Timely Diagnosis of Acute Lymphoblastic Leukemia Using Artificial Intelligence-Oriented Deep Learning Methods

Table 1

Results from ResNet-50, VGG-16, and the proposed convolutional neural network for classification into two classes of healthy and leukemia.

Network	Training accuracy	Test accuracy	Train loss	Test loss
ResNet-50	0.9576	0.8163	0.1064	0.7592
VGG-16	0.9741	0.8462	0.0653	0.9236
Proposed convolution	0.8579	0.8210	0.3356	0.4517