Claim Missing Document
Check
Articles

Found 11 Documents
Search

Transfer Learning Analysis VGG16 For the Detection of Tuberculosis Erna Dwi Astuti; Muslim Hidayat
Jurnal Teknik Elektro dan Komputer TRIAC Vol 12, No 1 (2025): Mei 2025
Publisher : Jurusan Teknik Elektro Universitas Trunojoyo Madura

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21107/triac.v12i1.28672

Abstract

- Indonesia is still one of the countries with the highest growth of TB disease in the world. TB is an infectious disease that can cause severe lung damage, even death. TB is a critical case to be detected early so that patients immediately get the proper treatment. The challenge is the difficulty in diagnosing symptoms that are not specific and similar to other diseases. Therefore, further research is needed to find a faster, more accurate, affordable TB detection method. VGG16 is one of the Convolutional Neural Network (CNN) architectures that has the characteristic of recognizing delicate patterns of chest X-ray images of TB patients. Transfer learning on VGG16 can increase the accuracy of detecting TB disease even though it uses a small amount of training data. The trial results show that the VGG16 transfer learning technique can produce better performance with an accuracy of 94%. The accuracy value can be used to benchmark that the VGG16 transfer learning technique is proven effective in detecting TB disease