Perbandingan Kemampuan Klasifikasi Citra X-ray Paru-paru menggunakan Transfer Learning ResNet-50 dan VGG-16

Tasya  Berliani; Enggalwiguno Rahardja; Lina Septiana

doi:10.28932/jmh.v5i2.6116

Authors

Tasya Berliani Universitas Kristen Krida Wacana
Enggalwiguno Rahardja Universitas Kristen Krida Wacana
Lina Septiana Universitas Kristen Krida Wacana

DOI:

https://doi.org/10.28932/jmh.v5i2.6116

Keywords:

Covid-19, ; klasifikasi citra medis, x-ray dada, , ResNet-50, VGG-16

Abstract

Di masa pandemi Covid-19, foto rontgen menjadi umum digunakan untuk memeriksa pasien diduga Covid-19. Pada citra x-ray paru-paru yang terkena Covid-19 ditemukan adanya bercak putih atau flek. Namun, paru-paru yang memiliki flek ini tidak selalu disebabkan oleh Covid-19. Tujuan penelitian ini adalah untuk mengklasifikasikan beberapa jenis penyakit paru-paru dari citra x-ray, yaitu paru-paru dengan Covid-19, paru-paru dengan pneumonia, dan paru-paru yang memiliki flek dibandingkan dengan yang normal. Proses klasifikasi data pada penelitian ini dilakukan dengan membandingkan dua model yaitu CNN VGG-16 dan ResNet-50 dengan model yang telah dilatih sebelumnya. Metrik evaluasi yang digunakan dalam penelitian ini terdiri dari akurasi, presisi, sensitivitas, spesifisitas, skor F1, dan kecepatan waktu inferensi. Hasil menunjukkan bahwa VGG-16 lebih unggul dari ResNet-50 dalam hal kecepatan inferensi namun tidak dalam hal metrik evaluasi lainnya. Perubahan parameter juga menunjukkan hasil yang berbeda, epoch 200 adalah nilai optimal. Untuk mendapatkan hasil yang optimal diperlukan finetuning dengan menyesuaikan kondisi data yang digunakan. Sebagai simpulan, VGG-16 memiliki kemampuan klasifikasi yang lebih baik dibandingkan ResNet-50, namun perlu terus dikembangkan dengan memperbanyak data klinis yang aktual.

Downloads

Download data is not yet available.

References

Kim H, Goo JM, Kim TJ, Kim HY, Gu G, Gil B, et al. Effectiveness of radiologist training in improving reader agreement for Lung-RADS 4X categorization. Eur Radiol. 2021. [Cited 2022 August 2]. Available from: https://doi.org/10.1007/s00330-021-07990-y.

Barlow WE, Chen Chi, Patricia A. Carney, Stephen H. Taplin, Carl D’Orsi, Gary Cutter, et al. Accuracy of Screening Mammography Interpretation by Characteristics of Radiologists, JNCI: Journal of the National Cancer Institute. 2014; 96 (24): 1840–1850.

Muller D, Kramer F. MIScnn: a framework for medical image segmentation with convolutional neural networks and deep learning. BMC Medical Imaging. 2021;21(1). [Cited 2022 August 5]. Available from: https://doi.org/10.1186/s12880-020-00543-7.

Lecun Y, Bottou L, Bengio Y, Haffner P. Gradient-based learning applied to document recognition. Proceedings of the IEEE. 1998;86(11):2278-324. doi: 10.1109/5.726791.

Zeiler MD, Fergus R. Visualizing and Understanding Convolutional Networks. In: Fleet D, Pajdla T, Schiele B, Tuytelaars T, editors. Computer Vision – ECCV 2014. Lecture Notes in Computer Science, vol 8689. Springer, Cham. 2014.

Krizhevsky A, Sutskever I, Hinton GE. ImageNet classification with deep convolutional neural networks. Commun ACM. 2017;60(6):84-90. doi: 10.1145/3065386.

Victor Ikechukwu A, Murali S, Deepu R, Shivamurthy RC. ResNet-50 vs VGG-19 vs training from scratch: A comparative analysis of the segmentation and classification of Pneumonia from chest X-ray images. Global Transitions Proceedings. 2021;2(2). doi: 10.1016/j.glopro.2021.100036.

Munir K, Elahi H, Ayub A, Frezza F, Rizzi A. Cancer Diagnosis Using Deep Learning: A Bibliographic Review. Cancers. 2019;11(9):1235. doi: 10.3390/cancers11091235.

Rajpal S, Lakhyani N, Singh AK, Kohli R, Kumar N. Using handpicked features in conjunction with ResNet-50 for improved detection of COVID-19 from chest X-ray images. Chaos Solitons Fractals. 2021;145:110749. doi: 10.1016/j.chaos.2021.110749.

Ismael AM, Sengur A. Deep learning approaches for COVID-19 detection based on chest X-ray images. Expert Systems with Applications. 2021;164:114054. doi: 10.1016/j.eswa.2021.114054.

Chowdhury MEH, Rahman T, Khandakar A, Mazhar R, Kadir MA, Mahbub ZB, et al. Can AI help in screening Viral and COVID-19 pneumonia? IEEE Access. 2020;8:132665-132676. [Cited 2022 January 19]. Available from: https://www.kaggle.com/tawsifurrahman/covid19-radiography-database. doi: 10.1109/ACCESS.2020.3014131.

Joseph Paul Cohen JP, Morrison P, Dao L. COVID-19 image data collection. arXiv:2003.11597, 2020. [Cited 2022 January 19]. Available from: https://github.com/ieee8023/covid-chestxray-dataset, https://www.kaggle.com/pranavraikokte/covid19-image-dataset.

Dwivedi P. Understanding and Coding a ResNet in Keras-Towards Data Science. Medium. Towards Data Science; 2019 [Cited 2022 January 20]. Available from: https://towardsdatascience.com/understanding-and-coding-a-resnet-in-keras-446d7ff84d33.

Müller D, Soto-Rey I, Kramer F. Towards a guideline for evaluation metrics in medical image segmentation. BMC Res Notes. 2022;15:210. doi: 10.1186/s13104-022-06096-y.

Hicks SA, Strümke I, Thambawita V, Hammou M, Riegler MA, Halvorsen P, et al. On evaluation metrics for medical applications of artificial intelligence. medRxiv. 2021. [Cited 2022 April 18]. Available from: https://doi.org/10.1101/2021.04.07.21254975.

Ham HS, Lee HS, Chae JW, Cho HC, Cho HC. Improvement of Gastroscopy Classification Performance Through Image Augmentation Using a Gradient-Weighted Class Activation Map. IEEE Access. 2022;10:99361-99369. doi: 10.1109/ACCESS.2022.3207839.

Junayed MS, Islam MB, Jeny AA, Sadeghzadeh A, Biswas T, Shah AFMS. ScarNet: Development and Validation of a Novel Deep CNN Model for Acne Scar Classification With a New Dataset. IEEE Access. 2022;10:1245-1258. doi: 10.1109/ACCESS.2021.3138021.

Tang GS, Chow LS, Solihin MI, Ramli N, Gowdh NF, Rahmat K. Detection of COVID-19 Using Deep Convolutional Neural Network on Chest X-ray (CXR) Images. 2021 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), ON, Canada, 2021, pp. 1-6. doi: 10.1109/CCECE53047.2021.9569064.

Hicks SA, Strümke I, Thambawita V, Hammou M, Riegler MA, Halvorsen P, et al. On evaluation metrics for medical applications of artificial intelligence. Sci Rep. 2022;12:5979. doi: 10.1038/s41598-022-09954-8.

Sait T, Rehmsmeier M. The Precision-Recall Plot Is More Informative than the ROC Plot When Evaluating Binary Classifiers on Imbalanced Datasets. PLOS ONE. 2015;10(3):e0118432. [Cited 2022 April 18]. Available from: https://doi.org/10.1371/journal.pone.0118432.