Аннотация
This research paper investigates the application of Convolutional Neural Networks (CNNs) for the classification of pneumonia using chest X-ray images. Through rigorous experimentation and data analysis, the study demonstrates the model's impressive learning capabilities, achieving a notable accuracy of 96% in pneumonia classification. The consistent decrease in training and validation losses across 25 learning epochs underscores the model's adaptability and proficiency. However, the research also highlights the challenge of dataset imbalance and the need for improved model interpretability. These findings emphasize the potential of deep learning models in enhancing pneumonia diagnosis but also underscore the importance of addressing existing limitations. The study calls for future research to explore techniques for addressing dataset imbalances, enhance model interpretability, and extend the scope to address nuanced diagnostic challenges within the field of pneumonia classification. Ultimately, this research contributes to the advancement of medical image analysis and the potential for deep learning models to aid in early and accurate pneumonia diagnosis, thereby improving patient care and clinical outcomes.