A Framework for Harmonization of Radiomics Data for Multicenter Studies and Clinical Trials

Abstract

PURPOSE Variability in computed tomography images intrinsic to individual scanners limits the application of radiomics in clinical and research settings. The development of reproducible and generalizable radiomics-based models to assess lesions requires harmonization of data. The purpose of this study was to develop, test, and analyze the efficacy of a radiomics data harmonization model. MATERIALS AND METHODS Radiomic features from biopsy-proven untreated hepatic metastasis (N = 380) acquired from 167 unique patients with pancreatic, colon, and breast cancers were analyzed. Radiomic features from volume-match 551 samples of normal liver tissue and 188 hepatic cysts were included as references. A novel linear mixed effect model was used to identify effects associated with lesion size, tissue type, and scanner model. Six separate machine learning models were then used to test the effectiveness of radiomic feature harmonization using multivariate analysis. RESULTS Proposed model identifies and removes scanner-associated effects while preserving cancer-specific functional dependence of radiomic features on the tumor size. Data harmonization improves the performance of classification models by reducing the scanner-associated variability. For example, the multiclass logistic regression model, LogitBoost, demonstrated the improvement in sensitivity in the range from 15% to 40% for each type of liver metastasis, whereas the overall model accuracy and the kappa coefficient increased by 5% and 8% accordingly. CONCLUSION The model removed scanner-associated effects while preserving cancer-specific functional dependence of radiomic features.

Comparing quantitative image parameters between animal and clinical CT-scanners: a translational phantom study analysis

“…There are several studies in the literature that have addressed the problem of the variability of radiomics parameters (27)(28)(29). Mackin et al investigated the significance of inter-scanner variability of CT image-based radiomics studies (27).…”

Evaluation of parametrial infiltration in patients with IB‐IIB cervical cancer by a radiomics model integrating features from tumoral and peritumoral regions in 18F‐fluorodeoxy glucose positron emission tomography/MR images

“…Furthermore, this study was a single‐center study with a small sample size, in view of the high cost of PET/MRI. Considering the limitations that an individual scanner brings to the clinical radiomics application, 37 we will carry out multicenter studies with more patients to test the generalization and robustness of the model proposed in the current study.…”

Artificial Intelligence-Assisted Processing of Anterior Segment OCT Images in the Diagnosis of Vitreoretinal Lymphoma

“…Third, this was a single-centre study. Because of these limitations, we could not perform a harmonisation of our data because it was not possible to apply the methods described in the literature [65][66][67][68].…”