Assessing the impact of Eight EfficientNetB (0- 7) Models for Leukemia Categorization

Leukemia is the foremost common sort of childhood cancer and accounts for roughly 25% of the pediatric cancers. These cells have been sectioned from minuscule pictures and are representative of pictures within the real-world since they contain a few recoloring clamor and light mistakes, in spite of the fact that these mistakes have generally been settled through innovative strategies. The errand of recognizing juvenile leukemic impacts from ordinary cells beneath the magnifying lens is challenging due to morphological similitude and in this way the ground truth names were clarified by oncologists to cure blood related diseases. In this research, comparison is performed on various EfficientNet models for classification of Leukemia infected blood cells. Main focus is to tackle one of the major childhood cancer types by creating an accurate EfficientNet model with precision value of 98.08% to classify normal from abnormal cell images.