Abstract
This paper presents an innovative approach using PyramidNet, a deep learning architecture renowned for hierarchical feature extraction, to automate the classification of medical images for disease diagnosis. The study utilizes a diverse dataset sourced from publicly available repositories, comprising various medical imaging modalities such as X-rays, MRIs, and histopathological slides. Rigorous preprocessing techniques, including normalization and augmentation, enhance the dataset’s suitability for deep learning applications. The PyramidNet model is trained using stochastic gradient descent with adaptive learning rates and weight decay, optimizing its performance for medical image classification tasks. Evaluation metrics, including accuracy (92.5%), precision (91.3%), recall (93.8%), and F1-score (92.5%), attest to the model’s robustness and efficacy. Integration of the validated PyramidNet model into clinical workflows demonstrates its potential to enhance diagnostic accuracy and efficiency in healthcare settings. This research underscores PyramidNet’s role in advancing automated medical image classification, highlighting its relevance for improving disease diagnosis and patient care outcomes
Keywords: Automated system, Deep learning, Disease diagnosis, Medical image classification, PyramidNet.