Sándor Kónya1, Tr Sai Natarajan2, Hassan Allouch3, Kais Abu Nahleh3, Omneya Yakout Dogheim4, Heinrich Boehm3. 1. Center for Diagnostic and Interventional Radiology and Neuroradiology, Bad Berka, Germany. 2. Independent Researcher, Chennai, India. 3. Department of Spinal Surgery, Zentralklinik Bad Berka, Bad Berka, Germany. 4. Department of Diagnostic and Interventional Radiology, Faculty of Medicine, University of Alexandria, Egypt.
Abstract
PURPOSE: This study investigated the segmentation metrics of different segmentation networks trained on 730 manually annotated lateral lumbar spine X-rays to test the generalization ability and robustness which are the basis of clinical decision support algorithms. METHODS: Instance segmentation networks were compared to semantic segmentation networks based on different metrics. The study cohort comprised diseased spines and postoperative images with metallic implants. RESULTS: However, the pixel accuracies and intersection over union are similarly high for the best performing instance and semantic segmentation models; the observed vertebral recognition rates of the instance segmentation models statistically significantly outperform the semantic models' recognition rates. CONCLUSION: The results of the instance segmentation models on lumbar spine X-ray perform superior to semantic segmentation models in the recognition rates even by images of severe diseased spines by allowing the segmentation of overlapping vertebrae, in contrary to the semantic models where such differentiation cannot be performed due to the fused binary mask of the overlapping instances. These models can be incorporated into further clinical decision support pipelines. Copyright:
PURPOSE: This study investigated the segmentation metrics of different segmentation networks trained on 730 manually annotated lateral lumbar spine X-rays to test the generalization ability and robustness which are the basis of clinical decision support algorithms. METHODS: Instance segmentation networks were compared to semantic segmentation networks based on different metrics. The study cohort comprised diseased spines and postoperative images with metallic implants. RESULTS: However, the pixel accuracies and intersection over union are similarly high for the best performing instance and semantic segmentation models; the observed vertebral recognition rates of the instance segmentation models statistically significantly outperform the semantic models' recognition rates. CONCLUSION: The results of the instance segmentation models on lumbar spine X-ray perform superior to semantic segmentation models in the recognition rates even by images of severe diseased spines by allowing the segmentation of overlapping vertebrae, in contrary to the semantic models where such differentiation cannot be performed due to the fused binary mask of the overlapping instances. These models can be incorporated into further clinical decision support pipelines. Copyright:
Authors: William Owens; Raymond Wiegand; Mark Studin; Donald Capoferri; Kenneth Barooha; Alexander Greaux; Robert Rattray; Adam Hutton; John Cintineo; Vipin Chaudhary Journal: Conf Proc IEEE Eng Med Biol Soc Date: 2017-07
Authors: Brian H Cho; Deepak Kaji; Zoe B Cheung; Ivan B Ye; Ray Tang; Amy Ahn; Oscar Carrillo; John T Schwartz; Aly A Valliani; Eric K Oermann; Varun Arvind; Daniel Ranti; Li Sun; Jun S Kim; Samuel K Cho Journal: Global Spine J Date: 2019-08-13