Vulnerabilities of radiomic signature development: The need for safeguards. (January 2019)

• Record Type: Journal Article
• Authors: Welch, Mattea L.; McIntosh, Chris; Haibe-Kains, Benjamin; Milosevic, Michael F.; Wee, Leonard; Dekker, Andre; Huang, Shao Hui; Purdie, Thomas G.; O'Sullivan, Brian; Aerts, Hugo J.W.L.; Jaffray, David A.
• Is Part Of: Radiotherapy and oncology. Volume 130(2019)
• Page Start: 2
• Page End: 9
• Publication Date: 2019-01
• Subjects: Radiomics -- Signature development -- Safeguards -- Volume -- Lung cancer -- Head and neck cancer; Oncology -- Periodicals; Radiotherapy -- Periodicals; Tumors -- Periodicals; Medical Oncology -- Periodicals; Neoplasms -- radiotherapy -- Periodicals; Radiotherapy -- Periodicals; Radiothérapie -- Périodiques; Cancérologie -- Périodiques; Tumeurs -- Périodiques; Electronic journals; 616.9940642
• DOI: 10.1016/j.radonc.2018.10.027
• Languages: English
• ISSNs: 0167-8140

Abstract

Highlights: Presented Safeguards ensure productive progress of the radiomic field. Radiomic models and features should be tested to determine added prognostic and predictive accuracy compared to accepted clinical factors. Radiomic features are susceptible to underlying dependencies and multi-collinearity within models. Open-source software should be used in radiomic developments to increase development accountability and facilitate inter-institutional research. Abstract: Purpose: Refinement of radiomic results and methodologies is required to ensure progression of the field. In this work, we establish a set of safeguards designed to improve and support current radiomic methodologies through detailed analysis of a radiomic signature. Methods: A radiomic model (MW2018) was fitted and externally validated using features extracted from previously reported lung and head and neck (H&N) cancer datasets using gross-tumour-volume contours, as well as from images with randomly permuted voxel index values; i.e. images without meaningful texture. To determine MW2018's added benefit, the prognostic accuracy of tumour volume alone was calculated as a baseline. Results: MW2018 had an external validation concordance index (c-index) of 0.64. However, a similar performance was achieved using features extracted from images with randomized signal intensities (c-index = 0.64 and 0.60 for H&N and lung, respectively). Tumour volume had a c-index = 0.64 and correlated strongly with three of the four model features. It was determined that the signature was a surrogate for tumour volume and that intensity and texture values were not pertinent for prognostication. Conclusion: Our experiments reveal vulnerabilities in radiomic signature development processes and suggest safeguards that can be used to refine methodologies, and ensure productive radiomic development using objective and independent features.