Search

Purpose: Predicting H3.1, TP53, and ACVR1 mutations in DIPG could aid in the selection of therapeutic options. The contribution of clinical data and multi-modal MRI were studied for these three predictive tasks. To keep the maximum number of subjects, which is essential for a rare disease, missing data were considered. A multi-modal model was proposed, collecting all available data for each patient, without performing any imputation., Methods: A retrospective cohort of 80 patients with confirmed DIPG and at least one of the four MR modalities (T1w, T1c, T2w, and FLAIR), acquired with two different MR scanners was built. A pipeline including standardization of MR data and extraction of radiomic features within the tumor was applied. The values of radiomic features between the two MR scanners were realigned using the ComBat method. For each prediction task, the most robust features were selected based on a recursive feature elimination with cross-validation. Five different models, one based on clinical data and one per MR modality, were developed using logistic regression classifiers. The prediction of the multi-modal model was defined as the average of all possible prediction results among five for each patient. The performances of the models were compared using a leave-one-out approach., Results: The percentage of missing modalities ranged from 6 to 11% across modalities and tasks. The performance of each individual model was dependent on each specific task, with an AUC of the ROC curve ranging from 0.63 to 0.80. The multi-modal model outperformed the clinical model for each prediction tasks, thus demonstrating the added value of MRI. Furthermore, regardless of performance criteria, the multi-modal model came in the first place or second place (very close to first). In the leave-one-out approach, the prediction of H3.1 (resp. ACVR1 and TP53) mutations achieved a balanced accuracy of 87.8% (resp. 82.1 and 78.3%)., Conclusion: Compared with a single modality approach, the multi-modal model combining multiple MRI modalities and clinical features was the most powerful to predict H3.1, ACVR1, and TP53 mutations and provided prediction, even in the case of missing modality. It could be proposed in the absence of a conclusive biopsy., Competing Interests: TE was employed by DOSIsoft SA. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2023 Khalid, Goya-Outi, Escobar, Dangouloff-Ros, Grigis, Philippe, Boddaert, Grill, Frouin and Frouin.)

Radiomics was proposed to identify tumor phenotypes non-invasively from quantitative imaging features. Calculating a large amount of information on images, allows the development of reliable classification models. In multi-modal imaging protocols, the question arises of adding an imaging modality to improve model performance. In addition, in the implementation of clinical protocols, some modalities are not acquired or are of insufficient quality and cannot be reliably taken into account. Furthermore, multi-scanner studies generate some variability in the acquisition and data. Some methodological solutions using ComBat and a multi-model approach were tested to take these two issues into account. It was applied to a cohort of 88 patients with Diffuse Intrinsic Pontine Glioma (DIPG). Sixteen models using radiomic features computed using 0, 1, 2, 3 or 4 MRI modalities were proposed. Based on Leave-One-Out Cross-Validation, F1 weighted scores ranged from 0.66 to 0.85. A model of majority voting using the prediction of all the models available for one given patient was finally applied, reducing drastically the number of unclassified patients.Clinical relevance- In case of patients with DIPG, the prediction of H3 mutation is of prime importance in case of inconclusive biopsy or in the absence of it. It could suggest orientations for new chemotherapy drugs associated with the radiation therapy.

Objective: Test a practical realignment approach to compensate the technical variability of MR radiomic features., Methods: T1 phantom images acquired on 2 scanners, FLAIR and contrast-enhanced T1-weighted (CE-T1w) images of 18 brain tumor patients scanned on both 1.5-T and 3-T scanners, and 36 T2-weighted (T2w) images of prostate cancer patients scanned in one of two centers were investigated. The ComBat procedure was used for harmonizing radiomic features. Differences in statistical distributions in feature values between 1.5- and 3-T images were tested before and after harmonization. The prostate studies were used to determine the impact of harmonization to distinguish between Gleason grades (GGs)., Results: In the phantom data, 40 out of 42 radiomic feature values were significantly different between the 2 scanners before harmonization and none after. In white matter regions, the statistical distributions of features were significantly different (p < 0.05) between the 1.5- and 3-T images for 37 out of 42 features in both FLAIR and CE-T1w images. After harmonization, no statistically significant differences were observed. In brain tumors, 41 (FLAIR) or 36 (CE-T1w) out of 42 features were significantly different between the 1.5- and 3-T images without harmonization, against 1 (FLAIR) or none (CE-T1w) with harmonization. In prostate studies, 636 radiomic features were significantly different between GGs after harmonization against 461 before. The ability to distinguish between GGs using radiomic features was increased after harmonization., Conclusion: ComBat harmonization efficiently removes inter-center technical inconsistencies in radiomic feature values and increases the sensitivity of studies using data from several scanners., Key Points: • Radiomic feature values obtained using different MR scanners or imaging protocols can be harmonized by combining off-the-shelf image standardization and feature realignment procedures. • Harmonized radiomic features enable one to pool data from different scanners and centers without a substantial loss of statistical power caused by intra- and inter-center variability. • The proposed realignment method is applicable to radiomic features from different MR sequences and tumor types and does not rely on any phantom acquisition.

Textural and shape analysis is gaining considerable interest in medical imaging, particularly to identify parameters characterizing tumor heterogeneity and to feed radiomic models. Here, we present a free, multiplatform, and easy-to-use freeware called LIFEx, which enables the calculation of conventional, histogram-based, textural, and shape features from PET, SPECT, MR, CT, and US images, or from any combination of imaging modalities. The application does not require any programming skills and was developed for medical imaging professionals. The goal is that independent and multicenter evidence of the usefulness and limitations of radiomic features for characterization of tumor heterogeneity and subsequent patient management can be gathered. Many options are offered for interactive textural index calculation and for increasing the reproducibility among centers. The software already benefits from a large user community (more than 800 registered users), and interactions within that community are part of the development strategy. Significance: This study presents a user-friendly, multi-platform freeware to extract radiomic features from PET, SPECT, MR, CT, and US images, or any combination of imaging modalities. Cancer Res; 78(16); 4786-9. ©2018 AACR ., (©2018 American Association for Cancer Research.)

Few methodological studies regarding widely used textural indices robustness in MRI have been reported. In this context, this study aims to propose some rules to compute reliable textural indices from multimodal 3D brain MRI. Diagnosis and post-biopsy MR scans including T1, post-contrast T1, T2 and FLAIR images from thirty children with diffuse intrinsic pontine glioma (DIPG) were considered. The hybrid white stripe method was adapted to standardize MR intensities. Sixty textural indices were then computed for each modality in different regions of interest (ROI), including tumor and white matter (WM). Three types of intensity binning were compared [Formula: see text]: constant bin width and relative bounds; [Formula: see text] constant number of bins and relative bounds; [Formula: see text] constant number of bins and absolute bounds. The impact of the volume of the region was also tested within the WM. First, the mean Hellinger distance between patient-based intensity distributions decreased by a factor greater than 10 in WM and greater than 2.5 in gray matter after standardization. Regarding the binning strategy, the ranking of patients was highly correlated for 188/240 features when comparing [Formula: see text] with [Formula: see text], but for only 20 when comparing [Formula: see text] with [Formula: see text], and nine when comparing [Formula: see text] with [Formula: see text]. Furthermore, when using [Formula: see text] or [Formula: see text] texture indices reflected tumor heterogeneity as assessed visually by experts. Last, 41 features presented statistically significant differences between contralateral WM regions when ROI size slightly varies across patients, and none when using ROI of the same size. For regions with similar size, 224 features were significantly different between WM and tumor. Valuable information from texture indices can be biased by methodological choices. Recommendations are to standardize intensities in MR brain volumes, to use intensity binning with constant bin width, and to define regions with the same volumes to get reliable textural indices.