東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Streamlining the Verification of Rad...

Choo, Neville Run Kang.

FindBook

Google Book

Amazon

博客來

Streamlining the Verification of Radiotherapy Contours by Identifying Clinically Relevant Subsections.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Streamlining the Verification of Radiotherapy Contours by Identifying Clinically Relevant Subsections./
作者:	Choo, Neville Run Kang.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2024,
面頁冊數:	56 p.
附註:	Source: Masters Abstracts International, Volume: 85-11.
Contained By:	Masters Abstracts International85-11.
標題:	Medicine. -
電子資源:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30997071
ISBN:	9798382744681

Streamlining the Verification of Radiotherapy Contours by Identifying Clinically Relevant Subsections.
Choo, Neville Run Kang.

Streamlining the Verification of Radiotherapy Contours by Identifying Clinically Relevant Subsections. - Ann Arbor : ProQuest Dissertations & Theses, 2024 - 56 p.

Source: Masters Abstracts International, Volume: 85-11.

Thesis (M.S.)--Duke University, 2024.

Purpose: New developments in radiation therapy such as AI based contouring and online adaptive radiotherapy have led to an increase in the number of structures and normal tissues being delineated with less human oversight, requiring review in a shorter timeframe. Addressing this problem, we aim to develop and validate a novel method for QA by streamlining the verification of radiotherapy contours. This method identifies subsections of organs at risk that could result in clinically relevant dose constraints being violated, should the contour be inaccurate. Methods: Structures with planning constraints are evaluated by adding increasing margins and checking against dose constraints. Structures that violate constraints when expanded with margins are flagged for manual review; The smallest margin that violates constraints is prioritized, with the location necessitating manual review presented in a suitable manner, such as the maximum dose point for the contours specified in this study. We applied this method in a retrospective analysis of 92 stereotactic radiosurgery plans, evaluating brainstem, optic nerves, and optic chiasm. Margins of 0, 1, 3, and 5mm were applied to define risk levels (very high, high, medium, low) for manual review. Contours and associated MR images were independently reviewed by 2 physicists with contouring experience using the locations flagged for review by our method to determine whether clinically relevant contouring errors existed. Results: 12 contours (brainstem n=10, optic nerve n=1, chiasm n=1) from 11 plans were flagged for manual review at risk levels of very high (n=1), high (n=4), medium (n=2), and low (n=5). Review by the 2 physicists indicated a mean offset value of 0.45mm, with a mean difference of 0.19mm (SD = 0.57mm) and a correlation value of 0.7 between the two sets of observations. Notably, one case exhibited a mean contouring error of 1.75mm, significantly beyond the standard tolerance for SRS of 1mm, suggesting a critical area of concern. Conclusion: Our results indicate that the method described here has potential to improve both the efficacy and efficiency of the plan review process. When applied to radiosurgery, efficacy improved as a number of previously unidentified contouring errors were identified in critical locations among the 12% of cases flagged for manual review, suggesting potential to reduce medical errors. Potential improvements in efficiency are highlighted by the 88% of cases for which the tool indicated that contouring errors would not have a clinically relevant dosimetric effect, indicating review is not necessary. Further investigation is warranted to explore the application of this method to other treatment sites.

ISBN: 9798382744681Subjects--Topical Terms:

641104
Medicine.
Subjects--Index Terms:

Adaptive radiotherapy

Streamlining the Verification of Radiotherapy Contours by Identifying Clinically Relevant Subsections.
LDR:03899nmm a2200397 4500 001 2399019
005 20240909062219.5
006 m o d
007 cr#unu||||||||
008 251215s2024 ||||||||||||||||| ||eng d
020 $a 9798382744681
035 $a (MiAaPQ)AAI30997071
035 $a AAI30997071
040 $a MiAaPQ $c MiAaPQ
100 1 $a Choo, Neville Run Kang. $3 3768987
245 1 0 $a Streamlining the Verification of Radiotherapy Contours by Identifying Clinically Relevant Subsections.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2024
300 $a 56 p.
500 $a Source: Masters Abstracts International, Volume: 85-11.
500 $a Advisor: Adamson, Justus.
502 $a Thesis (M.S.)--Duke University, 2024.
520 $a Purpose: New developments in radiation therapy such as AI based contouring and online adaptive radiotherapy have led to an increase in the number of structures and normal tissues being delineated with less human oversight, requiring review in a shorter timeframe. Addressing this problem, we aim to develop and validate a novel method for QA by streamlining the verification of radiotherapy contours. This method identifies subsections of organs at risk that could result in clinically relevant dose constraints being violated, should the contour be inaccurate. Methods: Structures with planning constraints are evaluated by adding increasing margins and checking against dose constraints. Structures that violate constraints when expanded with margins are flagged for manual review; The smallest margin that violates constraints is prioritized, with the location necessitating manual review presented in a suitable manner, such as the maximum dose point for the contours specified in this study. We applied this method in a retrospective analysis of 92 stereotactic radiosurgery plans, evaluating brainstem, optic nerves, and optic chiasm. Margins of 0, 1, 3, and 5mm were applied to define risk levels (very high, high, medium, low) for manual review. Contours and associated MR images were independently reviewed by 2 physicists with contouring experience using the locations flagged for review by our method to determine whether clinically relevant contouring errors existed. Results: 12 contours (brainstem n=10, optic nerve n=1, chiasm n=1) from 11 plans were flagged for manual review at risk levels of very high (n=1), high (n=4), medium (n=2), and low (n=5). Review by the 2 physicists indicated a mean offset value of 0.45mm, with a mean difference of 0.19mm (SD = 0.57mm) and a correlation value of 0.7 between the two sets of observations. Notably, one case exhibited a mean contouring error of 1.75mm, significantly beyond the standard tolerance for SRS of 1mm, suggesting a critical area of concern. Conclusion: Our results indicate that the method described here has potential to improve both the efficacy and efficiency of the plan review process. When applied to radiosurgery, efficacy improved as a number of previously unidentified contouring errors were identified in critical locations among the 12% of cases flagged for manual review, suggesting potential to reduce medical errors. Potential improvements in efficiency are highlighted by the 88% of cases for which the tool indicated that contouring errors would not have a clinically relevant dosimetric effect, indicating review is not necessary. Further investigation is warranted to explore the application of this method to other treatment sites.
590 $a School code: 0066.
650 4 $a Medicine. $3 641104
650 4 $a Medical imaging. $3 3172799
650 4 $a Biomedical engineering. $3 535387
653 $a Adaptive radiotherapy
653 $a Autocontouring
653 $a Contouring
653 $a Quality assurance
653 $a Radiation therapy
653 $a Treatment planning
690 $a 0564
690 $a 0574
690 $a 0541
710 2 $a Duke University. $b DKU - Medical Physics Master of Science Program. $3 3768988
773 0 $t Masters Abstracts International $g 85-11.
790 $a 0066
791 $a M.S.
792 $a 2024
793 $a English
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30997071