S. Barley, C. Antoine, G. Webster, M. Tiffany, N. Nundlall, R. Simmons, A. Hartley
{"title":"Atlas-based Auto-contouring - Balancing Accuracy with Efficiency in OnQ rts ®","authors":"S. Barley, C. Antoine, G. Webster, M. Tiffany, N. Nundlall, R. Simmons, A. Hartley","doi":"10.17925/EOH.2014.10.2.98","DOIUrl":null,"url":null,"abstract":"Atlas-based auto segmentation with OnQ rts ® has been shown to deliver time-savings for the delineation of organs at risk in head and neck patients being treated with intensity-modulated radiotherapy. However, as the initial time to set up atlases can be high in busy departments the optimal number of atlas cases needed for auto-contouring was investigated. Using conformity index and mean distance to conformity to compare automatically generated with gold standard clinical contours, it was found that the majority of contours were unaffected by reducing the number of atlas cases from 30 to 10. The optimum number of atlas cases, however, was considered to be 20 due to the reduction in accuracy of the mandible, larynx and brain, below this level.","PeriodicalId":38554,"journal":{"name":"European Oncology and Haematology","volume":"14 1","pages":"98"},"PeriodicalIF":0.0000,"publicationDate":"2014-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"European Oncology and Haematology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.17925/EOH.2014.10.2.98","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"Medicine","Score":null,"Total":0}
引用次数: 3
Abstract
Atlas-based auto segmentation with OnQ rts ® has been shown to deliver time-savings for the delineation of organs at risk in head and neck patients being treated with intensity-modulated radiotherapy. However, as the initial time to set up atlases can be high in busy departments the optimal number of atlas cases needed for auto-contouring was investigated. Using conformity index and mean distance to conformity to compare automatically generated with gold standard clinical contours, it was found that the majority of contours were unaffected by reducing the number of atlas cases from 30 to 10. The optimum number of atlas cases, however, was considered to be 20 due to the reduction in accuracy of the mandible, larynx and brain, below this level.