CausalCervixNet: convolutional neural networks with causal insight (CICNN) in cervical cancer cell classification—leveraging deep learning models for enhanced diagnostic accuracy

Table 10 The performance analysis of the base models

Dataset	Models	Avg. Precision	Avg. Recall	Avg. F1-score	Avg. Accuracy	AUC	Time (sec)
SIPaKMeD	VGG16	0.978	0.978	0.978	97.78%	0.986	5200
	VGG19	0.963	0.962	0.962	96.18%	0.976	5500
	ResNet-50	0.949	0.948	0.948	94.82%	0.968	6000
	XceptionNet	0.760	0.656	0.649	65.64%	0.785	6400
ShUCSEIT	VGG16	0.939	0.937	0.938	93.64%	0.960	3200
	VGG19	0.929	0.929	0.929	92.73%	0.955	3400
	ResNet-50	0.965	0.964	0.964	96.36%	0.977	3800
	XceptionNet	0.857	0.857	0.857	85.45%	0.909	4200
Herlev	VGG16	0.616	0.628	0.605	60.21%	0.768	5200
	VGG19	0.659	0.635	0.643	59.14%	0.762	5400
	ResNet-50	0.825	0.826	0.824	81.18%	0.890	5800
	XceptionNet	0.434	0.431	0.387	40.32%	0.652	6200

ISSN: 1471-2407