Skip to content
SDGs Center UDINUS

Evidence record · UDINUS · SDGs

Comparative Analysis of CNN and Fine-Tuned InceptionV3 With and Without Transposition Layer for Chest X-ray Image Pneumonia Detection

Rektorat UDINUS 2025 Published 2026-10-30

Evidence record · UDINUS · SDGs

Unit
Rektorat UDINUS
Period
2025
Goals
Record
75320642-4411-45ae-adf7-1e6a61983c47
Published
2026-10-30

Universitas Dian Nuswantoro (UDINUS), through the participation of Widhiyanti, Erna Amalia; Shidik, Guruh Fajar; Nuraini, Aisyah; Ramadhani, Talitha Olga; and Kusuma, Edi Jaya, conducted research on pneumonia detection through chest X-ray imaging. The activity addressed the role of chest X-ray-based detection in enabling swift and accurate clinical assessments. The research focused on the complex challenge of identifying subtle spatial patterns in medical images. Within this scope, UDINUS examined the performance of various neural network models for addressing these image-analysis challenges.

The activity considered the application of transfer learning, model fine-tuning, and structural modifications as approaches for improving diagnostic performance. The findings indicate that combining these approaches can significantly enhance diagnostic performance. The research therefore concerned the use and comparison of neural network models in relation to pneumonia detection through chest X-ray images, including the examination of model performance under structural modifications. The source identifies future directions involving more diverse datasets from multiple clinical environments and a broader range of architectural innovations. These directions are intended to improve model adaptability and reliability.

In relation to the stated goal of advancing health-focused research, this UDINUS activity provides research-based consideration of neural network methods for chest X-ray pneumonia detection. Its institutional relevance rests on examining computational approaches for a clinical assessment task while identifying further methodological directions for model adaptability and reliability. The activity is documented as a journal publication.

Reference https://www.scopus.com/pages/publications/105031616773?origin=resultslist