Cancer Cell Classification From Peripheral Blood Smear Data Using the YOLOv8 Architecture
Joao C. S. Nunes, José E. B. de S. Linhares, Miguel Postigo, Daniel Guzmán del Río, Angilberto Muniz Ferreira Sobrinho, Israel Gondres Torné · IEEE Access · 2025
The accurate classification of cancer cells in peripheral blood is essential for the diagnosis of leukemia and has traditionally been carried out by analyzing laboratory images. In this context, the use ofdeep learningtechniques facilitates decision-making and speeds up the early diagnosis of the disease, allowing preventive measures to be adopted for the patient. This study explores the application of the YOLOv8deep learningarchitecture for the classification of cancer cells in blood smear images, due to its ability to perform this task quickly and accurately. The models were trained on two datasets, ALL and C-NMC Leukemia, and evaluated using thetop-1 accuracy, top-5 accuracyandlossmetrics. The proposed approach achieved atop-1 accuracyof 99.982% andtop-5 accuracyof 100% on the validation and test subsets of the ALL dataset, with a finallossof 0.02952. For the C-NMC Leukemiadataset, the model obtained 66.897% and 89.612%top-1 accuracyin the validation and test subsets, respectively, while maintaining 100%top-5 accuracyin both, with alossof 0.18289. These results demonstrate the effectiveness of YOLOv8 in classifying cancer cells, especially in the ALL set. However, future strategies such as expanding the data set and fine-tuning the hyperparameters could contribute to better generalizing the model to different data distributions.