A5100775917- Radiomics and Machine Learning in Medical Imaging
- AI in cancer detection
- Advanced MRI Techniques and Applications
- Cardiac Imaging and Diagnostics
- Water Quality Monitoring Technologies
- Medical Imaging Techniques and Applications
- Advanced Neural Network Applications
- Water Quality Monitoring and Analysis
- Advanced Computational Techniques and Applications
- Medical Image Segmentation Techniques
- Digital Imaging for Blood Diseases
- Neural Networks and Applications
- Advanced Image Processing Techniques
- Epilepsy research and treatment
- Advanced X-ray and CT Imaging
- Domain Adaptation and Few-Shot Learning
- Machine Learning and ELM
- Image and Signal Denoising Methods
- Text and Document Classification Technologies
- Natural Language Processing Techniques
- Cardiovascular Function and Risk Factors
- Sports Dynamics and Biomechanics
- Machine Learning and Data Classification
- Winter Sports Injuries and Performance
- Advanced Photocatalysis Techniques
University of Oxford
2022-2025
University of Birmingham
2024-2025
Beihang University
2023-2025
Physical Sciences (United States)
2025
Hangzhou Dianzi University
2023-2024
University College London
2019-2023
National Hospital for Neurology and Neurosurgery
2019-2023
Queen Mary University of London
2020
Karolinska Institutet
2020
China XD Group (China)
2020
Recent years have seen increasing use of supervised learning methods for segmentation tasks. However, the predictive performance these algorithms depends on quality labels. This problem is particularly pertinent in medical image domain, where both annotation cost and inter-observer variability are high. In a typical label acquisition process, different human experts provide their estimates "true" labels under influence own biases competence levels. Treating noisy blindly as ground truth...
Supervised machine learning methods have been widely developed for segmentation tasks in recent years. However, the quality of labels has high impact on predictive performance these algorithms. This issue is particularly acute medical image domain, where both cost annotation and inter-observer variability are high. Different human experts contribute estimates "actual" a typical label acquisition process, influenced by their personal biases competency levels. The automatic algorithms limited...
In recent years, the deployment of supervised machine learning techniques for segmentation tasks has significantly increased. Nonetheless, annotation process extensive datasets remains costly, labor-intensive, and error-prone. While acquiring sufficiently large to train deep models is feasible, these often experience a distribution shift relative actual test data. This problem particularly critical in domain medical imaging, where it adversely affects efficacy automatic models. this work, we...
ABSTRACT Background Growth rods are the gold standard for treating early‐onset scoliosis (EOS). However, current treatments with growth do not optimize spinal in EOS patients, and frequent distraction surgeries significantly increase complications, imposing considerable economic psychological burdens on patients. An improved rod is urgently required to address need dynamic external regulation. Methods This study designed a novel (NGR) unidirectional sliding regulation capabilities. By...
Chronic kidney disease (CKD) is a major global health issue, affecting over 10% of the population and causing significant mortality. While biopsy remains gold standard for CKD diagnosis treatment, lack comprehensive benchmarks pathology segmentation hinders progress in field. To address this, we organized Kidney Pathology Image Segmentation (KPIs) Challenge, introducing dataset that incorporates preclinical rodent models with 10,000 annotated glomeruli from 60+ Periodic Acid Schiff...
Cardiac magnetic resonance (CMR) images play a growing role in the diagnostic imaging of cardiovascular diseases. Full coverage left ventricle (LV), from base to apex, is basic criterion for CMR image quality and necessary accurate measurement cardiac volume functional assessment. Incomplete LV identified through visual inspection, which time consuming usually done retrospectively assessment large cohorts. This paper proposes novel automatic method determining by using Fisher-discriminative...
Boards are crucial to shareholder wealth. Yet, little is known about how oversight affects director incentives. Using exogenous industry shocks institutional investor portfolios, we find that distraction weakens board oversight. Distracted institutions less likely discipline ineffective directors. Consequently, independent directors face weaker monitoring incentives and exhibit poor performance; also more frequently appointed. Moreover, the adverse effects of on various corporate governance...
We examine how friendly boards affect firm innovation. Using CEO-director social connections as a measure of board friendliness, we find that firms with create more patents and citations. The positive relation between innovation are pronounced when firms’ advisory needs higher or operate in innovative industries. Friendly also associated value, especially have is an important source value. Our results support the view on perspective directors serve valuable advisors to CEOs.
Image corruptions are common in the real world, for example images wild may come with unknown blur, bias field, noise, or other kinds of non-linear distributional shifts, thus hampering encoding methods and rendering downstream task unreliable. upgradation requires a complicated balance between high-level contextualised information spatial specific details. Existing approaches to solving problems designed focus on single corruption, which unavoidably results poor performance when...