AI-Powered Pneumonia Detection and Classification: An Extensive Exploration of Deep Learning Technique and Multi-Modal Imaging Techniques
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
Pneumonia is a severe respiratory illness that continues to be a major health challenge worldwide, especially impacting infants and older others individuals, patients with weakened immunity.Timely and precise identification is essential to avoid severe complications, yet current diagnostic practices using chest radiographs and CT scans are highly dependent on radiologists, making the process slower and prone to subjective errors.This work proposes an intelligent framework for pneumonia detection, classification, and diagnosis through the utilization of 3D Convolutional Neural networks combined with integrated medical imaging techniques.By analyzing volumetric image data, the proposed system is capable of capturing spatial and depth-related features that improve classification of pneumonia types, including bacterial, viral, and normal conditions.To further enhance reliability, the framework incorporates multi-modal imaging integration along with explainable AI techniques such as heatmap visualization to highlight infected lung regions, ensuring clinical interpretability.The system's effectiveness will be validated using publicly available datasets and evaluated with evaluation measure including accuracy, precision, recall.Additionally, a supportive interface will be designed for healthcare professionals, enabling faster, trustworthy, and interpretable diagnostic outcomes.This approach aims AI-based solution that improves healthcare decision-making and sets a new direction for medical image analysis in respiratory disease diagnosis.
Fields of Study & Subjects
COVID-19 diagnosis using AI
Radiomics and Machine Learning in Medical Imaging
Pneumonia and Respiratory Infections
Publication details
- Journal / source
- Advances in intelligent systems research/Advances in Intelligent Systems Research
- Pages
- 375–390
- Publication date
- 1 April 2026
- Language
- EN
- Document type
- Book chapter