Komparasi Supervised dan Self-Supervised Learning untuk Klasifikasi Sel Darah pada Skenario Keterbatasan Data Berlabel

Misella Mambu; Oktavian Abraham Lantang; Salaki Reynaldo Joshua

doi:10.30606/rjti.v5i2.4555

Authors

Misella Mambu Universitas Sam Ratulangi
Oktavian Abraham Lantang Universitas Sam Ratulangi
Salaki Reynaldo Joshua Universitas Sam Ratulangi

DOI:

https://doi.org/10.30606/rjti.v5i2.4555

Keywords:

Klasifikasi Sel Darah, Self-Supervised Learning, SimCLR, BYOL, ResNet-50, Grad-CAM.

Abstract

The development of blood cell image classification models is often hindered by the scarcity of labeled data. This study evaluates the effectiveness of supervised baseline models, supervised fine-tuning, and self-supervised learning (SSL) based on SimCLR and BYOL using the ResNet-50 and DenseNet-121 architectures. Data-limited simulation was performed using 10% labeled data for the classification stage and 90% unlabeled data for SSL pre-training on 17,092 images from 8 morphological classes. The results show that SSL achieves competitive performance, with the highest accuracy on ResNet-50 at 95.31% (SimCLR) and 94.27% (BYOL), and on DenseNet-121 at 93.75% (SimCLR) and 89.58% (BYOL). The SSL approach also demonstrated robustness against overfitting, good robustness against class imbalance, and Grad-CAM visualizations that align with medical expert assessments on ResNet-50. In conclusion, SSL effectively optimizes unlabeled data to improve the performance of classification models in scenarios with limited data annotation.

Downloads

Download data is not yet available.

References

S. B. Boly and M. Akbar, “Segmentasi Citra Sel Darah Menggunakan Convolutional Neural Network,” vol. 6, no. 2, pp. 390–398, 2024, [Online]. Available: https://www.kaggle.com/jeetblahiri

E. Retnoningsih and R. Pramudita, “Mengenal Machine Learning Dengan Teknik Supervised dan Unsupervised Learning Menggunakan Python,” BINA INSANI ICT JOURNAL, vol. 7, no. 2, pp. 156–165, 2020, [Online]. Available: https://www.python.org/

Ade Ryan Pratama, Farmin Wabula, Haekal Ilmandry, Maria Laura Isabela, Mugi Raharjo, and Ronald Sianipar, “Literature Review The Impact of Machine Learning in Modern Industries,” Nian Tana Sikka : Jurnal ilmiah Mahasiswa, vol. 3, no. 1, pp. 177–182, Jan. 2025, doi: 10.59603/niantanasikka.v3i1.680.

S. R. Joshua, S. Shin, J. H. Lee, and S. K. Kim, “Health to Eat: A Smart Plate with Food Recognition, Classification, and Weight Measurement for Type-2 Diabetic Mellitus Patients’ Nutrition Control,” Sensors, vol. 23, no. 3, Feb. 2023, doi: 10.3390/s23031656.

A. Chowdhury, J. Rosenthal, J. Waring, and R. Umeton, “Applying self-supervised learning to medicine: Review of the state of the art and medical implementations,” Sep. 01, 2021, MDPI. doi: 10.3390/informatics8030059.

N. Jia et al., “A fine-grained image classification algorithm based on self-supervised learning and multi-feature fusion of blood cells,” Sci. Rep., vol. 14, no. 1, p. 22964, Dec. 2024, doi: 10.1038/s41598-024-74753-2.

I. Putu Agus, K. Hidjah, N. Sulistianingsih, and G. Hendro, “Implementasi Arsitektur Deep Convolutional Neural Network (CNN) dengan Transfer Learning untuk Klasifikasi Penyakit Kulit,” JTIM 2025, vol. 7, no. 3, pp. 461–477, 2025, doi: 10.35746/jtim.v7i3.734.

Y. I. Salsabila, H. Mustofa, and M. A. Ulinuha, “KLASIFIKASI CITRA TUMOR OTAK MENGGUNAKAN TEKNIK TRANSFER LEARNING PADA ARSITEKTUR RESNET-50,” Journal of Information System, Informatics and Computing, vol. 9, no. 1, p. 128, Jun. 2025, doi: 10.52362/jisicom.v9i1.1925.

M. Putra et al., “Image Analysis of Skin Diseases Using DenseNet-121 Architecture,” Journal Medical Informatics Technology, pp. 73–78, Jun. 2025, doi: 10.37034/medinftech.v3i2.99.

O. Soedradjat, R. Magdalena, and N. Pratiwi, “DETEKSI GANGGUAN PARU-PARU BERBASIS CITRA X-RAY MENGGUNAKAN DEEP LEARNING,” 2022.

S. C. Huang, A. Pareek, M. Jensen, M. P. Lungren, S. Yeung, and A. S. Chaudhari, “Self-supervised learning for medical image classification: a systematic review and implementation guidelines,” Dec. 01, 2023, Nature Research. doi: 10.1038/s41746-023-00811-0.

C. Li, F. Li, L. Zhang, A. Yang, Z. Hu, and M. He, “Intrusion Detection for Industrial Control Systems Based on Improved Contrastive Learning SimCLR,” Applied Sciences (Switzerland), vol. 13, no. 16, Aug. 2023, doi: 10.3390/app13169227.

C. Agastya, S. Ghebremusse, I. Anderson, C. Reed, H. Vahabi, and A. Todeschini, “Self-supervised Contrastive Learning for Irrigation Detection in Satellite Imagery,” Aug. 2021, [Online]. Available: http://arxiv.org/abs/2108.05484

J.-B. Grill et al., “Bootstrap your own latent: A new approach to self-supervised Learning,” Sep. 2020, [Online]. Available: http://arxiv.org/abs/2006.07733

D. Zeng, Y. Wu, X. Hu, X. Xu, J. Hu, and Y. Shi, “Additional Positive Enables Better Representation Learning for Medical Images,” May 2023, [Online]. Available: http://arxiv.org/abs/2306.00112

H. Moujahid et al., “Combining cnn and grad-cam for covid-19 disease prediction and visual explanation,” Intelligent Automation and Soft Computing, vol. 32, no. 2, pp. 723–745, 2022, doi: 10.32604/iasc.2022.022179.

A. Verdaus Saputra and W. Setiawan, “KLASIFIKASI CITRA HISTOPATOLOGI KANKER PARU-PARU MENGGUNAKAN RESNET50 DENGAN INTERPRETASI GRAD-CAM,” 2026.

M. A. Hasanah, S. Soim, and A. S. Handayani, “Implementasi CRISP-DM Model Menggunakan Metode Decision Tree dengan Algoritma CART untuk Prediksi Curah Hujan Berpotensi Banjir,” 2021. [Online]. Available: http://jurnal.polibatam.ac.id/index.php/JAIC

O. Lantang, G. Terdik, A. Hajdu, and A. Tiba, “Comparison of single and ensemble-based convolutional neural networks for cancerous image classification,” Annales Mathematicae et Informaticae, vol. 54, pp. 45–56, 2021, doi: 10.33039/ami.2021.03.013.

S. R. Joshua, K. Y. Palilingan, S. P. Lengkong, and S. Park, “Deep Learning-Driven Solar Fault Detection in Solar–Hydrogen AIoT Systems: Implementing CNN VGG16, ResNet-50, DenseNet121, and EfficientNetB0 in a University-Based Framework,” Hydrogen, vol. 7, no. 1, p. 1, Dec. 2025, doi: 10.3390/hydrogen7010001.