Research Progress of Brain Tumor Segmentation Based on Convolutional Neural Network

Zhiwei Li; Hui Cao; Feng Yang; Bin Cao

doi:10.3788/LOP202158.2400003

Journals >Laser & Optoelectronics Progress >Volume 58 >Issue 24 >Page 2400003 > Article

Laser & Optoelectronics Progress
Vol. 58, Issue 24, 2400003 (2021)

Research Progress of Brain Tumor Segmentation Based on Convolutional Neural Network

Zhiwei Li¹, Hui Cao¹, Feng Yang¹, and Bin Cao^2、*

Author Affiliations

¹School of Intelligence and Information Engineering, Shandong University of Traditional Chinese Medicine, Jinan, Shandong 250355, China

²Affiliated Hospital of Shandong University of Traditional Chinese Medicine, Jinan, Shandong 250014, China

show less

DOI: 10.3788/LOP202158.2400003 Cite this Article Set citation alerts

Zhiwei Li, Hui Cao, Feng Yang, Bin Cao. Research Progress of Brain Tumor Segmentation Based on Convolutional Neural Network[J]. Laser & Optoelectronics Progress, 2021, 58(24): 2400003 Copy Citation Text

show less

Fig. 1. A network structure of FCN for brain tumor segmentation^[26]

Download full size

Fig. 2. Structure of U-Net network^[35]

Download full size

Method	Category	Dataset	Dice_WT	Dice_TC	Dice_ET
Havaei17	2D	BraTS2015	0.79	0.58	0.69
Zikic14	2D	BraTS2013	0.84	0.74	0.69
Pereira16	2D	BraTS2015	0.79	0.65	0.75
Tseng17	2D to 3D	BraTS2015	0.85	0.68	0.69
Urban14	3D	BraTS2013	0.86	0.75	0.73
Kamnitsal16	3D	BraTS2015	0.90	0.76	0.73
Casamitjana16	3D	BraTS2015	0.92	0.84	0.77
Chen18	3D	BraTS2017	0.72	0.83	0.81
Qamar18	3D	BraTS2018	0.87	0.81	0.84
Feng Bowen20	3D	BraTS2018	0.90	0.73	0.71
Mlynarski18	2D and 3D	BraTS2017	0.92	0.88	0.85

Table 1. Dice coefficients of different dimensions CNN in brain tumor images segmentation

Method	Category	Dataset	Dice_WT	Dice_TC	Dice_ET
Long15	2D FCN	BraTS2018	0.83	0.73	0.63
Shen17	2D FCN	BraTS2013	0.86	0.73	0.73
Shen17’	2D FCN	BraTS2013	0.87	0.82	0.75
Zhao17	2D FCN	BraTS2013	0.81	0.65	0.60
Zhao17’	2D FCN	BraTS2017	0.88	0.76	0.76
Wang17	2D to 3D FCN	BraTS2017	0.79	0.87	0.77
Puch19	3D FCN	BraTS2018	0.90	0.80	0.75
Ronneberger15	2D U-Net	BraTS2018	0.84	0.75	0.65
Dong17	2D U-Net	BraTS2018	0.88	0.87	0.81
Shaikh17	2D U-Net	BraTS2017	0.83	0.65	0.65
Liu18	2D U-Net	BraTS2015	0.87	0.62	0.68
Isensee18	2D U-Net	BraTS2017	0.86	0.78	0.65
Kong18	2D U-Net	BraTS2017	0.92	0.80	0.76
Ai Lingmei20	2D U-Net	BraTS2018	0.91	0.80	0.79
Chu Jinghui19	2.5D U-Net	BraTS2018	0.91	0.85	0.81
Menze17	3D U-Net	BraTS2017	0.88	0.76	0.72
Beers17	3D U-Net	BraTS2017	0.88	0.73	0.73
Feng18	3D U-Net	BraTS2018	0.91	0.84	0.79
He Cheng’en20	3D U-Net	BraTS2017	0.90	0.80	0.77

Table 2. Dice coefficients of different dimensions FCN in brain tumor images segmentation

Method	Category	Dataset	Dice_WT	Dice_TC	Dice_ET
Randhawa16	DCNN	BraTS2016	0.87	0.75	0.71
Hussain17	DCNN	BraTS2013	0.80	0.67	0.85
Ben Naceur18	DCNN	BraTS2017	0.89	0.76	0.81

Table 3. Dice coefficients of partial DCNN models in brain tumor images segmentation

Zhiwei Li, Hui Cao, Feng Yang, Bin Cao. Research Progress of Brain Tumor Segmentation Based on Convolutional Neural Network[J]. Laser & Optoelectronics Progress, 2021, 58(24): 2400003

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information