Melola Discovery

← Melola Discovery

Jurnal SINTA 2

Analisis Performa Pre-Trained Model Convolutional Neural Network dalam Mendeteksi Penyakit Tuberkulosis

Jurnal Teknologi Informasi dan Ilmu Komputer (JTIIK) · Vol. 8 No. 4 (2021) · pp. 805–814

DOI
10.25126/jtiik.2021844441
ISSN
2528-6579, 2355-7699
Published
2021-07-22
Publisher
Fakultas Ilmu Komputer, Universitas Brawijaya
Language
id

Abstract

Tuberkulosis (TB) merupakan salah satu penyakit berbahaya yang dapat menular lewat udara dan sering menyebabkan kematian apabila tidak cepat ditangani. Penyakit TB bisa disembuhkan dengan deteksi dini sehingga penderita dapat segera mendapatkan pengobatan yang tepat. Metode Convolutional Neural Network (CNN) digunakan untuk mendeteksi penyakit TB melalui foto rontgen dada. Penelitian ini bertujuan untuk menentukan model CNN yang mampu menghasilkan performa paling baik dalam mendeteksi penyakit TB. Pengujian dilakukan dengan menggunakan lima pre-trained model yang telah disediakan oleh Keras yaitu ResNet50, DenseNet121, MobileNet, Xception, InceptionV3, dan InceptionResNetV2. Perbedaan ukuran gambar yag digunakan pada saat pelatihan dan pengujian juga akan dianalisis pengaruhnya terhadap nilai akurasi yang dihasilkan dan waktu komputasinya. Hasil pengujian menunjukkan bahwa model DenseNet121 mampu menghasilkan nilai akurasi tertinggi dalam mendeteksi penyakit TB, yaitu 91,57%. Sedangkan model MobileNet merupakan model dengan waktu komputasi tercepat untuk semua ukuran gambar yang diuji. Semakin besar ukuran citra maka semakin tinggi nilai akurasinya, namun di sisi lain waktu komputasi juga akan semakin lama. Abstract Tuberculosis (TB) is one of the dangerous disease that can be transmitted through the air and often causes death if not treated quickly. This illness can be cured with early detection, so that sufferers can immediately get the right treatment. The Convolutional Neural Network (CNN) method is used to detect TB disease through chest X-rays. This study aims to determine which CNN model is able to produce the best performance in detecting TB disease. Testing was carried out using five pre-trained models provided by Keras namely ResNet50, DenseNet121, MobileNet, Xception, InceptionV3, and InceptionResNetV2. The difference in image size used during training and testing will also be analyzed for its effect on the resulting accuracy value and its computation time. The test results showed that the DenseNet121 model was able to produce the highest accuracy value in detecting TB disease, namely 91.57%. Meanwhile, the MobileNet model is the model with the fastest computation time for all image sizes tested. The bigger the image size, the higher the accuracy value, but on the other hand the computation time will also be longer.

Collaboration network of the authors

40 authors · 80 co-authorship links · showing the 40 most connected

Hover or tap a name to highlight its co-authors. Double-click to open the author's page.

  • University of Brawijaya
  • Affiliation unknown
  • Different organisations
  • Same organisation
  • Affiliation unknown
  • Thicker line = more joint articles · bigger circle = more articles
Show as table
Co-authorOrganisationJoint articles
Fajar PradanaUniversity of Brawijaya10
Retno Indah RokhmawatiUniversity of Brawijaya3
Muhammad WafiUniversity of Brawijaya3
Herman TolleUniversity of Brawijaya3
Aisyah AwalinaUniversity of Brawijaya3
Bayu PriyambadhaUniversity of Brawijaya2
Satrio Agung WicaksonoUniversity of Brawijaya2
Achmad BasukiUniversity of Brawijaya2
Imam CholissodinUniversity of Brawijaya2
Novanto YudistiraUniversity of Brawijaya2
Hurriyatul FitriyahUniversity of Brawijaya2
Rizal MaulanaUniversity of Brawijaya2
Randy Cahya WihandikaUniversity of Brawijaya2
Muhammad Ali FauziUniversity of Brawijaya2
Dahnial SyauqyUniversity of Brawijaya2
Firnanda Al Islama Achyunda PutraUniversity of Brawijaya1
Wayan Firdaus MahmudyUniversity of Brawijaya1
Joseph Ananda SugihdharmaUniversity of Brawijaya1
Rizal Setya PerdanaUniversity of Brawijaya1
Nathanael Victor DarenohUniversity of Brawijaya1
Aditya RachmadiUniversity of Brawijaya1
Ihwanudien Hasan RobbaniUniversity of Brawijaya1
Enny TrisnawatiUniversity of Brawijaya1
Rakhmadina NoviyantiUniversity of Brawijaya1
Afrizal RivaldiUniversity of Brawijaya1
Waskitha WijayaUniversity of Brawijaya1
Aulia Akhrian SyahidiUniversity of Brawijaya1
Fajerin BiabdillahUniversity of Brawijaya1
Nanang Yudi SetiawanUniversity of Brawijaya1
Iqbal Taufiq Ahmad NurUniversity of Brawijaya1
Indra SyahputraUniversity of Brawijaya1
Bayu SutawijayaUniversity of Brawijaya1
Mochammad Dearifaldi Al IkhsanUniversity of Brawijaya1
Sulandri SulandriUniversity of Brawijaya1
Rona SalsabilaUniversity of Brawijaya1
Budi Darma SetiawanUniversity of Brawijaya1
Muhammad Nugraha Delta RevanzaAffiliation unknown1

Affiliations come from OpenAlex; journal OAI-PMH metadata carries none. Authors without an OpenAlex ID are matched by name, which can merge or split people.

Metadata source: OAI-PMH (metadata jurnal) https://jtiik.ub.ac.id/index.php/jtiik/en/oai, first harvested 27 Okt 2026, last seen 27 Okt 2026.
Original metadata (oai_dc) 
<oai_dc:dc xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/  http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
	<dc:title xml:lang="id">Analisis Performa Pre-Trained Model Convolutional Neural Network dalam Mendeteksi Penyakit Tuberkulosis</dc:title>
	<dc:creator xml:lang="id">Rochmawanti, Ovy</dc:creator>
	<dc:creator xml:lang="id">Utaminingrum, Fitri</dc:creator>
	<dc:creator xml:lang="id">Bachtiar, Fitra A.</dc:creator>
	<dc:description xml:lang="id">Tuberkulosis (TB) merupakan salah satu penyakit berbahaya yang dapat menular lewat udara dan sering menyebabkan kematian apabila tidak cepat ditangani. Penyakit TB bisa disembuhkan dengan deteksi dini sehingga penderita dapat segera mendapatkan pengobatan yang tepat. Metode Convolutional Neural Network (CNN) digunakan untuk mendeteksi penyakit TB melalui foto rontgen dada. Penelitian ini bertujuan untuk menentukan model CNN yang mampu menghasilkan performa paling baik dalam mendeteksi penyakit TB. Pengujian dilakukan dengan menggunakan lima pre-trained model yang telah disediakan oleh Keras yaitu ResNet50, DenseNet121, MobileNet, Xception, InceptionV3, dan InceptionResNetV2. Perbedaan ukuran gambar yag digunakan pada saat pelatihan dan pengujian juga akan dianalisis pengaruhnya terhadap nilai akurasi yang dihasilkan dan waktu komputasinya. Hasil pengujian menunjukkan bahwa model DenseNet121 mampu menghasilkan nilai akurasi tertinggi dalam mendeteksi penyakit TB, yaitu 91,57%. Sedangkan model MobileNet merupakan model dengan waktu komputasi tercepat untuk semua ukuran gambar yang diuji. Semakin besar ukuran citra maka semakin tinggi nilai akurasinya, namun di sisi lain waktu komputasi juga akan semakin lama.  Abstract Tuberculosis (TB) is one of the dangerous disease that can be transmitted through the air and often causes death if not treated quickly. This illness can be cured with early detection, so that sufferers can immediately get the right treatment. The Convolutional Neural Network (CNN) method is used to detect TB disease through chest X-rays. This study aims to determine which CNN model is able to produce the best performance in detecting TB disease. Testing was carried out using five pre-trained models provided by Keras namely ResNet50, DenseNet121, MobileNet, Xception, InceptionV3, and InceptionResNetV2. The difference in image size used during training and testing will also be analyzed for its effect on the resulting accuracy value and its computation time. The test results showed that the DenseNet121 model was able to produce the highest accuracy value in detecting TB disease, namely 91.57%. Meanwhile, the MobileNet model is the model with the fastest computation time for all image sizes tested. The bigger the image size, the higher the accuracy value, but on the other hand the computation time will also be longer.</dc:description>
	<dc:publisher xml:lang="en">Fakultas Ilmu Komputer, Universitas Brawijaya</dc:publisher>
	<dc:date>2021-07-22</dc:date>
	<dc:type>info:eu-repo/semantics/article</dc:type>
	<dc:type>info:eu-repo/semantics/publishedVersion</dc:type>
	<dc:format>application/pdf</dc:format>
	<dc:identifier>https://jtiik.ub.ac.id/index.php/jtiik/article/view/4441</dc:identifier>
	<dc:identifier>10.25126/jtiik.2021844441</dc:identifier>
	<dc:source xml:lang="id">Jurnal Teknologi Informasi dan Ilmu Komputer; Vol 8 No 4: Agustus 2021; 805-814</dc:source>
	<dc:source xml:lang="en">Jurnal Teknologi Informasi dan Ilmu Komputer; Vol. 8 No. 4: Agustus 2021; 805-814</dc:source>
	<dc:source>2528-6579</dc:source>
	<dc:source>2355-7699</dc:source>
	<dc:source>10.25126/jtiik.202184</dc:source>
	<dc:language>id</dc:language>
	<dc:relation>https://jtiik.ub.ac.id/index.php/jtiik/article/view/4441/pdf</dc:relation>
	<dc:rights xml:lang="en">Copyright (c) 2021 Jurnal Teknologi Informasi dan Ilmu Komputer</dc:rights>
</oai_dc:dc>