Selected article for: "black box and deep learning"

Author: Chuansheng Zheng; Xianbo Deng; Qing Fu; Qiang Zhou; Jiapei Feng; Hui Ma; Wenyu Liu; Xinggang Wang
Title: Deep Learning-based Detection for COVID-19 from Chest CT using Weak Label
  • Document date: 2020_3_17
  • ID: ll4rxd9p_41
    Snippet: Limitations of this study There are still several limitations in this study. First, network design and training may be further improved. For example, the UNet model for lung segmentation did not utilize temporal information and it was trained using imperfect ground-truth masks, which could be improved by using 3D segmentation networks and adopting precise ground-truth annotated by experts. Second, the data used in this study came from a single ho.....
    Document: Limitations of this study There are still several limitations in this study. First, network design and training may be further improved. For example, the UNet model for lung segmentation did not utilize temporal information and it was trained using imperfect ground-truth masks, which could be improved by using 3D segmentation networks and adopting precise ground-truth annotated by experts. Second, the data used in this study came from a single hospital and cross-center validations were not performed. Third, when diagnosing COVID-19, the algorithm worked in a black-box manner, since the algorithm was based on deep learning and its explainability was still at an early stage. Related work of all limitations mentioned above will be addressed in our further studies.

    Search related documents:
    Co phrase search for related documents
    • black box and early stage: 1, 2
    • black box and ground truth: 1, 2, 3
    • black box and lung segmentation: 1
    • black box and network design: 1, 2
    • black box manner and deep learning: 1
    • black box manner and early stage: 1
    • deep learning and early stage: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25
    • deep learning and ground truth: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25
    • deep learning and lung segmentation: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25
    • deep learning and network design: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25
    • early stage and ground truth: 1, 2
    • early stage and lung segmentation: 1, 2, 3
    • early stage and network design: 1
    • ground truth and lung segmentation: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12
    • ground truth and network design: 1, 2, 3
    • ground truth mask and lung segmentation: 1
    • lung segmentation and network design: 1