A Deep Learning Framework for Leukemia Cancer Detection in Microscopic Blood Samples Using Squeeze and Excitation Learning

Table 5

Comparison with traditional deep learning model using ALLIDB1+ ALLIDB2 database.

Comparison of traditional deep learning model with the proposed model
Run#	Model	Loss value	Accuracy (%)	Precision (%)	Recall (%)	FScore (%)
1	CNN	0.92	95	95	94.5	94.5
2	CNN	1.67	92	92	92	91.5
3	CNN	1.73	91.4	90.5	92	91.5
Average	CNN	1.44	92.8	92.5	95.28	92.5
1	Proposed	0.16	97	97	97	97
2	Proposed	0.092	99	98.5	98.5	98.5
3	Proposed	0.099	99	99	99	99
Average	Proposed	0.117	98.3	98.16	98.16	98.43