Skip to content
SDGs Center UDINUS

Evidence record · UDINUS · SDGs

Ensemble CNN Models for Accurate Pneumonia Classification in Chest X-Ray Images

Rektorat UDINUS 2024 Published 2026-10-30

Evidence record · UDINUS · SDGs

Unit
Rektorat UDINUS
Period
2024
Goals
Record
f3e02687-600f-427c-8783-a303e1b804d4
Published
2026-10-30

Universitas Dian Nuswantoro conducted a health research activity concerning pneumonia classification in chest X-ray images. The activity involved UDINUS researchers Widiarto, Muhamad; Andono, Pulung Nurtantio; Supriyanto, Catur; Soeleman, Moch Arief; and Wijaya, Adi. The source describes pneumonia as a life-threatening respiratory disease, particularly in children and the elderly. It also explains that early detection through chest X-rays enables faster and more accurate treatment.

The activity used Convolutional Neural Networks (CNN) to classify lung X-ray images as either normal or pneumonia-infected. UDINUS compared two CNN models, InceptionResNetV2 and MobileNetV2, and introduced an ensemble approach that combined both models. The comparison addressed classification performance in medical image analysis. The source does not specify the dataset size or composition, the research setting, the study dates, or detailed implementation and evaluation procedures.

The ensemble model achieved a classification accuracy of 92.47%. In comparison, InceptionResNetV2 achieved 89.58% and MobileNetV2 achieved 90.38%. The source further states that the ensemble model enhanced precision, recall, and F1 score. These findings suggest that combining the two CNN models was an effective strategy for improving accuracy in pneumonia classification through medical imaging.

In relation to the stated goal concerning health, the activity addresses health research and the advancement of technology for disease diagnosis through medical imaging. It frames UDINUS institutional activity through research on a CNN-based method for classifying pneumonia in chest X-ray images. Further information is needed about the research context, data, procedures, and implementation.

Reference https://www.scopus.com/pages/publications/86000029517?origin=resultslist