A5059768795- Advanced Multi-Objective Optimization Algorithms
- Topic Modeling
- Metaheuristic Optimization Algorithms Research
- Evolutionary Algorithms and Applications
- Advanced Text Analysis Techniques
- Retinal Imaging and Analysis
- Glaucoma and retinal disorders
- Natural Language Processing Techniques
- Privacy-Preserving Technologies in Data
- Topology Optimization in Engineering
- Mobile Crowdsensing and Crowdsourcing
- Infrared Target Detection Methodologies
- Flow Measurement and Analysis
- Digital Imaging for Blood Diseases
- Retinal Diseases and Treatments
- Advanced Measurement and Detection Methods
- CO2 Reduction Techniques and Catalysts
- Energy Load and Power Forecasting
- Electrocatalysts for Energy Conversion
- Advanced Algorithms and Applications
- Evaluation Methods in Various Fields
- Heat Transfer and Optimization
- Industrial Vision Systems and Defect Detection
- Smart Grid and Power Systems
- Biomedical Text Mining and Ontologies
National University of Singapore
2009
Surrogate-assisted evolutionary algorithms for expensive optimization problems have gained considerable attention in recent years. In many real-world problems, we may face with multiple optimal solutions. Locating optima such is qualitatively challenging. This study proposes a surrogate-assisted differential evolution based on region decomposition to seek multimodal problems. this study, designed three major components: 1) the adaptive decomposition, 2) multilayer perceptron-based global...
Objectives. This study set out to develop and validate a risk prediction tool for the early detection of heart failure (HF) onset using real-world electronic health records (EHRs). Background. While existing HF assessment models have shown promise in clinical settings, they are often tailored specific medical conditions, limiting their generalizability. Moreover, most methods rely on hand-crafted features, making it difficult capture high-dimensional, sparse, temporal nature EHR data, thus...
This paper proposes a method to detect the macula in retinal fundus image automatically. The makes use of optic disc height obtained from ARGALI define region interest. Regions dark spots are then detected by finding coordinates with lowest pixel intensity and determining average neighbourhood intensities. These regions ranked determine containing macula. algorithm was tested on 162 images, an accuracy 98.8% achieved. results promising for further development this AMD studies physiology localization.
We propose a method for improving the accuracy of optic cup detected from ARGALI system. This makes use key points branching large vessels, analysis intensity variation and kinks small vessels to obtain an enhanced cup. Measures used assess detection showed 11% 40% improvement in mean average overlap relative area difference respectively over previous method. The CDR error was also shown be reduced less than 0.1CDR units. improved is more consistent with clinical ground truth, facilitating...