Validation of a fully automated liver segmentation algorithm using multi-scale deep reinforcement learning and comparison versus manual segmentation
作者: David J. WinkelThomas J. WeikertHanns-Christian BreitGuillaume ChabinEli GibsonTobias J. HeyeDorin ComaniciuDaniel T. Boll
作者单位: 1Department of Radiology, University Hospital of Basel, Basel, Switzerland
2Siemens Healthineers, Medical Imaging Technologies, Princeton, NJ, USA
刊名: European Journal of Radiology, 2020, Vol.126
来源数据库: Elsevier Journal
DOI: 10.1016/j.ejrad.2020.108918
关键词: Artificial intelligenceAlgorithmsReproducibility of resultsTomographyX-ray computedLiver
原始语种摘要: Abstract(#br)Purpose(#br)To evaluate the performance of an artificial intelligence (AI) based software solution tested on liver volumetric analyses and to compare the results to the manual contour segmentation.(#br)Materials and methods(#br)We retrospectively obtained 462 multiphasic CT datasets with six series for each patient: three different contrast phases and two slice thickness reconstructions (1.5/5 mm), totaling 2772 series. AI-based liver volumes were determined using multi-scale deep-reinforcement learning for 3D body markers detection and 3D structure segmentation. The algorithm was trained for liver volumetry on approximately 5000 datasets. We computed the absolute error of each automatically- and manually-derived volume relative to the mean manual volume. The mean processing...
全文获取路径: Elsevier  (合作)
影响因子:2.512 (2012)

  • segmentation 分段
  • manual 手操啄
  • algorithm 算法
  • liver 
  • segmented 分段[割]的
  • intelligence 智能
  • scale 度盘
  • automated 自动
  • automatically 自动地
  • agreement 协议