sparse exhibits irregular time-series patterns, which pose significant challenges to development accurate reliable predictive models RUL. To address these challenges, this paper presents a novel interpretable augmentation collaborative regression model (DL-SVRs). DL-SVRs deep framework with...
99mTc-pertechnetate thyroid scintigraphy is a valid complementary avenue for evaluating disease in the clinic, image feature of scintigram relatively simple but interpretation still has moderate consistency among physicians. Thus, we aimed to develop an artificial intelligence (AI) system automatically classify four patterns scintigram.We collected 3087 scintigrams from center 1 construct training dataset (n = 2468) and internal validating 619), another 302 cases 2 as external datasets. Four...
In this study, the volatile components in 40 samples of Tartary buckwheat and common from 6 major producing areas China were analyzed. A total 77 substances identified, among which aldehydes hydrocarbons main components. Odor activity value analysis revealed 26 aromatic compounds, with making a significant contribution to aroma buckwheat. Seven key compounds that could be used distinguish identified. The orthogonal partial least squares-discriminant was effectively classify different areas....
Hemerocallis L. possesses abundant germplasm resources and holds significant value in terms of ornamental, edible, medicinal aspects. However, the quality characteristics vary significantly depending on different varieties. Selection a high-quality variety with characteristic aroma can increase economic flowers. The analytic hierarchy process (AHP) is an effective decision-making method for comparing evaluating multiple dimensions. By applying AHP, aromatic character 60 varieties flowers...