A Mobile Deep Learning Model on Covid-19 CT-Scan Classification

(1) * Prastyo Eko Susanto Mail (Indonesian Islamic University, Indonesia)
(2) Arrie Kurniawardhan Mail (Department of Informatics, Universitas Islam Indonesia, Yogyakarta, Indonesia)
(3) Dhomas Hatta Fudholi Mail (Department of Informatics, Universitas Islam Indonesia, Yogyakarta, Indonesia)
(4) Ridho Rahmadi Mail (Department of Informatics, Universitas Islam Indonesia, Yogyakarta, Indonesia)
*corresponding author


COVID-19 pandemic is currently happening in the world. Previous studies have been done to diagnose COVID-19 by identifying CT-scan images through the development of the novel Joint Classification and Segmentation System models that work in real-time. In this study, the author focuses on a different motivation and innovation focused on the development of mobile deep learning. Mobile Net, a deep learning model as a method for classifying the disease COVID-19, is used as the base model. It has a good level of efficiency and reliability to be implemented on devices that have small memory and CPU specifications, such as mobile phones. The used data in this study is a CT-scan image of the lungs with a horizontal slice that has been classified as positive or negative for COVID-19. To give a broader analysis, the author compares and evaluates the model against other architectures, such as MobileNetV3 Large, MobileNetV3 Small, MobilenetV2, ResNet101, and EfficientNetB0. In terms of the developed mobile architecture model, the classification of COVID-19 using MobileNetV2 obtained the best result with 0.81 accuracy.


ct scan; covid-19; deep learning; MobileNet; classification




Article metrics

10.29099/ijair.v6i1.257 Abstract views : 819 | PDF views : 281




Full Text




W. H. Organization, “WHO Coronavirus (COVID-19) Dashboard,†World Health Organization, 2021. https://covid19.who.int/ (accessed Aug. 04, 2021).

S. T. P. COVID-19, “Peta Sebaran,†26 Agustus 2020, 2020. https://covid19.go.id/peta-sebaran (accessed Aug. 26, 2020).

Y.-H. Wu et al., “JCS: An Explainable COVID-19 Diagnosis System by Joint Classification and Segmentation,†pp. 1–11, 2020, [Online]. Available: http://arxiv.org/abs/2004.07054.

Y. Wang et al., “A survey on deploying mobile deep learning applications : A systemic and technical perspective,†Digit. Commun. Networks, no. June 2020, 2021, doi: 10.1016/j.dcan.2021.06.001.

A. Santoso and G. Ariyanto, “Implementasi Deep Learning Berbasis Keras Untuk Pengenalan Wajah,†Emit. J. Tek. Elektro, vol. 18, no. 01, pp. 15–21, 2018, doi: 10.23917/emitor.v18i01.6235.

C. Fan, Z. Zhang, and D. J. Crandall, “Deepdiary: Lifelogging image captioning and summarization,†J. Vis. Commun. Image Represent., vol. 55, no. March 2017, pp. 40–55, 2018, doi: 10.1016/j.jvcir.2018.05.008.

X. Yang, X. He, J. Zhao, Y. Zhang, S. Zhang, and P. Xie, “COVID-CT-Dataset: A CT Image Dataset about COVID-19,†Arxiv.Org, vol. XX, no. Xx, pp. 1–14, 2020, [Online]. Available: https://www.medrxiv.org/.

X. He et al., “Sample-Efficient Deep Learning for COVID-19 Diagnosis Based on CT Scans,†medRxiv, vol. XX, no. Xx, p. 2020.04.13.20063941, 2020, doi: 10.1101/2020.04.13.20063941.

J. P. Cohen, P. Morrison, and L. Dao, “COVID-19 Image Data Collection,†2020, [Online]. Available: http://arxiv.org/abs/2003.11597.

M. Ilyas, H. Rehman, and A. Nait-ali, “Detection of Covid-19 From Chest X-ray Images Using Artificial Intelligence: An Early Review,†pp. 1–8, 2020, [Online]. Available: http://arxiv.org/abs/2004.05436.

“Covid-19 database,†https://www.sirm.org/en/, 2020. https://www.sirm.org/en/category/articles/covid-19-database/ (accessed Nov. 21, 2020).

“Covid-19,†https://radiopaedia.org/, 2020. https://radiopaedia.org/search?utf8=✓&q=covid-19&scope=all〈=us (accessed Nov. 21, 2020).

“Eurorad,†https://www.eurorad.org/, 2020. https://www.eurorad.org/advanced-search?search=covid-19&sort_by=published_at&sort_order=ASC&page=1&filter%5B0%5D=section%3A40 (accessed Nov. 21, 2020).

J. Solawetz, “Why and How to Implement Random Rotate Data Augmentation,†https://blog.roboflow.com/, 2020. https://blog.roboflow.com/why-and-how-to-implement-random-rotate-data-augmentation/ (accessed Nov. 27, 2021).

J. Wang and L. Perez, “The Effectiveness of Data Augmentation in Image Classification using Deep Learning,†2017.

A. Howard et al., “Searching for mobileNetV3,†Proc. IEEE Int. Conf. Comput. Vis., vol. 2019-Octob, pp. 1314–1324, 2019, doi: 10.1109/ICCV.2019.00140.

K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,†Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 2016-Decem, pp. 770–778, 2016, doi: 10.1109/CVPR.2016.90.

M. Tan and Q. V. Le, “EfficientNet: Rethinking model scaling for convolutional neural networks,†36th Int. Conf. Mach. Learn. ICML 2019, vol. 2019-June, pp. 10691–10700, 2019.

R. Siswosudarmo, “Tes diagnostik (Diagnostic test),†J. Metodol. Penelit., p. 12, 2017, [Online]. Available: http://obgin-ugm.com/wp-content/uploads/2017/09/HRS-Kuliah-Tes-Diagnostik.pdf.

C. Goutte and E. Gaussier, “Ch10_Witnesses[8463].Pdf,†no. April, 2005, doi: 10.1007/978-3-540-31865-1.

B. Ramsay and E. Van Der Knaap, “Confusion Matrix-based Feature Selection Sofia Visa,†2018.

Kuliahkomputer, “Pengujian Dengan Confusion Matrix.†http://www.kuliahkomputer.com/2018/07/pengujian-dengan-confusion-matrix.html#:~:text=Confusion matrix adalah suatu metode,sebagai representasi hasil proses klasifikasi. (accessed Nov. 28, 2021).

S. Setiawan, “Membicarakan Precision, Recall, dan F1-Score,†https://medium.com, 2020. https://stevkarta.medium.com/membicarakan-precision-recall-dan-f1-score-e96d81910354 (accessed Dec. 01, 2021).

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.


The International Journal of Artificial Intelligence Research

Organized by: Departemen Teknik Informatika
Published by: STMIK Dharma Wacana
Jl. Kenanga No.03 Mulyojati 16C Metro Barat Kota Metro Lampung
phone. +62725-7850671
Fax. +62725-7850671

Email: jurnal.ijair@gmail.com

View IJAIR Statcounter

Creative Commons License
This work is licensed under  Creative Commons Attribution-ShareAlike 4.0 International License.