BMC Medical Imaging

Table 1 Study and comparison of the proposed method and various deep learning and machine learning methods

From: A dual autoencoder and singular value decomposition based feature optimization for the segmentation of brain tumor from MRI images

Method	Patch size	Type	DSC	PPV	Sensitivity
Proposed	16 × 64	Meningioma	0.84	0.88	0.89
	16 × 32		0.83	0.85	0.87
	16 × 16		0.81	0.82	0.83
	16 × 64	Glioma	0.82	0.84	0.86
	16 × 32		0.81	0.825	0.85
	16 × 16		0.78	0.80	0.81
				0.85	0.86
CNN [27]	16 × 16	Glioma	0.88	0.89	0.92
SVM [5]		Glioma	0.80	0.81	0.82
KNN, SVM [18]		Various	0.81	0.815	0.83
ANN [23]		Various	0.83	0.82	0.84
RescueNet [32]		Gliomas	0.94	0.85	0.88
3D-GAN [33]		Gliomas	0.87	0.88	0.88

Back to article page

ISSN: 1471-2342

Contact us

General enquiries: journalsubmissions@springernature.com