Two-stage CNN-based framework for leukocytes classification

Siraj Khan, Muhammad Sajjad, José Escorcia-Gutierrez, Sami Dhahbi, Mohammad Hijji, Khan Muhammad

Abstract: Highlights•Leukocytes are key markers in diagnosing diseases like malaria and viral infections.•Manual leukocyte counting is difficult and error-prone due to their complex structure.•Our two-stage CNN framework significantly enhances leukocyte analysis accuracy.•The framework achieves over 99 % accuracy in segmentation and classification tasks.•It significantly enhances diagnostic performance, improving patient outcomes.