Region-Based Segmentation and Classification for Ovarian Cancer Detection Using Convolution Neural Network

Table 1

Comparative analysis for proposed and existing techniques.

Study references	Techniques	Binary or multiclass	Modules	Accuracy	Other parameters
[9]	IABC-EMBOT, IHM-FFNN, PSO-RM, ABCO-BCD AND DNN-BCD	Binary	Negative, positive	0.975	—
[10]	FNN, ANFIS, ANNFIS	Binary	Negative, positive	0.92	Precision 0.944, recall 0.944
[11]	Deep type, state of the art	Multiclass	Normal, luminal A, B, basal and HER2	—	—
[12]	BPNN	Binary	Mutant as well as non-mutant sequences	0.998	Sensitivity = 0 Specificity = 0
[13]	CNN	Multiclass	—	0.956	—
[14]	DA	Multiclass	—	0.95	—
[15]	DNN + attention method	Multiclass	—	0.87	—
[16]	GCN	Multiclass	—	0.919	AUC = 0.84
[17]	DNN + SVM	Multiclass	Binary, miotic/non	0.83	F-score = 0.556 Accuracy = 0.8319
[18]	DNN	Multiclass	4 classes	0.97 by SVM	AUC = 0.82, accuracy = 0.8682
[19]	DL + ML	Multiclass	Auxiliary lymph node status, binary, cancer or not	0.84	Accuracy:0.98, AUC = 0.93
[21]	DL + ML	Multiclass	Binary, cancer or not	0.84	AUC:0.84