Dataset Information

Analysis of heterogeneity in T2-weighted MR images can differentiate pseudoprogression from progression in glioblastoma.

ABSTRACT: To develop an image analysis technique that distinguishes pseudoprogression from true progression by analyzing tumour heterogeneity in T2-weighted images using topological descriptors of image heterogeneity called Minkowski functionals (MFs).Using a retrospective patient cohort (n = 50), and blinded to treatment response outcome, unsupervised feature estimation was performed to investigate MFs for the presence of outliers, potential confounders, and sensitivity to treatment response. The progression and pseudoprogression groups were then unblinded and supervised feature selection was performed using MFs, size and signal intensity features. A support vector machine model was obtained and evaluated using a prospective test cohort.The model gave a classification accuracy, using a combination of MFs and size features, of more than 85% in both retrospective and prospective datasets. A different feature selection method (Random Forest) and classifier (Lasso) gave the same results. Although not apparent to the reporting radiologist, the T2-weighted hyperintensity phenotype of those patients with progression was heterogeneous, large and frond-like when compared to those with pseudoprogression.Analysis of heterogeneity, in T2-weighted MR images, which are acquired routinely in the clinic, has the potential to detect an earlier treatment response allowing an early change in treatment strategy. Prospective validation of this technique in larger datasets is required.

SUBMITTER: Booth TC

PROVIDER: S-EPMC5435159 | biostudies-literature | 2017

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Analysis of heterogeneity in T2-weighted MR images can differentiate pseudoprogression from progression in glioblastoma.

Booth Thomas C TC Larkin Timothy J TJ Yuan Yinyin Y Kettunen Mikko I MI Dawson Sarah N SN Scoffings Daniel D Canuto Holly C HC Vowler Sarah L SL Kirschenlohr Heide H Hobson Michael P MP Markowetz Florian F Jefferies Sarah S Brindle Kevin M KM

PloS one 20170517 5

<h4>Purpose</h4>To develop an image analysis technique that distinguishes pseudoprogression from true progression by analyzing tumour heterogeneity in T2-weighted images using topological descriptors of image heterogeneity called Minkowski functionals (MFs).<h4>Methods</h4>Using a retrospective patient cohort (n = 50), and blinded to treatment response outcome, unsupervised feature estimation was performed to investigate MFs for the presence of outliers, potential confounders, and sensitivity to ...[more]

PMID: 28520730

Similar Datasets

Project description:OBJECTIVES:The extent of peritumoral tumor cell infiltrations in glioblastoma contributes to poor prognosis. We aimed to assess additive prognostic value of Minkowski functionals in analyzing heterogeneity of peritumoral hyperintensity on T2WI in glioblastoma patients. METHODS:Clinical data (age, sex, extent of surgical resection), O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation status and pre-operative T2WI of 113 pathologically confirmed glioblastoma patients (from our institution, n = 61; from the Cancer Imaging Archive, n = 52) were retrospectively reviewed. The patients were randomly grouped into a training set (n = 80) and a test set (n = 33). Peritumoral T2 hyperintensity was manually segmented and Minkowski functionals-a texture analysis method capturing heterogeneity of MR images-were computed as a function of 11 grayscale thresholds. The Cox proportional hazards models were fitted with clinical variables, Minkowski functionals features as well as both combined. The risk prediction performances of the Minkowski functionals and combined models were validated on a separate test dataset. The sex-specific survival difference of the entire cohort was analyzed according to MGMT methylation status via Kaplan-Meier survival curves. RESULTS:Thirty-three Minkowski features (11 area, 11 perimeter and 11 genus) for each patient were acquired giving a total of 3729 features. Cox regression models fitted with clinical data, Minkowski features, and both combined had incremental concordance indices of 0.577 (P = 0.02), 0.706 (P = 0.02) and 0.714 (P = 0.01) respectively. The prediction error rate of the combined model-having clinical and Minkowski features-was lower than that of Minkowski functionals model (0.135 and 0.161, respectively) when validated on a test dataset. No sex-specific survival difference was found according to MGMT methylation status (male, P = 0.2; female, P = 0.22). CONCLUSIONS:Minkowski functionals features computed from peritumoral hyperintensity can capture heterogeneity of glioblastoma on T2WI and have additive prognostic value in predicting survival, demonstrating their potential in complementing currently available prognostic parameters.

Project description:PurposeThe purpose of this study was to evaluate the agreement between paraspinal muscle composition measurements obtained from fat-water images using % fat-signal fraction (%FSF) in comparison to those obtained from T2-weighted magnetic resonance images (MRI) using a thresholding method.MethodsA sample of 35 subjects (19 females, 16 males; 40.26 ± 11.3 years old) was selected from a cohort of patients with chronic low back pain (LBP). Axial T2-weighted and IDEAL (Lava-Flex, 2 echo sequence) fat and water MR images were obtained using a 3.0 Tesla GE scanner. Multifidus, erector spinae, and psoas major muscle composition measurements were acquired bilaterally at L4-L5 and L5-S1 using both imaging sequences and related measurement methods. All measurements were obtained by the same rater, with a minimum of 7 days between each method. Intra-class correlation coefficients (ICCs) were calculated to assess intra-rater reliability. Pearson Correlation and Bland-Altman 95% limits of agreement were used to assess the agreement between both measurement methods.ResultsThe intra-rater reliability was excellent for all measurements with ICCs varying between 0.851 and 0.997. Strong positive correlations indicating a strong relationship between composition measurements were obtained from fat-water and T2-weighted images for bilateral multifidus and erector spinae muscles at both spinal levels and the right psoas major muscle at L4-L5, with correlation coefficient r ranging between 0.67 and 0.92. Bland-Altman plots for bilateral multifidus and erector spinae muscles at both levels revealed excellent agreement between the two methods, however, systematic differences between both methods were evident for psoas major fat measurements.ConclusionOur findings suggest that utilizing fat-water and T2-weighted MR images are comparable for quantifying multifidus and erector spinae muscle composition but not of the psoas major. While this suggests that both methods could be used interchangeably for the multifidus and erector spinae, further evaluation is required to expand and confirm our findings to other spinal levels.

Project description:PurposeTo evaluate T2-weighted MRI features to differentiate adrenal metastases from lipid-poor adenomas.Materials and methodsWith institutional review board approval, this study retrospectively compared 40 consecutive patients (mean age, 66 years ± 10 [standard deviation]) with metastases to 23 patients (mean age, 60 years ± 15) with lipid-poor adenomas at 1.5- and 3-T MRI between June 2016 and March 2019. A blinded radiologist measured T2-weighted signal intensity (SI) ratio (SInodule/SIpsoas muscle), T2-weighted histogram features, and chemical shift SI index. Two blinded radiologists (radiologist 1 and radiologist 2) assessed T2-weighted SI and T2-weighted heterogeneity using five-point Likert scales.ResultsSubjectively, T2-weighted SI (P < .001 for radiologist 1 and radiologist 2) and T2-weighted heterogeneity (P < .001, for radiologist 1 and radiologist 2) were higher in metastases compared with adenomas when assessed by both radiologists. Agreement between the radiologists was substantial for T2-weighted SI (Cohen κ = 0.67) and T2-weighted heterogeneity (κ = 0.62). Metastases had higher T2-weighted SI ratio than adenomas (3.6 ± 1.7 [95% confidence interval {CI}: 0.2, 8.2] vs 2.2 ± 1.0 [95% CI: 0.6, 4.3], P < .001) and higher T2-weighted entropy (6.6 ± 0.6 [95% CI: 4.9, 7.5] vs 5.0 ± 0.8 [95% CI: 3.5, 6.6], P < .001). At multivariate analysis, T2-weighted entropy was the best differentiating feature (P < .001). Chemical shift SI index did not differ between metastases and adenomas (P = .748). Area under the receiver operating characteristic curve (AUC) for T2-weighted SI ratio and T2-weighted entropy were 0.76 (95% CI: 0.64, 0.88) and 0.94 (95% CI: 0.88, 0.99). The logistic regression model combining T2-weighted SI ratio with T2-weighted entropy yielded AUC of 0.95 (95% CI: 0.91, 0.99) and did not differ compared with T2-weighted entropy alone (P = .268). There was no difference in logistic regression model accuracy comparing the data by either field strength, 1.5- or 3-T MRI (P > .05).ConclusionLogistic regression models combining T2-weighted SI and T2-weighted heterogeneity can differentiate metastases from lipid-poor adenomas. Validation of these preliminary results is required.Keywords: Adrenal, MR-Imaging, UrinarySupplemental material is available for this article.© RSNA, 2020.

Project description:We conducted a feasibility study to predict malignant glioma grades via radiomic analysis using contrast-enhanced T1-weighted magnetic resonance images (CE-T1WIs) and T2-weighted magnetic resonance images (T2WIs). We proposed a framework and applied it to CE-T1WIs and T2WIs (with tumor region data) acquired preoperatively from 157 patients with malignant glioma (grade III: 55, grade IV: 102) as the primary dataset and 67 patients with malignant glioma (grade III: 22, grade IV: 45) as the validation dataset. Radiomic features such as size/shape, intensity, histogram, and texture features were extracted from the tumor regions on the CE-T1WIs and T2WIs. The Wilcoxon-Mann-Whitney (WMW) test and least absolute shrinkage and selection operator logistic regression (LASSO-LR) were employed to select the radiomic features. Various machine learning (ML) algorithms were used to construct prediction models for the malignant glioma grades using the selected radiomic features. Leave-one-out cross-validation (LOOCV) was implemented to evaluate the performance of the prediction models in the primary dataset. The selected radiomic features for all folds in the LOOCV of the primary dataset were used to perform an independent validation. As evaluation indices, accuracies, sensitivities, specificities, and values for the area under receiver operating characteristic curve (or simply the area under the curve (AUC)) for all prediction models were calculated. The mean AUC value for all prediction models constructed by the ML algorithms in the LOOCV of the primary dataset was 0.902 ± 0.024 (95% CI (confidence interval), 0.873-0.932). In the independent validation, the mean AUC value for all prediction models was 0.747 ± 0.034 (95% CI, 0.705-0.790). The results of this study suggest that the malignant glioma grades could be sufficiently and easily predicted by preparing the CE-T1WIs, T2WIs, and tumor delineations for each patient. Our proposed framework may be an effective tool for preoperatively grading malignant gliomas.

Dataset Information

Analysis of heterogeneity in T2-weighted MR images can differentiate pseudoprogression from progression in glioblastoma.

Publications

Analysis of heterogeneity in T2-weighted MR images can differentiate pseudoprogression from progression in glioblastoma.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets