David Bouget1, Arve Jørgensen2,3, Gabriel Kiss2,4, Haakon Olav Leira2,5, Thomas Langø6. 1. Department of Circulation and Medical Imaging, NTNU, Trondheim, Norway. david.bouget@ntnu.no. 2. Department of Circulation and Medical Imaging, NTNU, Trondheim, Norway. 3. Department of Radiology and Nuclear Medicine, St. Olavs Hospital, Trondheim, Norway. 4. Operating Room of the Future, St. Olavs Hospital, Trondheim, Norway. 5. Department of Thoracic Medicine, St. Olavs Hospital, Trondheim, Norway. 6. Medical Technology Department, SINTEF, Trondheim, Norway.
Abstract
PURPOSE: Accurate lung cancer diagnosis is crucial to select the best course of action for treating the patient. From a simple chest CT volume, it is necessary to identify whether the cancer has spread to nearby lymph nodes or not. It is equally important to know precisely where each malignant lymph node is with respect to the surrounding anatomical structures and the airways. In this paper, we introduce a new data-set containing annotations of fifteen different anatomical structures in the mediastinal area, including lymph nodes of varying sizes. We present a 2D pipeline for semantic segmentation and instance detection of anatomical structures and potentially malignant lymph nodes in the mediastinal area. METHODS: We propose a 2D pipeline combining the strengths of U-Net for pixel-wise segmentation using a loss function dealing with data imbalance and Mask R-CNN providing instance detection and improved pixel-wise segmentation within bounding boxes. A final stage performs pixel-wise labels refinement and 3D instance detection using a tracking approach along the slicing dimension. Detected instances are represented by a 3D pixel-wise mask, bounding volume, and centroid position. RESULTS: We validated our approach following a fivefold cross-validation over our new data-set of fifteen lung cancer patients. For the semantic segmentation task, we reach an average Dice score of 76% over all fifteen anatomical structures. For the lymph node instance detection task, we reach 75% recall for 9 false positives per patient, with an average centroid position estimation error of 3 mm in each dimension. CONCLUSION: Fusing 2D networks' results increases pixel-wise segmentation results while enabling good instance detection. Better leveraging of the 3D information and station mapping for the detected lymph nodes are the next steps.
PURPOSE: Accurate lung cancer diagnosis is crucial to select the best course of action for treating the patient. From a simple chest CT volume, it is necessary to identify whether the cancer has spread to nearby lymph nodes or not. It is equally important to know precisely where each malignant lymph node is with respect to the surrounding anatomical structures and the airways. In this paper, we introduce a new data-set containing annotations of fifteen different anatomical structures in the mediastinal area, including lymph nodes of varying sizes. We present a 2D pipeline for semantic segmentation and instance detection of anatomical structures and potentially malignant lymph nodes in the mediastinal area. METHODS: We propose a 2D pipeline combining the strengths of U-Net for pixel-wise segmentation using a loss function dealing with data imbalance and Mask R-CNN providing instance detection and improved pixel-wise segmentation within bounding boxes. A final stage performs pixel-wise labels refinement and 3D instance detection using a tracking approach along the slicing dimension. Detected instances are represented by a 3D pixel-wise mask, bounding volume, and centroid position. RESULTS: We validated our approach following a fivefold cross-validation over our new data-set of fifteen lung cancerpatients. For the semantic segmentation task, we reach an average Dice score of 76% over all fifteen anatomical structures. For the lymph node instance detection task, we reach 75% recall for 9 false positives per patient, with an average centroid position estimation error of 3 mm in each dimension. CONCLUSION: Fusing 2D networks' results increases pixel-wise segmentation results while enabling good instance detection. Better leveraging of the 3D information and station mapping for the detected lymph nodes are the next steps.
Authors: Shuai Wang; Yingying Zhu; Sungwon Lee; Daniel C Elton; Thomas C Shen; Youbao Tang; Yifan Peng; Zhiyong Lu; Ronald M Summers Journal: Med Image Anal Date: 2022-01-08 Impact factor: 8.545
Authors: Che Wei Chang; Feipei Lai; Mesakh Christian; Yu Chun Chen; Ching Hsu; Yo Shen Chen; Dun Hao Chang; Tyng Luen Roan; Yen Che Yu Journal: JMIR Med Inform Date: 2021-12-02