• Laser & Optoelectronics Progress
  • Vol. 57, Issue 22, 221004 (2020)
Lingmei Ai* and Kangzhen Shi*
Author Affiliations
  • College of Computer Science, Shaanxi Normal University, Xi'an, Shaanxi 710119, China
  • show less
    DOI: 10.3788/LOP57.221004 Cite this Article Set citation alerts
    Lingmei Ai, Kangzhen Shi. Low-Grade Gliomas MR Image Segmentation Based on Conditional Generative Adversarial Networks[J]. Laser & Optoelectronics Progress, 2020, 57(22): 221004 Copy Citation Text show less
    Overall framework of segmentation LGG
    Fig. 1. Overall framework of segmentation LGG
    Structure of CGAN. (a) Generator network; (b) discriminator network
    Fig. 2. Structure of CGAN. (a) Generator network; (b) discriminator network
    U-Net model of segmentation LGG
    Fig. 3. U-Net model of segmentation LGG
    MR images and segmentation mask of LGG. (a) Pre-contrast; (b) FLAIR; (c) post-contrast; (d) segmentation mask
    Fig. 4. MR images and segmentation mask of LGG. (a) Pre-contrast; (b) FLAIR; (c) post-contrast; (d) segmentation mask
    LGG images generated by CGAN. (a) Pre-contrast; (b) FLAIR; (c) post-contrast; (d) segmentation mask
    Fig. 5. LGG images generated by CGAN. (a) Pre-contrast; (b) FLAIR; (c) post-contrast; (d) segmentation mask
    Training process on different datasets. (a) Dataset1 and dataset2; (b) dataset3 and dataset4
    Fig. 6. Training process on different datasets. (a) Dataset1 and dataset2; (b) dataset3 and dataset4
    Segmentation results of different methods. (a) Dataset1; (b) dataset2; (c) dataset3; (d) dataset4
    Fig. 7. Segmentation results of different methods. (a) Dataset1; (b) dataset2; (c) dataset3; (d) dataset4
    DatasetSensitivity /%Specificity /%Accuracy /%Dice /%Jaccard /%MCC
    Dataset179.9599.8399.6281.4371.89104.7074
    Dataset283.5999.8999.7086.8077.31118.4762
    Dataset384.2199.7799.6082.4172.89102.7782
    Dataset484.2999.8799.6986.4276.87116.3193
    Table 1. Results of the four trained models on the testset
    Lingmei Ai, Kangzhen Shi. Low-Grade Gliomas MR Image Segmentation Based on Conditional Generative Adversarial Networks[J]. Laser & Optoelectronics Progress, 2020, 57(22): 221004
    Download Citation