To carry out their activities, Research Teams of the Frédéric Joliot Institute for Life Sciences have developed high-profile technological platforms in many areas : biomedical imaging, structural biology, metabolomics, High-Throughput screening, level 3 microbiological safety laboratory...
All the news of the Institute of life sciences Frédéric Joliot
Scientific result | Medical imaging | Health ＆ life sciences
According to a study published in the Journal of Nuclear Medicine, Irène Buvat's Research Team (IMIV, SHFJ), in collaboration with the Vincent Frouin's one (UNATI / NeuroSpin) proposed an unprecedented imaging approach based on the use of the ComBat harmonization method derived from genomics. This method correctly estimates the "center" effect that affects the images and thus makes it possible to analyze together radiomic biomarkers of positron emission tomography (PET) images from different centers.
IntroductionSeveral reports have shown that radiomic feature values are affected by the acquisition and reconstruction parameters, thus hampering multicenter studies. We propose a method to standardize features measured from Positron Emission Tomography (PET) images obtained using different imaging protocols to remove the center effect while preserving patient-specific effects. MethodsPre-treatment 18F-FDG-PET images of patients with breast cancer were included. In Department A, 63 patients were scanned using a Gemini Time-Of-Flight-PET/Computed-Tomography scanner including 16 triple-negative lesions (TN). In Department B, 74 patients underwent a PET on a GE Discovery 690 Scanner including 15 TN lesions. PET images from Department A were also smoothed using a Gaussian filter to mimic data from a third Department called Department A-S. The primary tumor was segmented to get a tumor volume of interest (VOI) and a spherical VOI was also set in healthy liver tissue. Three Standardized Uptake Values (SUVs) and 6 textural features were computed in all VOI using LIFEx software. A harmonization method, ComBat, initially described for genomic data, was used to estimate the department effect based on the observed feature values. Feature distributions in each department were compared before and after harmonization. ResultsIn healthy liver tissue, the feature distributions were significantly different for 4 out of 9 features between Departments A and B, and for 6 out of 9 between Departments A and A-S (p<0.05, Wilcoxon's test). After ComBat, none of the 9 feature distributions significantly differ between two departments (p>0.1). The same trend was observed in tumors with a realignment of feature values between the departments after ComBat. Identification of TN lesions was largely enhanced after harmonization when the cut-off values were determined on data from one department and applied to data from the other department. ConclusionComBat is efficient at removing the multicenter effect for textural features and SUVs. The method is easy to use, retains biological variations not related to a center effect, and does not require any feature recalculation. Such a harmonization allows for multicenter studies, external validation of radiomic models or cut-off values, and should facilitate the use of radiomic models in clinical practice.Read the French version.
F Orlhac, S Boughdad, C Philippe, H Stalla-Bourdillon, C Nioche, L Champion, M Soussan, F Frouin, V Frouin and I Buvat. A post-reconstruction harmonization method for multicenter radiomic studies in PET (2018) J Nucl Med, sous presse http://dx.doi.org/10.2967/jnumed.117.199935
CEA is a French government-funded technological research organisation in four main areas: low-carbon energies, defense and security, information technologies and health technologies. A prominent player in the European Research Area, it is involved in setting up collaborative projects with many partners around the world.