AI-Enhanced Pneumonia Detection with Visual Interpretability
International Journal of Innovations in Science and Technology · 17.05.2025 · 118
DOI: 10.33411/ijist/1291 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Innovations in Science and Technology, online erschienen am 17.05.2025. Verfasst von Abrar Ahmed Shahok und 3 weiteren Autor:innen. Die Arbeit nennt 10 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Pneumonia is a serious lung infection that can be life-threatening, particularly for young children, the elderly, and people with weakened immune systems. Early detection is crucial but difficult because pneumonia signs on X-rays can be subtle. Many AI tools can help diagnose pneumonia, but they often work like “black boxes,” making it hard for doctors to trust their decisions. This study introduces a mobile app that uses Convolutional Neural Networks (CNNs) to detect pneumonia from X-rays. To improve transparency, we use Explainable AI (XAI) to highlight the areas of the X-ray that influenced the diagnosis. Additionally, we integrate a Large Language Model (LLM) to generate clear, structured medical reports. Our goal is to create a trustworthy and user-friendly tool for doctors in real-world settings.
Zitieren
APA
Shahok, A. A., Memon, F. A., Jalbani, K., & Shoaib, M. (2025). AI-Enhanced Pneumonia Detection with Visual Interpretability. International Journal of Innovations in Science and Technology, 118. https://doi.org/10.33411/ijist/1291
Vancouver
Shahok AA, Memon FA, Jalbani K, Shoaib M. AI-Enhanced Pneumonia Detection with Visual Interpretability. International Journal of Innovations in Science and Technology. 2025:118. doi:10.33411/ijist/1291
BibTeX
@article{Shahok2025,
author = {Shahok, Abrar Ahmed and Memon, Faizan Ali and Jalbani, Kaleemullah and Shoaib, M.},
title = {AI-Enhanced Pneumonia Detection with Visual Interpretability},
journal = {International Journal of Innovations in Science and Technology},
year = {2025},
pages = {118},
doi = {10.33411/ijist/1291}
}
EUVIMED