TY - JOUR
T1 - Machine learning applications in head and neck radiation oncology
T2 - Lessons from open-source radiomics challenges
AU - on behalf of MICCAI/M.D. Anderson Cancer Center Head and Neck Quantitative Imaging Working Group
AU - Elhalawani, Hesham
AU - Lin, Timothy A.
AU - Volpe, Stefania
AU - Mohamed, Abdallah S.R.
AU - White, Aubrey L.
AU - Zafereo, James
AU - Wong, Andrew J.
AU - Berends, Joel E.
AU - AboHashem, Shady
AU - Williams, Bowman
AU - Aymard, Jeremy M.
AU - Kanwar, Aasheesh
AU - Perni, Subha
AU - Rock, Crosby D.
AU - Cooksey, Luke
AU - Campbell, Shauna
AU - Yang, Pei
AU - Nguyen, Khahn
AU - Ger, Rachel B.
AU - Cardenas, Carlos E.
AU - Fave, Xenia J.
AU - Sansone, Carlo
AU - Piantadosi, Gabriele
AU - Marrone, Stefano
AU - Liu, Rongjie
AU - Huang, Chao
AU - Yu, Kaixian
AU - Li, Tengfei
AU - Yu, Yang
AU - Zhang, Youyi
AU - Zhu, Hongtu
AU - Morris, Jeffrey S.
AU - Baladandayuthapani, Veerabhadran
AU - Shumway, John W.
AU - Ghosh, Alakonanda
AU - Pöhlmann, Andrei
AU - Phoulady, Hady A.
AU - Goyal, Vibhas
AU - Canahuate, Guadalupe
AU - Marai, Elisabeta G.
AU - Vock, David
AU - Lai, Stephen Y.
AU - Mackin, Dennis S.
AU - Court, Laurence E.
AU - Freymann, John
AU - Farahani, Keyvan
AU - Kaplathy-Cramer, Jayashree
AU - Fuller, Clifton D.
N1 - Publisher Copyright:
© 2018 Elhalawani, Lin, Volpe, Mohamed, White, Zafereo, Wong, Berends, AboHashem,Williams, Aymard, Kanwar, Perni, Rock, Cooksey, Campbell, Yang, Nguyen, Ger, Cardenas, Fave, Sansone, Piantadosi, Marrone, Liu, Huang, Yu, Li, Yu, Zhang, Zhu, Morris, Baladandayuthapani, Shumway, Ghosh, Pöhlmann, Phoulady, Goyal, Canahuate, Marai, Vock, Lai, Mackin, Court, Freymann, Farahani, Kaplathy-Cramer and Fuller.
PY - 2018/8/17
Y1 - 2018/8/17
N2 - Radiomics leverages existing image datasets to provide non-visible data extraction via image post-processing, with the aim of identifying prognostic, and predictive imaging features at a sub-region of interest level. However, the application of radiomics is hampered by several challenges such as lack of image acquisition/analysis method standardization, impeding generalizability. As of yet, radiomics remains intriguing, but not clinically validated. We aimed to test the feasibility of a non-custom-constructed platform for disseminating existing large, standardized databases across institutions for promoting radiomics studies. Hence, University of Texas MD Anderson Cancer Center organized two public radiomics challenges in head and neck radiation oncology domain. This was done in conjunction with MICCAI 2016 satellite symposium using Kaggle-in-Class, a machine-learning and predictive analytics platform. We drew on clinical data matched to radiomics data derived from diagnostic contrast-enhanced computed tomography (CECT) images in a dataset of 315 patients with oropharyngeal cancer. Contestants were tasked to develop models for (i) classifying patients according to their human papillomavirus status, or (ii) predicting local tumor recurrence, following radiotherapy. Data were split into training, and test sets. Seventeen teams from various professional domains participated in one or both of the challenges. This review paper was based on the contestants' feedback; provided by 8 contestants only (47%). Six contestants (75%) incorporated extracted radiomics features into their predictive model building, either alone (n = 5; 62.5%), as was the case with the winner of the "HPV" challenge, or in conjunction with matched clinical attributes (n = 2; 25%). Only 23% of contestants, notably, including the winner of the "local recurrence" challenge, built their model relying solely on clinical data. In addition to the value of the integration of machine learning into clinical decision-making, our experience sheds light on challenges in sharing and directing existing datasets toward clinical applications of radiomics, including hyper-dimensionality of the clinical/imaging data attributes. Our experience may help guide researchers to create a framework for sharing and reuse of already published data that we believe will ultimately accelerate the pace of clinical applications of radiomics; both in challenge or clinical settings.
AB - Radiomics leverages existing image datasets to provide non-visible data extraction via image post-processing, with the aim of identifying prognostic, and predictive imaging features at a sub-region of interest level. However, the application of radiomics is hampered by several challenges such as lack of image acquisition/analysis method standardization, impeding generalizability. As of yet, radiomics remains intriguing, but not clinically validated. We aimed to test the feasibility of a non-custom-constructed platform for disseminating existing large, standardized databases across institutions for promoting radiomics studies. Hence, University of Texas MD Anderson Cancer Center organized two public radiomics challenges in head and neck radiation oncology domain. This was done in conjunction with MICCAI 2016 satellite symposium using Kaggle-in-Class, a machine-learning and predictive analytics platform. We drew on clinical data matched to radiomics data derived from diagnostic contrast-enhanced computed tomography (CECT) images in a dataset of 315 patients with oropharyngeal cancer. Contestants were tasked to develop models for (i) classifying patients according to their human papillomavirus status, or (ii) predicting local tumor recurrence, following radiotherapy. Data were split into training, and test sets. Seventeen teams from various professional domains participated in one or both of the challenges. This review paper was based on the contestants' feedback; provided by 8 contestants only (47%). Six contestants (75%) incorporated extracted radiomics features into their predictive model building, either alone (n = 5; 62.5%), as was the case with the winner of the "HPV" challenge, or in conjunction with matched clinical attributes (n = 2; 25%). Only 23% of contestants, notably, including the winner of the "local recurrence" challenge, built their model relying solely on clinical data. In addition to the value of the integration of machine learning into clinical decision-making, our experience sheds light on challenges in sharing and directing existing datasets toward clinical applications of radiomics, including hyper-dimensionality of the clinical/imaging data attributes. Our experience may help guide researchers to create a framework for sharing and reuse of already published data that we believe will ultimately accelerate the pace of clinical applications of radiomics; both in challenge or clinical settings.
KW - Big data
KW - Head and neck
KW - Machine learning
KW - Radiation oncology
KW - Radiomics challenge
UR - http://www.scopus.com/inward/record.url?scp=85051839540&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85051839540&partnerID=8YFLogxK
U2 - 10.3389/fonc.2018.00294
DO - 10.3389/fonc.2018.00294
M3 - Article
C2 - 30175071
AN - SCOPUS:85051839540
SN - 2234-943X
VL - 8
JO - Frontiers in Oncology
JF - Frontiers in Oncology
IS - AUG
M1 - 294
ER -