방사선종양학

본문글자크기
  • [Breast .] Assessment of deep learning-based auto-contouring on interobserver consistency in target volume and organs-at-risk delineation for breast cancer: Implications for RTQA program in a multi-institutional study

    [Breast .] Assessment of deep learning-based auto-contouring on interobserver consistency in target volume and organs-at-risk delineation for breast cancer: Implications for RTQA program in a multi-institutional study

    연세의대, 서울의대 / 최민서, 장지석, 김진성*, 신경환*

  • 출처
    Breast .
  • 등재일
    2024 Feb:73:103599.
  • 저널이슈번호
  • 내용

    바로가기  >

    Abstract
    Purpose: To quantify interobserver variation (IOV) in target volume and organs-at-risk (OAR) contouring across 31 institutions in breast cancer cases and to explore the clinical utility of deep learning (DL)-based auto-contouring in reducing potential IOV.

    Methods and materials: In phase 1, two breast cancer cases were randomly selected and distributed to multiple institutions for contouring six clinical target volumes (CTVs) and eight OAR. In Phase 2, auto-contour sets were generated using a previously published DL Breast segmentation model and were made available for all participants. The difference in IOV of submitted contours in phases 1 and 2 was investigated quantitatively using the Dice similarity coefficient (DSC) and Hausdorff distance (HD). The qualitative analysis involved using contour heat maps to visualize the extent and location of these variations and the required modification.

    Results: Over 800 pairwise comparisons were analysed for each structure in each case. Quantitative phase 2 metrics showed significant improvement in the mean DSC (from 0.69 to 0.77) and HD (from 34.9 to 17.9 mm). Quantitative analysis showed increased interobserver agreement in phase 2, specifically for CTV structures (5-19 %), leading to fewer manual adjustments. Underlying IOV differences causes were reported using a questionnaire and hierarchical clustering analysis based on the volume of CTVs.

    Conclusion: DL-based auto-contours improved the contour agreement for OARs and CTVs significantly, both qualitatively and quantitatively, suggesting its potential role in minimizing radiation therapy protocol deviation.

     

     

    Affiliations

    Min Seo Choi 1, Jee Suk Chang 1, Kyubo Kim 2, Jin Hee Kim 3, Tae Hyung Kim 4, Sungmin Kim 5, Hyejung Cha 6, Oyeon Cho 7, Jin Hwa Choi 8, Myungsoo Kim 9, Juree Kim 10, Tae Gyu Kim 11, Seung-Gu Yeo 12, Ah Ram Chang 13, Sung-Ja Ahn 14, Jinhyun Choi 15, Ki Mun Kang 16, Jeanny Kwon 17, Taeryool Koo 18, Mi Young Kim 19, Seo Hee Choi 20, Bae Kwon Jeong 21, Bum-Sup Jang 22, In Young Jo 23, Hyebin Lee 24, Nalee Kim 25, Hae Jin Park 26, Jung Ho Im 27, Sea-Won Lee 28, Yeona Cho 29, Sun Young Lee 30, Ji Hyun Chang 22, Jaehee Chun 1, Eung Man Lee 31, Jin Sung Kim 32, Kyung Hwan Shin 33, Yong Bae Kim 1
    1Department of Radiation Oncology, Yonsei University College of Medicine, Seoul, Republic of Korea.
    2Department of Radiation Oncology, Ewha Womans University College of Medicine, Seoul, Republic of Korea; Department of Radiation Oncology, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seongnam, Republic of Korea.
    3Department of Radiation Oncology, Dongsan Medical Center, Keimyung University School of Medicine, Daegu, Republic of Korea.
    4Department of Radiation Oncology, Nowon Eulji Medical Center, Eulji University School of Medicine, Seoul, Republic of Korea.
    5Department of Radiation Oncology, Dong-A University Hospital, Dong-A University College of Medicine, Busan, Republic of Korea.
    6Department of Radiation Oncology, Yonsei University Wonju College of Medicine, Wonju, Republic of Korea.
    7Department of Radiation Oncology, Ajou University School of Medicine, Suwon, Republic of Korea.
    8Department of Radiation Oncology, Chung-Ang University Hospital, Seoul, Republic of Korea.
    9Department of Radiation Oncology, Incheon St Mary's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea.
    10Department of Radiation Oncology, Ilsan CHA Medical Center, CHA University School of Medicine, Goyang, Republic of Korea.
    11Department of Radiation Oncology, Samsung Changwon Hospital, Sungkyunkwan University School of Medicine, Changwon, Republic of Korea.
    12Department of Radiation Oncology, Soonchunhyang University College of Medicine, Soonchunhyang University Hospital, Bucheon, Republic of Korea.
    13Department of Radiation Oncology, Soonchunhyang University College of Medicine, Seoul, Republic of Korea.
    14Department of Radiation Oncology, Chonnam National University Medical School, Gwangju, Republic of Korea.
    15Department of Radiation Oncology, Jeju National University Hospital, Jeju University College of Medicine, Republic of Korea.
    16Gyeongsang National University Changwon Hospital, Gyeongsang National University College of Medicine, Jinju, Republic of Korea.
    17Department of Radiation Oncology, Chungnam National University School of Medicine, Daejeon, Republic of Korea.
    18Department of Radiation Oncology, Hallym University Sacred Heart Hospital, Hallym University College of Medicine, Anyang, Republic of Korea.
    19Department of Radiation Oncology, Kyungpook National University Chilgok Hospital, Daegu, Republic of Korea.
    20Department of Radiation Oncology, Yongin Severance Hospital, Yonsei University College of Medicine, Yongin, Republic of Korea.
    21Department of Radiation Oncology, Gyeongsang National University Hospital, Gyeongsang National University College of Medicine, Jinju, Republic of Korea.
    22Department of Radiation Oncology, Seoul National University College of Medicine, Seoul, Republic of Korea.
    23Department of Radiation Oncology, Soonchunhyang University Hospital, Cheonan, Republic of Korea.
    24Department of Radiation Oncology, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea.
    25Department of Radiation Oncology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea.
    26Department of Radiation Oncology, Hanyang University College of Medicine, Seoul, Republic of Korea.
    27Department of Radiation Oncology, CHA Bundang Medical Center, CHA University School of Medicine, Seongnam, Republic of Korea.
    28Department of Radiation Oncology, Eunpyeong St. Mary's Hospital, Catholic University of Korea College of Medicine, Seoul, Republic of Korea.
    29Department of Radiation Oncology, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, Republic of Korea.
    30Department of Radiation Oncology, Chonbuk National University Hospital, Jeonju, Republic of Korea.
    31Department of Radiation Oncology, Ewha Womans University College of Medicine, Seoul, Republic of Korea.
    32Department of Radiation Oncology, Yonsei University College of Medicine, Seoul, Republic of Korea. Electronic address: jinsung@yuhs.ac.
    33Department of Radiation Oncology, Seoul National University College of Medicine, Seoul, Republic of Korea. Electronic address: radiat@snu.ac.kr.

  • 키워드
    Auto-contouring; Breast cancer; Deep learning; Inter-observer variation; RTQA.
  • 연구소개
    AI 기반 자동 컨투어링은 현재 임상에서 방사선 치료 계획 단계에서 적용되어 많이 사용되고 있지만 아직까지 품질 보증 측면에서 어떠한 영향이 있는지는 보고된 바가 없었습니다. 본 연구에서는 국내 31개 기관에 동일한 환자 케이스를 배포하여 컨투어링 방식을 조사하였고, AI 기반 자동 컨투어링 툴의 사용 유무에 따라 기관 별로 얼마나 차이가 생기는지 분석하였습니다. 결과적으로 AI 기반 자동 컨투어링 툴이 기관 및 의사 간의 차이를 유의미하게 줄여주는 것을 발견하였고, 이는 향 후 대규모 임상 연구에서 도움이 될만한 정보라고 생각합니다.
  • 편집위원

    31개 기관에서 Target Volume과 주변 장기에 대하여 Deep Learning을 통해 자동 컨투어가 Interobserver Variation을 감소시켜줄 수 있는가에 대한 현실적인 답변을 제시

    2024-03-28 17:38:21

  • 덧글달기
    덧글달기
       IP : 3.135.200.211

    등록