Klasifikasi Subtipe Leukemia Limfoblastik Akut (LLA) pada Citra Mikroskopis Sel Darah Menggunakan Arsitektur EfficientNet-B3 dengan Dataset Seimbang

Agustina, Ni Putu Dina
Wijayakusuma, I Gusti Ngurah Lanang

Jurnal Locus Penelitian dan Pengabdian
Vol. 4 No. 6 (2025): JURNAL LOCUS: Penelitian & Pengabdian

Acute Lymphoblastic Leukemia (ALL) is one of the most common types of blood cancer that affects children and requires fast and accurate diagnosis. This study proposes a classification model for subtypes of acute lymphoblastic leukemia (ALL) based on microscopic blood cell images using the EfficientNet-B3 architecture. With a transfer learning approach and a balanced dataset, the model achieves a testing accuracy of 97.50% and an average F1-Score of 0.97. Overall, the macro average and weighted average values show consistent results, with precision and recall of 0.98 and an F1-Score of 0.97. This indicates that the model excels not only in one or two classes but demonstrates uniform performance across all classes, making it a robust classification tool for automatic leukemia diagnosis applications.