https://scholars.lib.ntu.edu.tw/handle/123456789/579835
Title: | Multiple Resolution Residually Connected Feature Streams for Automatic Lung Tumor Segmentation From CT Images | Authors: | Jiang, Jue Hu, Yu-Chi CHIA-JU LIU Halpenny, Darragh Hellmann, Matthew D Deasy, Joseph O Mageras, Gig Veeraraghavan, Harini |
Keywords: | Deep learning | detection | longitudinal | lung cancer | segmentation;Deep learning; detection; longitudinal; lung cancer; segmentation | Issue Date: | 2019 | Journal Volume: | 38 | Journal Issue: | 1 | Source: | IEEE transactions on medical imaging | Abstract: | Volumetric lung tumor segmentation and accurate longitudinal tracking of tumor volume changes from computed tomography images are essential for monitoring tumor response to therapy. Hence, we developed two multiple resolution residually connected network (MRRN) formulations called incremental-MRRN and dense-MRRN. Our networks simultaneously combine features across multiple image resolution and feature levels through residual connections to detect and segment the lung tumors. We evaluated our method on a total of 1210 non-small cell (NSCLC) lung tumors and nodules from three data sets consisting of 377 tumors from the open-source Cancer Imaging Archive (TCIA), 304 advanced stage NSCLC treated with anti- PD-1 checkpoint immunotherapy from internal institution MSKCC data set, and 529 lung nodules from the Lung Image Database Consortium (LIDC). The algorithm was trained using 377 tumors from the TCIA data set and validated on the MSKCC and tested on LIDC data sets. The segmentation accuracy compared to expert delineations was evaluated by computing the dice similarity coefficient, Hausdorff distances, sensitivity, and precision metrics. Our best performing incremental-MRRN method produced the highest DSC of 0.74 ± 0.13 for TCIA, 0.75±0.12 for MSKCC, and 0.68±0.23 for the LIDC data sets. There was no significant difference in the estimations of volumetric tumor changes computed using the incremental-MRRN method compared with the expert segmentation. In summary, we have developed a multi-scale CNN approach for volumetrically segmenting lung tumors which enables accurate, automated identification of and serial measurement of tumor volumes in the lung. |
URI: | https://scholars.lib.ntu.edu.tw/handle/123456789/579835 | ISSN: | 02780062 | DOI: | 10.1109/TMI.2018.2857800 | SDG/Keyword: | Biological organs; Computerized tomography; Deep learning; Diseases; Error detection; Feature extraction; Image resolution; Media streaming; Medical imaging; Tumors; Cancer; Longitudinal; Lung; Lung Cancer; Streaming media; Image segmentation; immunostimulating agent; programmed death 1 ligand 1; algorithm; Article; cancer diagnosis; cancer immunotherapy; dense multiple resolution residually connected network; diagnostic accuracy; feature extraction; human; image segmentation; incremental multiple resolution residually connected network; lung nodule; lung tumor; major clinical study; oncological parameters; quantitative analysis; radiological parameters; random forest; tumor number; tumor volume; x-ray computed tomography; computer assisted diagnosis; diagnostic imaging; factual database; lung; lung tumor; procedures; x-ray computed tomography; Algorithms; Databases, Factual; Deep Learning; Humans; Image Interpretation, Computer-Assisted; Lung; Lung Neoplasms; Tomography, X-Ray Computed |
Appears in Collections: | 醫學院附設醫院 (臺大醫院) |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.