Gray-level discretization impacts reproducible MRI radiomics texture features

Abstract

Objectives: To assess the influence of gray-level discretization on inter- and intra-observer reproducibility of texture radiomics features on clinical MR images.

Materials and methods: We studied two independent MRI datasets of 74 lacrymal gland tumors and 30 breast lesions from two different centers. Two pairs of readers performed three two-dimensional delineations for each dataset. Texture features were extracted using two radiomics softwares (Pyradiomics and an in-house software). Reproducible features were selected using a combination of intra-class correlation coefficient (ICC) and concordance and coherence coefficient (CCC) with 0.8 and 0.9 as thresholds, respectively. We tested six absolute and eight relative gray-level discretization methods and analyzed the distribution and highest number of reproducible features obtained for each discretization. We also analyzed the number of reproducible features extracted from computer simulated delineations representative of inter-observer variability.

Results: The gray-level discretization method had a direct impact on texture feature reproducibility, independent of observers, software or method of delineation (simulated vs. human). The absolute discretization consistently provided statistically significantly more reproducible features than the relative discretization. Varying the bin number of relative discretization led to statistically significantly more variable results than varying the bin size of absolute discretization.

Conclusions: When considering inter-observer reproducible results of MRI texture radiomics features, an absolute discretization should be favored to allow the extraction of the highest number of potential candidates for new imaging biomarkers. Whichever the chosen method, it should be systematically documented to allow replicability of results.

Fig 1. Example of computer simulations of delineations.

The reference manual delineation is in red. Computer simulations with corresponding Dice coefficient are superimposed.

Fig 2. Flowchart of the study.

Reader 2.1 = Reader 2, delineation session 1; Reader 2.2 = Reader 2, delineation session 2; Reader 3.1 = Reader 3, delineation session 1; Reader 3.2 = Reader 3, delineation session 2; ROIs = Regions Of Interest; ICC = Intra-class correlation coefficient; CCC = Concordance correlation coefficient.

Fig 3. Highest number of reproducible texture features obtained for each experiment according to the discretization method.

FBS = Fixed Bin Size (absolute discretization); FBN = Fixed Bin Number (relative discretization). * 6 MR sequences; ** 1 MR sequence.

Fig 4. Highest number of reproducible texture features using simulated ROIs on DATASET 1 with the Pyradiomics software according to the Dice coefficient ranging from 1.0 to 0.65, for FBS versus FBN methods.

FBS = Fixed Bin Size (absolute discretization); FBN = Fixed Bin Number (relative discretization).