Detection of Leukemia Using Inception-V3 and GoogLeNet
Yessi Jusman · 2023
Leukemia is a type of cancer affecting the blood that usually involves the white blood cells, which are strong infection fighters. The purpose of this research is to find out the classification results of the two models, namely, GoogLeNet and Inception-V3 in classifying two classes of leukemia images (normal images and abnormal images). This study uses a trained model called a pretrained model from ImageNet. The results of this study indicate that both training and testing, these two models (obtain very good results, where each has 100% accuracy). In addition, in the assessment of accuracy, an assessment is also carried out using the confusion matrix where the results of both also get equally good results. that is 100%. In addition to the assessment based on accuracy, the assessment was also carried out based on the time needed for the model to classify leukemia cells. In terms of time measurement, GoogLeNet is faster than Inception-V3. This study can be concluded that these two models can classify very well but in in terms of time-based assessment, GoogLeNet is faster than Inception-V3.