Saif Baig, Igor Vidic, George M Mastorakos, Robert X Smith, Nathan White, Suzie Bash, Anders M Dale, Carrie R McDonald, Thomas Beaumont, Tyler M Seibert, Jona Hattangadi-Gluth, Santosh Kesari, Nikdokht Farid, Jeffrey D Rudie
{"title":"治疗前和治疗后弥漫性胶质瘤组织亚区(包括切除腔)的分割。","authors":"Saif Baig, Igor Vidic, George M Mastorakos, Robert X Smith, Nathan White, Suzie Bash, Anders M Dale, Carrie R McDonald, Thomas Beaumont, Tyler M Seibert, Jona Hattangadi-Gluth, Santosh Kesari, Nikdokht Farid, Jeffrey D Rudie","doi":"10.1093/noajnl/vdae140","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Evaluating longitudinal changes in gliomas is a time-intensive process with significant interrater variability. Automated segmentation could reduce interrater variability and increase workflow efficiency for assessment of treatment response. We sought to evaluate whether neural networks would be comparable to expert assessment of pre- and posttreatment diffuse gliomas tissue subregions including resection cavities.</p><p><strong>Methods: </strong>A retrospective cohort of 647 MRIs of patients with diffuse gliomas (average 55.1 years; 29%/36%/34% female/male/unknown; 396 pretreatment and 251 posttreatment, median 237 days post-surgery) from 7 publicly available repositories in The Cancer Imaging Archive were split into training (536) and test/generalization (111) samples. T1, T1-post-contrast, T2, and FLAIR images were used as inputs into a 3D nnU-Net to predict 3 tumor subregions and resection cavities. We evaluated the performance of networks trained on pretreatment training cases (Pre-Rx network), posttreatment training cases (Post-Rx network), and both pre- and posttreatment cases (Combined networks).</p><p><strong>Results: </strong>Segmentation performance was as good as or better than interrater reliability with median dice scores for main tumor subregions ranging from 0.82 to 0.94 and strong correlations between manually segmented and predicted total lesion volumes (0.94 < <i>R</i> <sup>2</sup> values < 0.98). The Combined network performed similarly to the Pre-Rx network on pretreatment cases and the Post-Rx network on posttreatment cases with fewer false positive resection cavities (7% vs 59%).</p><p><strong>Conclusions: </strong>Neural networks that accurately segment pre- and posttreatment diffuse gliomas have the potential to improve response assessment in clinical trials and reduce provider burden and errors in measurement.</p>","PeriodicalId":94157,"journal":{"name":"Neuro-oncology advances","volume":"6 1","pages":"vdae140"},"PeriodicalIF":3.7000,"publicationDate":"2024-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11407510/pdf/","citationCount":"0","resultStr":"{\"title\":\"Segmentation of pre- and posttreatment diffuse glioma tissue subregions including resection cavities.\",\"authors\":\"Saif Baig, Igor Vidic, George M Mastorakos, Robert X Smith, Nathan White, Suzie Bash, Anders M Dale, Carrie R McDonald, Thomas Beaumont, Tyler M Seibert, Jona Hattangadi-Gluth, Santosh Kesari, Nikdokht Farid, Jeffrey D Rudie\",\"doi\":\"10.1093/noajnl/vdae140\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Background: </strong>Evaluating longitudinal changes in gliomas is a time-intensive process with significant interrater variability. Automated segmentation could reduce interrater variability and increase workflow efficiency for assessment of treatment response. We sought to evaluate whether neural networks would be comparable to expert assessment of pre- and posttreatment diffuse gliomas tissue subregions including resection cavities.</p><p><strong>Methods: </strong>A retrospective cohort of 647 MRIs of patients with diffuse gliomas (average 55.1 years; 29%/36%/34% female/male/unknown; 396 pretreatment and 251 posttreatment, median 237 days post-surgery) from 7 publicly available repositories in The Cancer Imaging Archive were split into training (536) and test/generalization (111) samples. T1, T1-post-contrast, T2, and FLAIR images were used as inputs into a 3D nnU-Net to predict 3 tumor subregions and resection cavities. We evaluated the performance of networks trained on pretreatment training cases (Pre-Rx network), posttreatment training cases (Post-Rx network), and both pre- and posttreatment cases (Combined networks).</p><p><strong>Results: </strong>Segmentation performance was as good as or better than interrater reliability with median dice scores for main tumor subregions ranging from 0.82 to 0.94 and strong correlations between manually segmented and predicted total lesion volumes (0.94 < <i>R</i> <sup>2</sup> values < 0.98). The Combined network performed similarly to the Pre-Rx network on pretreatment cases and the Post-Rx network on posttreatment cases with fewer false positive resection cavities (7% vs 59%).</p><p><strong>Conclusions: </strong>Neural networks that accurately segment pre- and posttreatment diffuse gliomas have the potential to improve response assessment in clinical trials and reduce provider burden and errors in measurement.</p>\",\"PeriodicalId\":94157,\"journal\":{\"name\":\"Neuro-oncology advances\",\"volume\":\"6 1\",\"pages\":\"vdae140\"},\"PeriodicalIF\":3.7000,\"publicationDate\":\"2024-08-16\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11407510/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Neuro-oncology advances\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1093/noajnl/vdae140\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2024/1/1 0:00:00\",\"PubModel\":\"eCollection\",\"JCR\":\"Q1\",\"JCRName\":\"CLINICAL NEUROLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Neuro-oncology advances","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1093/noajnl/vdae140","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/1/1 0:00:00","PubModel":"eCollection","JCR":"Q1","JCRName":"CLINICAL NEUROLOGY","Score":null,"Total":0}
Segmentation of pre- and posttreatment diffuse glioma tissue subregions including resection cavities.
Background: Evaluating longitudinal changes in gliomas is a time-intensive process with significant interrater variability. Automated segmentation could reduce interrater variability and increase workflow efficiency for assessment of treatment response. We sought to evaluate whether neural networks would be comparable to expert assessment of pre- and posttreatment diffuse gliomas tissue subregions including resection cavities.
Methods: A retrospective cohort of 647 MRIs of patients with diffuse gliomas (average 55.1 years; 29%/36%/34% female/male/unknown; 396 pretreatment and 251 posttreatment, median 237 days post-surgery) from 7 publicly available repositories in The Cancer Imaging Archive were split into training (536) and test/generalization (111) samples. T1, T1-post-contrast, T2, and FLAIR images were used as inputs into a 3D nnU-Net to predict 3 tumor subregions and resection cavities. We evaluated the performance of networks trained on pretreatment training cases (Pre-Rx network), posttreatment training cases (Post-Rx network), and both pre- and posttreatment cases (Combined networks).
Results: Segmentation performance was as good as or better than interrater reliability with median dice scores for main tumor subregions ranging from 0.82 to 0.94 and strong correlations between manually segmented and predicted total lesion volumes (0.94 < R2 values < 0.98). The Combined network performed similarly to the Pre-Rx network on pretreatment cases and the Post-Rx network on posttreatment cases with fewer false positive resection cavities (7% vs 59%).
Conclusions: Neural networks that accurately segment pre- and posttreatment diffuse gliomas have the potential to improve response assessment in clinical trials and reduce provider burden and errors in measurement.