Selected article for: "classification segmentation and segmentation classification"
Author: Amine Amyar; Romain Modzelewski; Su Ruan
Title: Multi-task Deep Learning Based CT Imaging Analysis For COVID-19: Classification and Segmentation
  • Document date: 2020_4_21
    • ID: hiac6ur7_42
    Snippet: The main results of the three experiments are shown in Table 2. Metrics include: dice coefficient, accuracy, sensibility, specificity and the area under the ROC curve. Experiment 1: As shown in Table 2 Table 2 , the best result for image segmentation was obtained using our method with a dice coef of 78.52% versus 69.09% and 67.14% using U-NET with 256 x 256 and 512 x 512 resolutions respectively. The combination of the reconstruction, segmentatio.....
    Document: The main results of the three experiments are shown in Table 2. Metrics include: dice coefficient, accuracy, sensibility, specificity and the area under the ROC curve. Experiment 1: As shown in Table 2 Table 2 , the best result for image segmentation was obtained using our method with a dice coef of 78.52% versus 69.09% and 67.14% using U-NET with 256 x 256 and 512 x 512 resolutions respectively. The combination of the reconstruction, segmentation and classification results in a higher accuracy to detect infection regions, compared to the use of the U-NET model alone.

    Search related documents:
    Co phrase search for related documents
    • dice coefficient and image segmentation: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14
    • dice coefficient and infection region: 1
    • good result and ROC curve: 1, 2, 3
    • good result and ROC curve area: 1, 2
    • high accuracy and image segmentation: 1, 2, 3, 4, 5, 6, 7, 8
    • high accuracy and ROC curve: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21
    • high accuracy and ROC curve area: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13
    • high accuracy and ROC curve area specificity: 1, 2
    • high accuracy and Table show: 1
    • image segmentation and infection region: 1
    • image segmentation and ROC curve: 1, 2, 3, 4, 5
    • image segmentation and ROC curve area: 1, 2, 3, 4
    • infection region and main result: 1
    • ROC curve and Table show: 1