Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Dec 7;66(24).
doi: 10.1088/1361-6560/ac39e5.

Evaluation of conventional and deep learning based image harmonization methods in radiomics studies

F Tixier  1   2 , V Jaouen  1   3 , C Hognon  1   4 , O Gallinato  4 , T Colin  4 , D Visvikis  1
Affiliations

Evaluation of conventional and deep learning based image harmonization methods in radiomics studies

F Tixier et al. Phys Med Biol. .

Abstract

Objective.To evaluate the impact of image harmonization on outcome prediction models using radiomics.Approach.234 patients from the Brain Tumor Image Segmentation Benchmark (BRATS) dataset with T1 MRI were enrolled in this study. Images were harmonized to a reference image using histogram matching (HHM) and a generative adversarial network (GAN)-based method (HGAN). 88 radiomics features were extracted on HHM, HGANand original (HNONE) images. Wilcoxon paired test was used to identify features significantly impacted by the harmonization protocol used. Radiomic prediction models were built using feature selection with the Least Absolute Shrinkage and Selection Operator (LASSO) and Kaplan-Meier analysis.Main results.More than 50% of the features (49/88) were statistically modified by the harmonization with HHMand 55 with HGAN(adjustedp-value < 0.05). The contribution of histogram and texture features selected by the LASSO, in comparison to shape features that were not impacted by harmonization, was higher in harmonized datasets (47% for Hnone, 62% for HHMand 71% for HGAN). Both image-based harmonization methods allowed to split patients into two groups with significantly different survival (p<0.05). With the HGANimages, we were also able to build and validate a model using only features impacted by the harmonization (median survivals of 189 versus 437 days,p= 0.006)Significance.Data harmonization in a multi-institutional cohort allows to recover the predictive value of some radiomics features that was lost due to differences in the image properties across centers. In terms of ability to build survival prediction models in the BRATS dataset, the loss of power from impacted histogram and heterogeneity features was compensated by the selection of additional shape features. The harmonization using a GAN-based approach outperformed the histogram matching technique, supporting the interest for the development of new advanced harmonization techniques for radiomic analysis purposes.

Keywords: harmonization; machine learning; radiomics.

PubMed Disclaimer

Publication types

LinkOut - more resources