Machine learning applications in head and neck radiation oncology: Lessons from open-source radiomics challenges

on behalf of MICCAI/M.D. Anderson Cancer Center Head and Neck Quantitative Imaging Working Group

Research output: Contribution to journalArticlepeer-review

10 Scopus citations

Abstract

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.

Original languageEnglish (US)
Article number294
JournalFrontiers in Oncology
Volume8
Issue numberAUG
DOIs
StatePublished - Aug 17 2018

Bibliographical note

Funding Information:
Multiple funders/agencies contributed to personnel salaries or project support during the manuscript preparation interval. Dr. HE is supported in part by the philanthropic donations from the Family of Paul W. Beach to Dr. G. Brandon Gunn, MD. This research was supported by the Andrew Sabin Family Foundation; Dr. CF is a Sabin Family Foundation Fellow. Drs. SL, AM, and CF receive funding support from the National Institutes of Health (NIH)/National Institute for Dental and Craniofacial Research (1R01DE025248-01/R56DE025248-01). Drs. GM, DV, GC, and CF are supported via a National Science Foundation (NSF), Division of Mathematical Sciences, Joint NIH/NSF Initiative on Quantitative Approaches to Biomedical Big Data (QuBBD) Grant (NSF 1557679). Dr. CF received grant and/or salary support from the NIH/National Cancer Institute (NCI) Head and Neck Specialized Programs of Research Excellence (SPORE) Developmental Research Program Award (P50 CA097007-10) and the Paul Calabresi Clinical Oncology Program Award (K12 CA088084-06), the Center for Radiation Oncology Research (CROR) at MD Anderson Cancer Center Seed Grant; and the MD Anderson Institutional Research Grant (IRG) Program. Dr. JK-C is supported by the National Cancer Institute (U24 CA180927-03, U01 CA154601-06. Mr. Kanwar was supported by a 2016-2017 Radiological Society of North America Education and Research Foundation Research Medical Student Grant Award (RSNA RMS1618). GM's work is partially supported by National Institutes of Health (NIH) awards NCI-R01-CA214825, NCI-R01CA225190, and NLM-R01LM012527; by National Science Foundation (NSF) award CNS-1625941 and by The Joseph and Bessie Feinberg Foundation. Dr. CF received a General Electric Healthcare/MD Anderson Center for Advanced Biomedical Imaging In-Kind Award and an Elekta AB/MD Anderson Department of Radiation Oncology Seed Grant. Dr. CF has also received speaker travel funding from Elekta AB. None of these industrial partners' equipment was directly used or experimented with in the present work.

Keywords

  • Big data
  • Head and neck
  • Machine learning
  • Radiation oncology
  • Radiomics challenge

Fingerprint Dive into the research topics of 'Machine learning applications in head and neck radiation oncology: Lessons from open-source radiomics challenges'. Together they form a unique fingerprint.

Cite this