Deep learning for accurate detection of pneumonia from chest x-rays using VGG-16 architecture
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
The early detection and classification of comorbidities associated to diabetes is crucial for effective disease therapy and improved patient outcomes. A novel approach to the classification of diabetic comorbidities utilising multiple labels is presented in this study, which makes use of Beta Ant Colony Optimisation (BACO). The beta approach for colony optimisation (BACO) is an upgrade to the original ant colony optimisation (ACO) method, which finds the optimal classification parameters by balancing exploration and exploitation using beta distribution. This strategy, when combined with ML models, allows for accurate multi-label predictions and efficient feature selection in the proposed technique. The study's massive dataset consists of biochemical, demographic, and clinical details of people with diabetes. Across all labels, BACO achieves better performance than standard optimisation approaches in terms of accuracy, recall, precision, and F1-score. The proposed method outperforms baseline techniques in multi-label classification tasks and decreases computational complexity without sacrificing resilience to data imbalance. The findings demonstrate the potential of BACO to improve diabetes management via precision medicine. By allowing for more precise comorbidity identification, this approach enhances our understanding of the intricate relationship between diabetes and its comorbidities, leading to more personalised treatment strategies. Our next step is to test the model using larger datasets collected from many centres.
Fields of Study & Subjects
COVID-19 diagnosis using AI
Metaheuristic Optimization Algorithms Research
Artificial Intelligence in Healthcare
Publication details
- Journal / source
- AIP conference proceedings
- Volume
- 3401
- Pages
- 020142–020142
- Publication date
- 1 January 2026
- Language
- EN