Computational and Mathematical Methods in Medicine Glioma Brain Tumor Detection and Classification Using Convolutional Neural Network

Table 2

BraTS dataset-based classification performance of CD_BLNL model.

Manifold	Mode of operation	Precision	Recall	Accuracy	Stability loss	score
I	Training	96.1	95.1	97.1	97.1	95.1
	Testing	95.2	95.6	97.3	96.9	95.2
II	Training	94.8	96.1	97.01	96.8	96.1
	Testing	95.01	95.3	97.03	96.9	95.3
III	Training	96.1	95.3	96.89	97.1	95.7
	Testing	94.9	95.1	97.4	97.1	95.9
IV	Training	95.3	96.1	97.2	96.9	95.9
	Testing	94.8	95.6	97.45	96.09	95.3
V	Training	95.6	96.2	97.8	97.01	96.1
	Testing	94.7	95.7	96.9	96.9	96.3
VI	Training	96.1	95.1	97.1	97.1	95.1
	Testing	95.2	95.6	97.2	96.9	95.2
VII	Training	94.8	96.11	97.08	96.8	96.1
	Testing	95.07	95.3	97.07	96.9	95.3
VIII	Training	96.1	95.32	96.82	97.1	95.7
	Testing	94.9	95.12	97.41	97.1	95.9
IX	Training	95.3	96.11	97.3	96.9	95.9
	Testing	94.79	95.63	97.44	96.09	95.3
X	Training	95.59	96.28	97.85	97.01	96.1
	Testing	94.74	95.74	96.89	96.9	96.8
Average (%)		95.255	95.6205	97.212	96.88	95.715