• Pratham R
• Gowthami Bhandarkar
• Rindiya N
• Praveen Bagali
• Nithya BP

Artificial intelligence (AI) integration into medical imaging has revolutionized diagnosis, especially in the detection of lung conditions like pneumonia. In this work, the effectiveness of deep transfer learning models VGG16, ResNet50, InceptionNet-v3, and EfficientNet is assessed in classifying pneumonia from chest X-rays. These models were trained on the RSNA Pneumonia Detection Challenge dataset (32,227 images) and evaluated using validation accuracy and AUC-ROC scores. VGG16 showed better classification performance (88% accuracy, 91.8% AUC-ROC), whereas EfficientNet was 99% confident in localizing lung inflammation. Our results highlight the promise of AI- based diagnostics to complement clinical decision-making in limited-resource environments.

Deep learning, convolutional neural networks, medical image analysis, screening for pneumonia, transfer learning, X-ray classification

"Deep Learning Approaches for Automated Pneumonia Diagnosis in Chest X-rays: A Comparative Study", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 7, page no.642-644, July-2025, Available :http://www.jetir.org/papers/JETIRGX06122.pdf

"Deep Learning Approaches for Automated Pneumonia Diagnosis in Chest X-rays: A Comparative Study", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 7, page no. pp642-644, July-2025, Available at : http://www.jetir.org/papers/JETIRGX06122.pdf