Whole-Tumor Histogram and Texture Analyses of DTI for Evaluation of IDH1-Mutation and 1p/19q-Codeletion Status in World Health Organization Grade II Gliomas

Machine learning based radiomics approach for outcome prediction of meningioma - a systematic review

Introduction

Meningioma is the most common brain tumor in adults. Magnetic resonance imaging (MRI) is the preferred imaging modality for assessing tumor outcomes. Radiomics, an advanced imaging technique, assesses tumor heterogeneity and identifies predictive markers, offering a non-invasive alternative to biopsies. Machine learning (ML) based radiomics models enhances diagnostic and prognostic accuracy of tumors. Comprehensive review on ML-based radiomics models for predicting meningioma recurrence and survival are lacking. Hence, the aim of the study is to summarize the performance measures of ML based radiomics models in the prediction of outcomes such as progression/recurrence (P/R) and overall survival analysis of meningioma.

Methods

Data bases such as Scopus, Web of Science, PubMed, and Embase were used to conduct a literature search in order to find pertinent original articles that concentrated on meningioma outcome prediction. PRISMA (Preferred reporting items for systematic reviews and meta-analysis) recommendations were used to extract data from selected studies.

Results

Eight articles were included in the study. MRI Radiomics-based models combined with clinical and pathological data showed strong predictive performance for meningioma recurrence. A decision tree model achieved 90% accuracy, outperforming an apparent diffusion coefficient (ADC) based model (83%). A support vector machine (SVM) model reached an area under curve (AUC) of 0.80 with radiomic features, improving to 0.88 with ADC integration. A combined clinico-pathological radiomics model (CPRM) achieved an AUC of 0.88 in testing. Key predictors of recurrence include ADC values, radiomic scores, ki-67 index, and Simpson grading. For predicting overall survival analysis of meningioma, the combined clinicopathological and radiomic features achieved an AUC of 0.78.

Conclusion

Integrating radiomics with clinical and pathological data through ML models greatly improved the outcome prediction for meningioma. These ML models surpass conventional MRI in predicting meningioma recurrence and aggressiveness, providing crucial insights for personalized treatment and surgical planning.

Multimodal MRI and 1H-MRS for Preoperative Stratification of High-Risk Molecular Subtype in Adult-Type Diffuse Gliomas

Isocitrate dehydrogenase (IDH) and O6-methylguanine-DNA methyltransferase (MGMT) genes are critical molecular markers in determining treatment options and predicting the prognosis of adult-type diffuse gliomas. Objectives: this study aimed to investigate whether multimodal MRI enables the differentiation of genotypes in adult-type diffuse gliomas. Methods: a total of 116 adult-type diffuse glioma patients (61 males, 51.5 (37, 62) years old) who underwent multimodal MRI before surgery were retrospectively analysed. Multimodal MRI included conventional MRI, proton magnetic resonance spectroscopy (1H-MRS), and diffusion tensor imaging (DTI). Conventional visual features, N-acetyl-aspartate (NAA)/Creatine (Cr), Choline (Cho)/Cr, Cho/NAA, fractional anisotropy (FA), mean diffusivity (MD), and diffusion histogram parameters were extracted on the whole tumour. Multimodal MRI parameters of IDH-mutant and IDH-wildtype gliomas were compared using the Mann-Whitney U test, Student's t-test, or Pearson chi-square tests. Logistic regression was used to select the MRI parameters to predict IDH-mutant gliomas. Furthermore, multimodal MRI parameters were selected to establish models for predicting MGMT methylation in the IDH-wildtype gliomas. The performance of models was evaluated by the receiver operating characteristics curve. Results: a total of 56 patients with IDH-mutant gliomas and 60 patients with IDH-wildtype glioblastomas (GBM) (37 with methylated MGMT and 17 with unmethylated MGMT) were diagnosed by 2021 WHO classification criteria. The enhancement degree (OR = 4.298, p < 0.001), necrosis/cyst (OR = 5.381, p = 0.011), NAA/Cr (OR = 0.497, p = 0.037), FA-Skewness (OR = 0.497, p = 0.033), MD-Skewness (OR = 1.849, p = 0.035), FAmean (OR = 1.924, p = 0.049) were independent factors for the multimodal combined prediction model in predicting IDH-mutant gliomas. The combined modal based on conventional MRI, 1H-MRS, DTI parameters, and histogram performed best in predicting IDH-wildtype status (AUC = 0.890). However, only NAA/Cr (OR = 0.17, p = 0.043) and FA (OR = 0.38, p = 0.015) were associated with MGMT methylated in IDH-wildtype GBM. The combination of NAA/Cr and FA-Median is more accurate for predicting MGMT methylation levels than using these elements alone (AUC, 0.847 vs. 0.695/0.684). Conclusions: multimodal MRI based on conventional MRI, 1H-MRS, and DTI can provide compound imaging markers for stratified individual diagnosis of IDH mutant and MGMT promoter methylation in adult-type diffuse gliomas.

Identification of schizophrenia by applying interpretable radiomics modeling with structural magnetic resonance imaging of the cerebellum

AIMS

The cerebellum is involved in higher-order mental processing as well as sensorimotor functions. Although structural abnormalities in the cerebellum have been demonstrated in schizophrenia, neuroimaging techniques are not yet applicable to identify them given the lack of biomarkers. We aimed to develop a robust diagnostic model for schizophrenia using radiomic features from T1-weighted magnetic resonance imaging (T1-MRI) of the cerebellum.

METHODS

A total of 336 participants (174 schizophrenia; 162 healthy controls [HCs]) were allocated to training (122 schizophrenia; 115 HCs) and test (52 schizophrenia; 47 HCs) cohorts. We obtained 2568 radiomic features from T1-MRI of the cerebellar subregions. After feature selection, a light gradient boosting machine classifier was trained. The discrimination and calibration of the model were evaluated. SHapley Additive exPlanations (SHAP) was applied to determine model interpretability.

RESULTS

We identified 17 radiomic features to differentiate participants with schizophrenia from HCs. In the test cohort, the radiomics model had an area under the curve, accuracy, sensitivity, and specificity of 0.89 (95% confidence interval: 0.82-0.95), 78.8%, 88.5%, and 75.4%, respectively. The model explanation by SHAP suggested that the second-order size zone non-uniformity feature from the right lobule IX and first-order energy feature from the right lobules V and VI were highly associated with the risk of schizophrenia.

CONCLUSION

The radiomics model focused on the cerebellum demonstrates robustness in diagnosing schizophrenia. Our results suggest that microcircuit disruption in the posterior cerebellum is a disease-defining feature of schizophrenia, and radiomics modeling has potential for supporting biomarker-based decision-making in clinical practice.

Predicting brain age gap with radiomics and automl: A Promising approach for age-Related brain degeneration biomarkers

The Brain Age Gap (BAG), which refers to the difference between chronological age and predicted neuroimaging age, is proposed as a potential biomarker for age-related brain degeneration. However, existing brain age prediction models usually rely on a single marker and can not discover meaningful hidden information in radiographic images. This study focuses on the application of radiomics, an advanced imaging analysis technique, combined with automated machine learning to predict BAG. Our methods achieve a promising result with a mean absolute error of 1.509 using the Alzheimer's Disease Neuroimaging Initiative dataset. Furthermore, we find that the hippocampus and parahippocampal gyrus play a significant role in predicting age with interpretable method called SHapley Additive exPlanations. Additionally, our investigation of age prediction discrepancies between patients with Alzheimer's disease (AD) and those with mild cognitive impairment (MCI) reveals a notable correlation with clinical cognitive assessment scale scores. This suggests that BAG has the potential to serve as a biomarker to support the diagnosis of AD and MCI. Overall, this study presents valuable insights into the application of neuroimaging models in the diagnosis of neurodegenerative diseases.

Predicting Isocitrate Dehydrogenase Status in Non-Contrast-Enhanced Adult-Type Astrocytic Tumors Using Diffusion Tensor Imaging and 11C-Methionine, 11C-Choline, and 18F-Fluorodeoxyglucose PET

We aimed to differentiate the isocitrate dehydrogenase (IDH) status among non-enhanced astrocytic tumors using preoperative MRI and PET. We analyzed 82 patients with non-contrast-enhanced, diffuse, supratentorial astrocytic tumors (IDH mutant [IDH-mut], 55 patients; IDH-wildtype [IDH-wt], 27 patients) who underwent MRI and PET between May 2012 and December 2022. We calculated the fractional anisotropy (FA) and mean diffusivity (MD) values using diffusion tensor imaging. We evaluated the tumor/normal brain uptake (T/N) ratios using 11C-methionine, 11C-choline, and 18F-fluorodeoxyglucose PET; extracted the parameters with significant differences in distinguishing the IDH status; and verified their diagnostic accuracy. Patients with astrocytomas were significantly younger than those with glioblastomas. The following MRI findings were significant predictors of IDH-wt instead of IDH-mut: thalamus invasion, contralateral cerebral hemisphere invasion, location adjacent to the ventricular walls, higher FA value, and lower MD value. The T/N ratio for all tracers was significantly higher for IDH-wt than for IDH-mut. In a composite diagnosis based on nine parameters, including age, 84.4% of cases with 0-4 points were of IDH-mut; conversely, 100% of cases with 6-9 points were of IDH-wt. Composite diagnosis using all parameters, including MRI and PET findings with significant differences, may help guide treatment decisions for early-stage gliomas.

Rethinking extent of resection of contrast-enhancing and non-enhancing tumor: different survival impacts on adult-type diffuse gliomas in 2021 World Health Organization classification

OBJECTIVES

Extent of resection (EOR) of contrast-enhancing (CE) and non-enhancing (NE) tumors may have different impacts on survival according to types of adult-type diffuse gliomas in the molecular era. This study aimed to evaluate the impact of EOR of CE and NE tumors in glioma according to the 2021 World Health Organization classification.

METHODS

This retrospective study included 1193 adult-type diffuse glioma patients diagnosed between 2001 and 2021 (183 oligodendroglioma, 211 isocitrate dehydrogenase [IDH]-mutant astrocytoma, and 799 IDH-wildtype glioblastoma patients) from a single institution. Patients had complete information on IDH mutation, 1p/19q codeletion, and O6-methylguanine-methyltransferase (MGMT) status. Cox survival analyses were performed within each glioma type to assess predictors of overall survival, including clinical, imaging data, histological grade, MGMT status, adjuvant treatment, and EOR of CE and NE tumors. Subgroup analyses were performed in patients with CE tumor.

RESULTS

Among 1193 patients, 935 (78.4%) patients had CE tumors. In entire oligodendrogliomas, gross total resection (GTR) of NE tumor was not associated with survival (HR = 0.56, p = 0.223). In 86 (47.0%) oligodendroglioma patients with CE tumor, GTR of CE tumor was the only independent predictor of survival (HR = 0.16, p = 0.004) in multivariable analysis. GTR of CE and NE tumors was independently associated with better survival in IDH-mutant astrocytoma and IDH-wildtype glioblastoma (all ps < 0.05).

CONCLUSIONS

GTR of both CE and NE tumors may significantly improve survival within IDH-mutant astrocytomas and IDH-wildtype glioblastomas. In oligodendrogliomas, the EOR of CE tumor may be crucial in survival; aggressive GTR of NE tumor may be unnecessary, whereas GTR of the CE tumor is recommended.

CLINICAL RELEVANCE STATEMENT

Surgical strategies on contrast-enhancing (CE) and non-enhancing (NE) tumors should be reassessed considering the different survival outcomes after gross total resection depending on CE and NE tumors in the 2021 World Health Organization classification of adult-type diffuse gliomas.

KEY POINTS

The survival impact of extent of resection of contrast-enhancing (CE) and non-enhancing (NE) tumors was evaluated in adult-type diffuse gliomas. Gross total resection of both CE and NE tumors may improve survival in isocitrate dehydrogenase (IDH)-mutant astrocytomas and IDH-wildtype glioblastomas, while only gross total resection of the CE tumor improves survival in oligodendrogliomas. Surgical strategies should be reconsidered according to types in adult-type diffuse gliomas.

Advancing noninvasive glioma classification with diffusion radiomics: Exploring the impact of signal intensity normalization

Background

This study investigates the influence of diffusion-weighted Magnetic Resonance Imaging (DWI-MRI) on radiomic-based prediction of glioma types according to molecular status and assesses the impact of DWI intensity normalization on model generalizability.

Methods

Radiomic features, compliant with image biomarker standardization initiative standards, were extracted from preoperative MRI of 549 patients with diffuse glioma, known IDH, and 1p19q-status. Anatomical sequences (T1, T1c, T2, FLAIR) underwent N4-Bias Field Correction (N4) and WhiteStripe normalization (N4/WS). Apparent diffusion coefficient (ADC) maps were normalized using N4 or N4/z-score. Nine machine-learning algorithms were trained for multiclass prediction of glioma types (IDH-mutant 1p/19q codeleted, IDH-mutant 1p/19q non-codeleted, IDH-wild type). Four approaches were compared: Anatomical, anatomical + ADC naive, anatomical + ADC N4, and anatomical + ADC N4/z-score. The University of California San Francisco (UCSF)-glioma dataset (n = 409) was used for external validation.

Results

Naïve-Bayes algorithms yielded overall the best performance on the internal test set. Adding ADC radiomics significantly improved AUC from 0.79 to 0.86 (P = .011) for the IDH-wild-type subgroup, but not for the other 2 glioma subgroups (P > .05). In the external UCSF dataset, the addition of ADC radiomics yielded a significantly higher AUC for the IDH-wild-type subgroup (P ≤ .001): 0.80 (N4/WS anatomical alone), 0.81 (anatomical + ADC naive), 0.81 (anatomical + ADC N4), and 0.88 (anatomical + ADC N4/z-score) as well as for the IDH-mutant 1p/19q non-codeleted subgroup (P < .012 each).

Conclusions

ADC radiomics can enhance the performance of conventional MRI-based radiomic models, particularly for IDH-wild-type glioma. The benefit of intensity normalization of ADC maps depends on the type and context of the used data.

Combined texture analysis of dynamic contrast-enhanced MRI with histogram analysis of diffusion kurtosis imaging for predicting IDH mutational status in gliomas

BACKGROUND

Non-invasive detection of isocitrate dehydrogenase (IDH) mutational status in gliomas is clinically meaningful for molecular stratification of glioma; however, it remains challenging.

PURPOSE

To investigate the usefulness of texture analysis (TA) of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and histogram analysis of diffusion kurtosis imaging (DKI) maps for evaluating IDH mutational status in gliomas.

MATERIAL AND METHODS

This retrospective study enrolled 84 patients with histologically confirmed gliomas, comprising IDH-mutant (n = 34) and IDH-wildtype (n = 50). TA was performed for the quantitative parameters derived by DCE-MRI. Histogram analysis was performed for the quantitative parameters derived by DKI. Unpaired Student's t-test was used to identify IDH-mutant and IDH-wildtype gliomas. Logistic regression and receiver operating characteristic (ROC) curve analyses were used to compare the diagnostic performance of each parameter and their combination for predicting the IDH mutational status in gliomas.

RESULTS

Significant statistical differences in the TA of DCE-MRI and histogram analysis of DKI were observed between IDH-mutant and IDH-wildtype gliomas (all P < 0.05). Using multivariable logistic regression, the entropy of Ktrans, skewness of Ve, and Kapp-90th had higher prediction potential for IDH mutations with areas under the ROC curve (AUC) of 0.915, 0.735, and 0.830, respectively. A combination of these analyses for the identification of IDH mutation improved the AUC to 0.978, with a sensitivity and specificity of 94.1% and 96.0%, respectively, which was higher than the single analysis (P < 0.05).

CONCLUSION

Integrating the TA of DCE-MRI and histogram analysis of DKI may help to predict the IDH mutational status.

The 2021 WHO Classification for Gliomas and Implications on Imaging Diagnosis: Part 1-Key Points of the Fifth Edition and Summary of Imaging Findings on Adult-Type Diffuse Gliomas

The fifth edition of the World Health Organization (WHO) classification of central nervous system tumors published in 2021 advances the role of molecular diagnostics in the classification of gliomas by emphasizing integrated diagnoses based on histopathology and molecular information and grouping tumors based on genetic alterations. Importantly, molecular biomarkers that provide important prognostic information are now a parameter for establishing tumor grades in gliomas. Understanding the 2021 WHO classification is crucial for radiologists for daily imaging interpretation as well as communication with clinicians. Although imaging features are not included in the 2021 WHO classification, imaging can serve as a powerful tool to impact the clinical practice not only prior to tissue confirmation but beyond. This review represents the first of a three-installment review series on the 2021 WHO classification for gliomas, glioneuronal tumors, and neuronal tumors and implications on imaging diagnosis. This Part 1 Review focuses on the major changes to the classification of gliomas and imaging findings on adult-type diffuse gliomas. EVIDENCE LEVEL: 3. TECHNICAL EFFICACY: Stage 3.

Advanced MR Techniques for Preoperative Glioma Characterization: Part 1

Preoperative clinical magnetic resonance imaging (MRI) protocols for gliomas, brain tumors with dismal outcomes due to their infiltrative properties, still rely on conventional structural MRI, which does not deliver information on tumor genotype and is limited in the delineation of diffuse gliomas. The GliMR COST action wants to raise awareness about the state of the art of advanced MRI techniques in gliomas and their possible clinical translation or lack thereof. This review describes current methods, limits, and applications of advanced MRI for the preoperative assessment of glioma, summarizing the level of clinical validation of different techniques. In this first part, we discuss dynamic susceptibility contrast and dynamic contrast-enhanced MRI, arterial spin labeling, diffusion-weighted MRI, vessel imaging, and magnetic resonance fingerprinting. The second part of this review addresses magnetic resonance spectroscopy, chemical exchange saturation transfer, susceptibility-weighted imaging, MRI-PET, MR elastography, and MR-based radiomics applications. Evidence Level: 3 Technical Efficacy: Stage 2.

Combining Multi-Shell Diffusion with Conventional MRI Improves Molecular Diagnosis of Diffuse Gliomas with Deep Learning

The WHO classification since 2016 confirms the importance of integrating molecular diagnosis for prognosis and treatment decisions of adult-type diffuse gliomas. This motivates the development of non-invasive diagnostic methods, in particular MRI, to predict molecular subtypes of gliomas before surgery. At present, this development has been focused on deep-learning (DL)-based predictive models, mainly with conventional MRI (cMRI), despite recent studies suggesting multi-shell diffusion MRI (dMRI) offers complementary information to cMRI for molecular subtyping. The aim of this work is to evaluate the potential benefit of combining cMRI and multi-shell dMRI in DL-based models. A model implemented with deep residual neural networks was chosen as an illustrative example. Using a dataset of 146 patients with gliomas (from grade 2 to 4), the model was trained and evaluated, with nested cross-validation, on pre-operative cMRI, multi-shell dMRI, and a combination of the two for the following classification tasks: (i) IDH-mutation; (ii) 1p/19q-codeletion; and (iii) three molecular subtypes according to WHO 2021. The results from a subset of 100 patients with lower grades gliomas (2 and 3 according to WHO 2016) demonstrated that combining cMRI and multi-shell dMRI enabled the best performance in predicting IDH mutation and 1p/19q codeletion, achieving an accuracy of 75 ± 9% in predicting the IDH-mutation status, higher than using cMRI and multi-shell dMRI separately (both 70 ± 7%). Similar findings were observed for predicting the 1p/19q-codeletion status, with the accuracy from combining cMRI and multi-shell dMRI (72 ± 4%) higher than from each modality used alone (cMRI: 65 ± 6%; multi-shell dMRI: 66 ± 9%). These findings remain when we considered all 146 patients for predicting the IDH status (combined: 81 ± 5% accuracy; cMRI: 74 ± 5%; multi-shell dMRI: 73 ± 6%) and for the diagnosis of the three molecular subtypes according to WHO 2021 (combined: 60 ± 5%; cMRI: 57 ± 8%; multi-shell dMRI: 56 ± 7%). Together, these findings suggest that combining cMRI and multi-shell dMRI can offer higher accuracy than using each modality alone for predicting the IDH and 1p/19q status and in diagnosing the three molecular subtypes with DL-based models.

Quality reporting of radiomics analysis in pituitary adenomas: promoting clinical translation

OBJECTIVE

To evaluate the quality of radiomics studies on pituitary adenoma according to the radiomics quality score (RQS) and Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD).

METHODS

PubMed MEDLINE and EMBASE were searched to identify radiomics studies on pituitary adenomas. From 138 articles, 20 relevant original research articles were included. Studies were scored based on RQS and TRIPOD guidelines.

RESULTS

Most included studies did not perform pre-processing; isovoxel resampling, signal intensity normalization, and N4 bias field correction were performed in only five (25%), eight (40%), and four (20%) studies, respectively. Only two (10%) studies performed external validation. The mean RQS and basic adherence rate were 2.8 (7.6%) and 26.6%, respectively. There was a low adherence rate for conducting comparison to "gold-standard" (20%), multiple segmentation (25%), and stating potential clinical utility (25%). No study stated the biological correlation, conducted a test-retest or phantom study, was a prospective study, conducted cost-effectiveness analysis, or provided open-source code and data, which resulted in low-level evidence. The overall adherence rate for TRIPOD was 54.6%, and it was low for reporting the title (5%), abstract (0%), explaining the sample size (10%), and suggesting a full prediction model (5%).

CONCLUSION

The radiomics reporting quality for pituitary adenoma is insufficient. Pre-processing is required for feature reproducibility and external validation is necessary. Feature reproducibility, clinical utility demonstration, higher evidence levels, and open science are required. Titles, abstracts, and full prediction model suggestions should be improved for transparent reporting.

ADVANCES IN KNOWLEDGE

Despite the rapidly increasing number of radiomics researches on pituitary adenoma, the quality of science in these researches is unknown. Our study indicates that the overall quality needs to be significantly improved in radiomics studies on pituitary adenoma, and since the concept of RQS and IBSI is still unfamiliar to clinicians and radiologist researchers, our study may help to reach higher technical and clinical impact in the future study.

An interpretable radiomics model to select patients for radiotherapy after surgery for WHO grade 2 meningiomas

Objectives

This study investigated whether radiomic features can improve the prediction accuracy for tumor recurrence over clinicopathological features and if these features can be used to identify high-risk patients requiring adjuvant radiotherapy (ART) in WHO grade 2 meningiomas.

Methods

Preoperative magnetic resonance imaging (MRI) of 155 grade 2 meningioma patients with a median follow-up of 63.8 months were included and allocated to training (n = 92) and test sets (n = 63). After radiomic feature extraction (n = 200), least absolute shrinkage and selection operator feature selection with logistic regression classifier was performed to develop two models: (1) a clinicopathological model and (2) a combined clinicopathological and radiomic model. The probability of recurrence using the combined model was analyzed to identify candidates for ART.

Results

The combined clinicopathological and radiomics model exhibited superior performance for the prediction of recurrence compared with the clinicopathological model in the training set (area under the curve [AUC] 0.78 vs. 0.67, P = 0.042), which was also validated in the test set (AUC 0.77 vs. 0.61, P = 0.192). In patients with a high probability of recurrence by the combined model, the 5-year progression-free survival was significantly improved with ART (92% vs. 57%, P = 0.024), and the median time to recurrence was longer (54 vs. 17 months after surgery).

Conclusions

Radiomics significantly contributes added value in predicting recurrence when integrated with the clinicopathological features in patients with grade 2 meningiomas. Furthermore, the combined model can be applied to identify high-risk patients who require ART.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13014-022-02090-7.

A Survey of Radiomics in Precision Diagnosis and Treatment of Adult Gliomas

Glioma is the most common primary malignant tumor of the adult central nervous system (CNS), which mostly shows invasive growth. In most cases, surgery is often difficult to completely remove, and the recurrence rate and mortality of patients are high. With the continuous development of molecular genetics and the great progress of molecular biology technology, more and more molecular biomarkers have been proved to have important guiding significance in the individualized diagnosis, treatment, and prognosis evaluation of glioma. With the updates of the World Health Organization (WHO) classification of tumors of the CNS in 2021, the diagnosis and treatment of glioma has entered the era of precision medicine in the true sense. Due to its ability to non-invasively achieve accurate identification of glioma from other intracranial tumors, and to predict the grade, genotyping, treatment response, and prognosis of glioma, which provides a scientific basis for the clinical application of individualized diagnosis and treatment model of glioma, radiomics has become a research hotspot in the field of precision medicine. This paper reviewed the research related to radiomics of adult gliomas published in recent years and summarized the research proceedings of radiomics in differential diagnosis, preoperative grading and genotyping, treatment and efficacy evaluation, and survival prediction of adult gliomas.

Adding radiomics to the 2021 WHO updates may improve prognostic prediction for current IDH-wildtype histological lower-grade gliomas with known EGFR amplification and TERT promoter mutation status

OBJECTIVES

To assess whether radiomic features could improve the accuracy of survival predictions of IDH-wildtype (IDHwt) histological lower-grade gliomas (LGGs) over clinicopathological features.

METHODS

Preoperative MRI data of 61 patients with IDHwt histological LGGs were included as the institutional training set. The test set consisted of 32 patients from The Cancer Genome Atlas. Radiomic features (n = 186) were extracted using conventional MRIs. The radiomics risk score (RRS) for overall survival (OS) was derived from the elastic net. Multivariable Cox regression analyses with clinicopathological features (including epidermal growth factor receptor [EGFR] amplification and telomerase reverse transcriptase promoter [TERTp] mutation status) and the RRS were performed. The integrated area under the receiver operating curves (iAUCs) from the models with and without the RRS were compared. The net reclassification index (NRI) for 1-year OS was also calculated. The prognostic value of the RRS was evaluated using the external validation set.

RESULTS

The RRS independently predicted OS (hazard ratio = 48.08; p = 0.001). Compared with the clinicopathological model alone, adding the RRS had a better OS prediction performance (iAUCs 0.775 vs. 0.910), which was internally validated (iAUCs 0.726 vs. 0.884, 1-year OS NRI = 0.497), and a similar trend was found on external validation (iAUCs 0.683 vs. 0.705, 1-year OS NRI = 0.733). The prognostic significance of the RRS was confirmed in the external validation set (p = 0.001).

CONCLUSIONS

Integrating radiomics with clinicopathological features (including EGFR amplification and TERTp mutation status) can improve survival prediction in patients with IDHwt LGGs.

KEY POINTS

• Radiomics risk score has the potential to improve survival prediction when added to clinicopathological features (iAUCs increased from 0.775 to 0.910). • NRIs for 1-year OS showed that the radiomics risk score had incremental value over the clinicopathological model. • The prognostic significance of the radiomics risk score was confirmed in the external validation set (p = 0.001).

An interpretable radiomics model for the diagnosis of panic disorder with or without agoraphobia using magnetic resonance imaging

BACKGROUND

Early and accurate diagnosis of panic disorder with or without agoraphobia (PDA) is crucial to reducing disease burden and individual suffering. However, its diagnosis is challenging for lack of validated biomarkers. This study aimed to investigate whether radiomic features extracted from T1-weighted images (T1) of major fear-circuit structures (amygdala, insula, and anterior cingulate cortex [ACC]) could differentiate patients with PDA from healthy controls (HCs).

METHODS

The 213 participants (93 PDA, 120 HCs) were allocated to training (n = 149) and test (n = 64) sets after undergoing magnetic resonance imaging. Radiomic features (n = 1498) were extracted from T1 of the studied structures. Machine learning models were trained after feature selection and then validated in the test set. SHapley Additive exPlanations (SHAP) explored the model interpretability.

RESULTS

We identified 29 radiomic features to differentiate participants with PDA from HCs. The area under the curve, accuracy, sensitivity, and specificity of the best performing radiomics model in the test set were 0.84 (95% confidence interval: 0.74-0.95), 81.3%, 75.0%, and 86.1%, respectively. The SHAP model explanation suggested that the energy features extracted from the bilateral long insula gyrus and central sulcus of the insula and right ACC were highly associated with the risk of PDA.

LIMITATIONS

This was a cross-sectional study with a relatively small sample size, and the causality of changes in radiomic features and their biological and clinical meanings remained to be elucidated.

CONCLUSIONS

Our findings suggest that radiomic features from the fear-circuit structures could unveil hidden microstructural aberrations underlying the pathogenesis of PDA that could help identify PDA.

Differentiation of progressive disease from pseudoprogression using MRI histogram analysis in patients with treated glioblastoma

PURPOSE

Conventional magnetic resonance imaging (MRI) technics are insufficient in the differentiation of tumor progression from post-treatment changes in patients with treated glioblastoma. Previous studies have suggested that histogram analysis is a useful tool in the assessment of treatment response in different cancer types. The aim of the study was to to evaluate the effectiveness of MRI histogram analysis in the differentiation of tumor progression from pseudoprogression in patients with treated glioblastoma.

METHODS

Forty-six patients with glioblastoma who newly developed enhancing lesions following chemoradiation treatment were included in this retrospective study. Histogram analysis was performed from new enhancing lesions on T1-weighted contrast-enhanced MRI. Histogram analysis findings of patients with progression (23) and pseudoprogression (23) were compared.

RESULTS

Mean, minimum, median, maximum, standard deviation, variance, entropy, skewness, uniformity values were found to be significantly higher in progressive disease (p < 0.05). A receiver-operating characteristic (ROC) curve analysis was performed for mean value, and area under the curve (AUC) was found as 0.975. When the threshold value was selected as 528.86, two groups could be differentiated with 95.7% sensitivity and 87.0% specificity.

CONCLUSION

MRI histogram analysis can be used for the differentiation of progressive disease from pseudoprogression.

A fully automatic multiparametric radiomics model for differentiation of adult pilocytic astrocytomas from glioblastomas

OBJECTIVES

To develop a fully automatic radiomics model to differentiate adult pilocytic astrocytomas (PA) from high-grade gliomas (HGGs).

METHODS

This retrospective study included 302 adult patients with PA (n = 62) or HGG (n = 240). The patients were randomly divided into training (n = 211) and test (n = 91) sets. Clinical data were obtained, and radiomic features (n = 372) were extracted from multiparametric MRI with automatic tumour segmentation. After feature selection with F-score, a Light Gradient Boosting Machine classifier with subsampling was trained to develop three models: (1) clinical model, (2) radiomics model, and (3) combined clinical and radiomics model. Human performance was also assessed. The performance of the classifier was validated in the test set. SHapley Additive exPlanations (SHAP) was applied to explore the interpretability of the model.

RESULTS

A total of 15 radiomic features were selected. In the test set, the combined clinical and radiomics model (area under the curve [AUC], 0.93) showed a significantly higher performance than the clinical model (AUC, 0.79, p = 0.037) and had a similar performance to the radiomics model (AUC, 0.92, p = 0.828). The combined clinical and radiomics model also showed a significantly higher performance than humans (AUC, 0.76-0.81, p < 0.05). The model explanation by SHAP suggested that lower intratumoural heterogeneity from T2-weighted images was highly associated with PA diagnosis.

CONCLUSIONS

The fully automatic combined clinical and radiomics model may be helpful for differentiating adult PAs from HGGs.

KEY POINTS

• Differentiating adult PAs from HGGs is challenging because PAs may manifest a large spectrum of imaging presentations, often including aggressive imaging features. • The fully automatic combined clinical and radiomics model showed a significantly higher performance than the clinical model or humans. • The model explanation by SHAP suggested that second-order features from T2-weighted imaging were important in diagnosis and might reflect the underlying pathophysiology that PAs have lesser tissue heterogeneity than HGGs.

An interpretable multiparametric radiomics model for the diagnosis of schizophrenia using magnetic resonance imaging of the corpus callosum

There is a growing need to develop novel strategies for the diagnosis of schizophrenia using neuroimaging biomarkers. We investigated the robustness of the diagnostic model for schizophrenia using radiomic features from T1-weighted and diffusion tensor images of the corpus callosum (CC). A total of 165 participants [86 schizophrenia and 79 healthy controls (HCs)] were allocated to training (N = 115) and test (N = 50) sets. Radiomic features of the CC subregions were extracted from T1-weighted, apparent diffusion coefficient (ADC), and fractional anisotropy (FA) images (N = 1605). Following feature selection, various combinations of classifiers were trained, and Bayesian optimization was adopted in the best performing classifier. Discrimination, calibration, and clinical utility of the model were assessed. An online calculator was constructed to offer the probability of having schizophrenia. SHapley Additive exPlanations (SHAP) was applied to explore the interpretability of the model. We identified 30 radiomic features to differentiate participants with schizophrenia from HCs. The Bayesian optimized model achieved the highest performance, with an area under the curve (AUC), accuracy, sensitivity, and specificity of 0.89 (95% confidence interval: 0.81-0.98), 80.0, 83.3, and 76.9%, respectively, in the test set. The final model offers clinical probability in an online calculator. The model explanation by SHAP suggested that second-order features from the posterior CC were highly associated with the risk of schizophrenia. The multiparametric radiomics model focusing on the CC shows its robustness for the diagnosis of schizophrenia. Radiomic features could be a potential source of biomarkers that support the biomarker-based diagnosis of schizophrenia and improve the understanding of its neurobiology.

Pre- and Post-Treatment Imaging of Primary Central Nervous System Tumors in the Molecular and Genetic Era

Recent advances in the molecular and genetic characterization of central nervous system (CNS) tumors have ushered in a new era of tumor classification, diagnosis, and prognostic assessment. In this emerging and rapidly evolving molecular genetic era, imaging plays a critical role in the preoperative diagnosis and surgical planning, molecular marker prediction, targeted treatment planning, and post-therapy assessment of CNS tumors. This review provides an overview of the current imaging methods relevant to the molecular genetic classification of CNS tumors. Specifically, we focused on 1) the correlates between imaging features and specific molecular genetic markers and 2) the post-therapy imaging used for therapeutic assessment.

Assessing Versatile Machine Learning Models for Glioma Radiogenomic Studies across Hospitals

Simple Summary

Radiogenomics enables prediction of the status and prognosis of patients using non-invasively obtained imaging data. Current machine learning (ML) methods used in radiogenomics require huge datasets, which involve the handling of large heterogeneous datasets from multiple cohorts/hospitals. In this study, two different glioma datasets were used to test various ML and image pre-processing methods to confirm whether the models trained on one dataset are universally applicable to other datasets. Our result suggested that the ML method that yielded the highest accuracy in a single dataset was likely to be overfitted. We demonstrated that implementation of standardization and dimension reduction procedures prior to classification, enabled the development of ML methods that are less affected by the multiple cohort difference. We advocate using caution in interpreting the results of radiogenomic studies of the training and testing datasets that are small or mixed, with a view to implementing practical ML methods in radiogenomics.

Abstract

Radiogenomics use non-invasively obtained imaging data, such as magnetic resonance imaging (MRI), to predict critical biomarkers of patients. Developing an accurate machine learning (ML) technique for MRI requires data from hundreds of patients, which cannot be gathered from any single local hospital. Hence, a model universally applicable to multiple cohorts/hospitals is required. We applied various ML and image pre-processing procedures on a glioma dataset from The Cancer Image Archive (TCIA, n = 159). The models that showed a high level of accuracy in predicting glioblastoma or WHO Grade II and III glioma using the TCIA dataset, were then tested for the data from the National Cancer Center Hospital, Japan (NCC, n = 166) whether they could maintain similar levels of high accuracy. Results: we confirmed that our ML procedure achieved a level of accuracy (AUROC = 0.904) comparable to that shown previously by the deep-learning methods using TCIA. However, when we directly applied the model to the NCC dataset, its AUROC dropped to 0.383. Introduction of standardization and dimension reduction procedures before classification without re-training improved the prediction accuracy obtained using NCC (0.804) without a loss in prediction accuracy for the TCIA dataset. Furthermore, we confirmed the same tendency in a model for IDH1/2 mutation prediction with standardization and application of dimension reduction that was also applicable to multiple hospitals. Our results demonstrated that overfitting may occur when an ML method providing the highest accuracy in a small training dataset is used for different heterogeneous data sets, and suggested a promising process for developing an ML method applicable to multiple cohorts.

Radiomics With Ensemble Machine Learning Predicts Dopamine Agonist Response in Patients With Prolactinoma

CONTEXT

Early identification of the response of prolactinoma patients to dopamine agonists (DA) is crucial in treatment planning.

OBJECTIVE

To develop a radiomics model using an ensemble machine learning classifier with conventional magnetic resonance images (MRIs) to predict the DA response in prolactinoma patients.

DESIGN

Retrospective study.

SETTING

Severance Hospital, Seoul, Korea.

PATIENTS

A total of 177 prolactinoma patients who underwent baseline MRI (109 DA responders and 68 DA nonresponders) were allocated to the training (n = 141) and test (n = 36) sets. Radiomic features (n = 107) were extracted from coronal T2-weighed MRIs. After feature selection, single models (random forest, light gradient boosting machine, extra-trees, quadratic discrimination analysis, and linear discrimination analysis) with oversampling methods were trained to predict the DA response. A soft voting ensemble classifier was used to achieve the final performance. The performance of the classifier was validated in the test set.

RESULTS

The ensemble classifier showed an area under the curve (AUC) of 0.81 [95% confidence interval (CI), 0.74-0.87] in the training set. In the test set, the ensemble classifier showed an AUC, accuracy, sensitivity, and specificity of 0.81 (95% CI, 0.67-0.96), 77.8%, 78.6%, and 77.3%, respectively. The ensemble classifier achieved the highest performance among all the individual models in the test set.

CONCLUSIONS

Radiomic features may be useful biomarkers to predict the DA response in prolactinoma patients.

Machine learning applications to neuroimaging for glioma detection and classification: An artificial intelligence augmented systematic review

Glioma is the most common primary intraparenchymal tumor of the brain and the 5-year survival rate of high-grade glioma is poor. Magnetic resonance imaging (MRI) is essential for detecting, characterizing and monitoring brain tumors but definitive diagnosis still relies on surgical pathology. Machine learning has been applied to the analysis of MRI data in glioma research and has the potential to change clinical practice and improve patient outcomes. This systematic review synthesizes and analyzes the current state of machine learning applications to glioma MRI data and explores the use of machine learning for systematic review automation. Various datapoints were extracted from the 153 studies that met inclusion criteria and analyzed. Natural language processing (NLP) analysis involved keyword extraction, topic modeling and document classification. Machine learning has been applied to tumor grading and diagnosis, tumor segmentation, non-invasive genomic biomarker identification, detection of progression and patient survival prediction. Model performance was generally strong (AUC = 0.87 ± 0.09; sensitivity = 0.87 ± 0.10; specificity = 0.0.86 ± 0.10; precision = 0.88 ± 0.11). Convolutional neural network, support vector machine and random forest algorithms were top performers. Deep learning document classifiers yielded acceptable performance (mean 5-fold cross-validation AUC = 0.71). Machine learning tools and data resources were synthesized and summarized to facilitate future research. Machine learning has been widely applied to the processing of MRI data in glioma research and has demonstrated substantial utility. NLP and transfer learning resources enabled the successful development of a replicable method for automating the systematic review article screening process, which has potential for shortening the time from discovery to clinical application in medicine.

Magnetic Resonance Imaging Parameters for Noninvasive Prediction of Epidermal Growth Factor Receptor Amplification in Isocitrate Dehydrogenase-Wild-Type Lower-Grade Gliomas: A Multicenter Study

BACKGROUND

The epidermal growth factor receptor (EGFR) amplification status of isocitrate dehydrogenase-wild-type (IDHwt) lower-grade gliomas (LGGs; grade II/III) is one of the key markers for diagnosing molecular glioblastoma. However, the association between EGFR status and imaging parameters is unclear.

OBJECTIVE

To identify noninvasive imaging parameters from diffusion-weighted and dynamic susceptibility contrast imaging for predicting the EGFR amplification status of IDHwt LGGs.

METHODS

A total of 86 IDHwt LGG patients with known EGFR amplification status (62 nonamplified and 24 amplified) from 3 tertiary institutions were included. Qualitative and quantitative imaging features, including histogram parameters from apparent diffusion coefficient (ADC), normalized cerebral blood volume (nCBV), and normalized cerebral blood flow (nCBF), were assessed. Univariable and multivariable logistic regression models were constructed.

RESULTS

On multivariable analysis, multifocal/multicentric distribution (odds ratio [OR] = 11.77, P = .006), mean ADC (OR = 0.01, P = .044), 5th percentile of ADC (OR = 0.01, P = .046), and 95th percentile of nCBF (OR = 1.24, P = .031) were independent predictors of EGFR amplification. The diagnostic performance of the model with qualitative imaging parameters increased significantly when quantitative imaging parameters were added, with areas under the curves of 0.81 and 0.93, respectively (P = .004).

CONCLUSION

The presence of multifocal/multicentric distribution patterns, lower mean ADC, lower 5th percentile of ADC, and higher 95th percentile of nCBF may be useful imaging biomarkers for EGFR amplification in IDHwt LGGs. Moreover, quantitative imaging biomarkers may add value to qualitative imaging parameters.

Quality assessment of meningioma radiomics studies: Bridging the gap between exploratory research and clinical applications

PURPOSE

To evaluate the quality of radiomics studies on meningiomas, using a radiomics quality score (RQS), Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD), and the Image Biomarker Standardization Initiative (IBSI).

METHODS

PubMed MEDLINE and Embase were searched to identify radiomics studies on meningiomas. Of 138 identified articles, 25 relevant original research articles were included. Studies were scored according to the RQS, TRIPOD guidelines, and items in IBSI.

RESULTS

Only four studies (16 %) performed external validation. The mean RQS was 5.6 out of 36 (15.4 %), and the basic adherence rate was 26.8 %. The adherence rate was low for stating biological correlation (4%), conducting calibration statistics (12 %), multiple segmentation (16 %), and stating potential clinical utility (16 %). None of the studies conducted a test‒retest or phantom study, stated a comparison to a 'gold standard', conducted prospective studies or cost-effectivity analysis, or opened code and data to the public, resulting in low RQS. The overall adherence rate for TRIPOD was 54.1 %, with low scores for reporting the title (4%), abstract (0%), blind assessment of the outcome (8%), and explaining the sample size (0%). According to IBSI items, only 6 (24 %), 6 (24 %), and 3 (12 %) studies performed N4 bias-field correction, isovoxel resampling, and grey-level discretization, respectively. No study performed skull stripping.

CONCLUSION

The quality of radiomics studies for meningioma is insufficient. Acknowledgement of RQS, TRIPOD, and IBSI reporting guidelines may improve the quality of meningioma radiomics studies and enable their clinical application.

Clinical and diffusion parameters may noninvasively predict TERT promoter mutation status in grade II meningiomas

BACKGROUND AND PURPOSE

Increasing evidence suggests that genomic and molecular markers need to be integrated in grading of meningioma. Telomerase reverse transcriptase promoter (TERTp) mutation is receiving attention due to its clinical relevance in the treatment of meningiomas. The predictive ability of conventional and diffusion MRI parameters for determining the TERTp mutation status in grade II meningiomas has yet been identified.

MATERIAL AND METHODS

In this study, 63 patients with surgically confirmed grade II meningiomas (56 TERTp wildtype, 7 TERTp mutant) were included. Conventional imaging features were qualitatively assessed. The maximum diameter, volume of the tumors and histogram parameters from the apparent diffusion coefficient (ADC) were assessed. Independent clinical and imaging risk factors for TERTp mutation were investigated using multivariable logistic regression. The discriminative value of the prediction models with and without imaging features was evaluated.

RESULTS

In the univariable regression, older age (odds ratio [OR] = 1.13, P = 0.005), larger maximum diameter (OR = 1.09, P = 0.023), larger volume (OR = 1.04, P = 0.014), lower mean ADC (OR = 0.02, P = 0.025), and lower ADC 10th percentile (OR = 0.01, P = 0.014) were predictors of TERTp mutation. In multivariable regression, age (OR = 1.13, P = 0.009) and ADC 10th percentile (OR = 0.01, P = 0.038) were independent predictors of variables for predicting the TERTp mutation status. The performance of the prediction model increased upon inclusion of imaging parameters (area under the curves of 0.86 and 0.91, respectively, without and with imaging parameters).

CONCLUSION

Older age and lower ADC 10th percentile may be useful parameters to predict TERTp mutation in grade II meningiomas.

Prediction of Lower Grade Insular Glioma Molecular Pathology Using Diffusion Tensor Imaging Metric-Based Histogram Parameters

Objectives

To explore whether a simplified lesion delineation method and a set of diffusion tensor imaging (DTI) metric-based histogram parameters (mean, 25th percentile, 75th percentile, skewness, and kurtosis) are efficient at predicting the molecular pathology status (MGMT methylation, IDH mutation, TERT promoter mutation, and 1p19q codeletion) of lower grade insular gliomas (grades II and III).

Methods

40 lower grade insular glioma patients in two medical centers underwent preoperative DTI scanning. For each patient, the entire abnormal area in their b-non (b0) image was defined as region of interest (ROI), and a set of histogram parameters were calculated for two DTI metrics, fractional anisotropy (FA) and mean diffusivity (MD). Then, we compared how these DTI metrics varied according to molecular pathology and glioma grade, with their predictive performance individually and jointly assessed using receiver operating characteristic curves. The reliability of the combined prediction was evaluated by the calibration curve and Hosmer and Lemeshow test.

Results

The mean, 25th percentile, and 75th percentile of FA were associated with glioma grade, while the mean, 25th percentile, 75th percentile, and skewness of both FA and MD predicted IDH mutation. The mean, 25th percentile, and 75th percentile of FA, and all MD histogram parameters significantly distinguished TERT promoter status. Similarly, all MD histogram parameters were associated with 1p19q status. However, none of the parameters analyzed for either metric successfully predicted MGMT methylation. The 25th percentile of FA yielded the highest prediction efficiency for glioma grade, IDH mutation, and TERT promoter mutation, while the 75th percentile of MD gave the best prediction of 1p19q codeletion. The combined prediction could enhance the discrimination of grading, IDH and TERT mutation, and also with a good fitness.

Conclusions

Overall, more invasive gliomas showed higher FA and lower MD values. The simplified ROI delineation method presented here based on the combination of appropriate histogram parameters yielded a more practical and efficient approach to predicting molecular pathology in lower grade insular gliomas. This approach could help clinicians to determine the extent of tumor resection required and reduce complications, enabling more precise treatment of insular gliomas in combination with radiotherapy and chemotherapy.

Differentiating Between Multiple Myeloma and Metastasis Subtypes of Lumbar Vertebra Lesions Using Machine Learning-Based Radiomics

Objective

To determine whether machine learning based on conventional magnetic resonance imaging (MRI) sequences have the potential for the differential diagnosis of multiple myeloma (MM), and different tumor metastasis lesions of the lumbar vertebra.

Methods

We retrospectively enrolled 107 patients newly diagnosed with MM and different metastasis of the lumbar vertebra. In total 60 MM lesions and 118 metastasis lesions were selected for training classifiers (70%) and subsequent validation (30%). Following segmentation, 282 texture features were extracted from both T1WI and T2WI images. Following regression analysis using the least absolute shrinkage and selection operator (LASSO) algorithm, the following machine learning models were selected: Support‐Vector Machine (SVM), K-Nearest Neighbor (KNN), Random Forest (RF), Artificial Neural Networks (ANN), and Naïve Bayes (NB) using 10-fold cross validation, and the performances were evaluated using a confusion matrix. Matthews correlation coefficient (MCC), sensitivity, specificity, and accuracy of the models were also calculated.

Results

To differentiate MM and metastasis, 13 features in the T1WI images and 9 features in the T2WI images were obtained. Among the 10 classifiers, the ANN classifier from the T2WI images achieved the best performance (MCC = 0.605) with accuracy, sensitivity, and specificity of 0.815, 0.879, and 0.790, respectively, in the validation cohort. To differentiate MM and metastasis subtypes, eight features in the T1WI images and seven features in the T2WI images were obtained. Among the 10 classifiers, the ANN classifier from the T2WI images achieved the best performance (MCC = 0.560, 0.412, 0.449), respectively, with accuracy = 0.648; sensitivity 0.714, 0.821, 0.897 and specificity 0.775, 0.600, 0.640 for the MM, lung, and other metastases, respectively, in the validation cohort.

Conclusions

Machine learning–based classifiers showed a satisfactory performance in differentiating MM lesions from those of tumor metastasis. While their value for distinguishing myeloma from different metastasis subtypes was moderate.

Differentiation of recurrent glioblastoma from radiation necrosis using diffusion radiomics with machine learning model development and external validation

The purpose of this study was to establish a high-performing radiomics strategy with machine learning from conventional and diffusion MRI to differentiate recurrent glioblastoma (GBM) from radiation necrosis (RN) after concurrent chemoradiotherapy (CCRT) or radiotherapy. Eighty-six patients with GBM were enrolled in the training set after they underwent CCRT or radiotherapy and presented with new or enlarging contrast enhancement within the radiation field on follow-up MRI. A diagnosis was established either pathologically or clinicoradiologically (63 recurrent GBM and 23 RN). Another 41 patients (23 recurrent GBM and 18 RN) from a different institution were enrolled in the test set. Conventional MRI sequences (T2-weighted and postcontrast T1-weighted images) and ADC were analyzed to extract 263 radiomic features. After feature selection, various machine learning models with oversampling methods were trained with combinations of MRI sequences and subsequently validated in the test set. In the independent test set, the model using ADC sequence showed the best diagnostic performance, with an AUC, accuracy, sensitivity, specificity of 0.80, 78%, 66.7%, and 87%, respectively. In conclusion, the radiomics models models using other MRI sequences showed AUCs ranging from 0.65 to 0.66 in the test set. The diffusion radiomics may be helpful in differentiating recurrent GBM from RN..

Qualitative and Quantitative MRI Analysis in IDH1 Genotype Prediction of Lower-Grade Gliomas: A Machine Learning Approach

Purpose

Preoperative prediction of isocitrate dehydrogenase 1 (IDH1) mutation in lower-grade gliomas (LGGs) is crucial for clinical decision-making. This study aimed to examine the predictive value of a machine learning approach using qualitative and quantitative MRI features to identify the IDH1 mutation in LGGs.

Materials and Methods

A total of 102 LGG patients were allocated to training (n = 67) and validation (n = 35) cohorts and were subject to Visually Accessible Rembrandt Images (VASARI) feature extraction (23 features) from conventional multimodal MRI and radiomics feature extraction (56 features) from apparent diffusion coefficient maps. Feature selection was conducted using the maximum Relevance Minimum Redundancy method and 0.632+ bootstrap method. A machine learning model to predict IDH1 mutation was then established using a random forest classifier. The predictive performance was evaluated using receiver operating characteristic (ROC) curves.

Results

After feature selection, the top 5 VASARI features were enhancement quality, deep white matter invasion, tumor location, proportion of necrosis, and T1/FLAIR ratio, and the top 10 radiomics features included 3 histogram features, 3 gray-level run-length matrix features, and 3 gray-level size zone matrix features and one shape feature. Using the optimal VASARI or radiomics feature sets for IDH1 prediction, the trained model achieved an area under the ROC curve (AUC) of 0.779 ± 0.001 or 0.849 ± 0.008 on the validation cohort, respectively. The fusion model that integrated outputs of both optimal VASARI and radiomics models improved the AUC to 0.879.

Conclusion

The proposed machine learning approach using VASARI and radiomics features can predict IDH1 mutation in LGGs.

Predicting Amyloid Pathology in Mild Cognitive Impairment Using Radiomics Analysis of Magnetic Resonance Imaging

BACKGROUND

Noninvasive identification of amyloid-β (Aβ) is important for better clinical management of mild cognitive impairment (MCI) patients.

OBJECTIVE

To investigate whether radiomics features in the hippocampus in MCI improve the prediction of cerebrospinal fluid (CSF) Aβ42 status when integrated with clinical profiles.

METHODS

A total of 407 MCI subjects from the Alzheimer's Disease Neuroimaging Initiative were allocated to training (n = 324) and test (n = 83) sets. Radiomics features (n = 214) from the bilateral hippocampus were extracted from magnetic resonance imaging (MRI). A cut-off of <192 pg/mL was applied to define CSF Aβ42 status. After feature selection, random forest with subsampling methods were utilized to develop three models with which to predict CSF Aβ42: 1) a radiomics model; 2) a clinical model based on clinical profiles; and 3) a combined model based on radiomics and clinical profiles. The prediction performances thereof were validated in the test set. A prediction model using hippocampus volume was also developed and validated.

RESULTS

The best-performing radiomics model showed an area under the curve (AUC) of 0.674 in the test set. The best-performing clinical model showed an AUC of 0.758 in the test set. The best-performing combined model showed an AUC of 0.823 in the test set. The hippocampal volume model showed a lower performance, with an AUC of 0.543 in the test set.

CONCLUSION

Radiomics models from MRI can help predict CSF Aβ42 status in MCI patients and potentially triage the patients for invasive and costly Aβ tests.

Radiomics Approach for Prediction of Recurrence in Non-Functioning Pituitary Macroadenomas

Objectives

A subset of non-functioning pituitary macroadenomas (NFPAs) may exhibit early progression/recurrence (P/R) after surgical resection. The purpose of this study was to apply radiomics in predicting P/R in NFPAs.

Methods

Only patients who had undergone preoperative MRI and postoperative MRI follow-ups for more than 1 year were included in this study. From September 2010 to December 2017, 50 eligible patients diagnosed with pathologically confirmed NFPAs were identified. Preoperative coronal T2WI and contrast-enhanced (CE) T1WI imaging were analyzed by computer algorithms. For each imaging sequence, 32 first-order features and 75 texture features were extracted. Support vector machine (SVM) classifier was utilized to evaluate the importance of extracted parameters, and the most significant three parameters were used to build the prediction model. The SVM score was calculated based on the three selected features.

Results

Twenty-eight patients exhibited P/R (28/50, 56%) after surgery. The median follow-up time was 38 months, and the median time to P/R was 20 months. Visual disturbance, hypopituitarism, extrasellar extension, compression of the third ventricle, large tumor height and volume, failed optic chiasmatic decompression, and high SVM score were more frequently encountered in the P/R group (p < 0.05). In multivariate Cox hazards analysis, symptoms of sex hormones, hypopituitarism, and SVM score were high risk factors for P/R (p < 0.05) with hazard ratios of 10.71, 2.68, and 6.88. The three selected radiomics features were T1 surface-to-volume radio, T1 GLCM-informational measure of correlation, and T2 NGTDM-coarseness. The radiomics predictive model shows 25 true positive, 16 true negative, 6 false positive, and 3 false negative cases, with an accuracy of 82% and AUC of 0.78 in differentiating P/R from non-P/R NFPAs. For SVM score, optimal cut-off value of 0.537 and AUC of 0.87 were obtained for differentiation of P/R. Higher SVM scores were associated with shorter progression-free survival (p < 0.001).

Conclusions

Our preliminary results showed that objective and quantitative MR radiomic features can be extracted from NFPAs. Pending more studies and evidence to support the findings, radiomics analysis of preoperative MRI may have the potential to offer valuable information in treatment planning for NFPAs.

Radiomics model predicts granulation pattern in growth hormone-secreting pituitary adenomas

PURPOSE

To investigate whether radiomic features from magnetic resonance image (MRI) can predict the granulation pattern of growth hormone (GH)-secreting pituitary adenoma patients.

METHODS

Sixty-nine pathologically proven acromegaly patients (densely granulated [DG] = 50, sparsely granulated [SG] = 19) were included. Radiomic features (n = 214) were extracted from contrast-enhancing and total tumor portions from T2-weighted (T2) MRIs. Imaging features were selected using a least absolute shrinkage and selection operator (LASSO) logistic regression model with fivefold cross-validation. Diagnostic performance for predicting granulation pattern was compared with that for qualitative T2 signal intensity assessment and T2 relative signal intensity (rSI) using the area under the receiver operating characteristics curve (AUC).

RESULTS

Four significant radiomic features from the contrast-enhancing tumor (1 from shape, 1 from first order feature, and 2 from second order features) were selected by LASSO for model construction. The radiomics model showed an AUC, accuracy, sensitivity, and specificity of 0.834 (95% confidence interval [CI] 0.738-0.930), 73.7%, 74.0%, and 73.9%, respectively. The radiomics model showed significantly better performance than the model using qualitative T2 signal intensity assessment (AUC 0.597 [95% CI 0.447-0.747], P = 0.009) and T2 rSI (AUC 0.647 [95% CI 0.523-0.759], P = 0.037).

CONCLUSION

Radiomic features may be useful biomarkers to differentiate granulation pattern of GH-secreting pituitary adenoma patients, and showed better performance than qualitative assessment or rSI evaluation.

Predicting Isocitrate Dehydrogenase (IDH) Mutation Status in Gliomas Using Multiparameter MRI Radiomics Features

BACKGROUND

Accurate and noninvasive detection of isocitrate dehydrogenase (IDH, including IDH1 and IDH2) status is clinically meaningful for molecular stratification of glioma, but remains challenging.

PURPOSE

To establish a model for classifying IDH status in gliomas based on multiparametric MRI.

STUDY TYPE

Retrospective, radiomics.

POPULATION

In all, 105 consecutive cases of grade II-IV glioma with 50 IDH1 or IDH2 mutant (IDHm) and 55 IDH wildtype (IDHw) were separated into a training cohort (n = 73) and a test cohort (n = 32).

FIELD STRENGTH/SEQUENCE

Contrast-enhanced T₁ -weighted (CE-T₁ W), T₂ -weighted (T₂ W), and arterial spin labeling (ASL) images were acquired at 3.0T.

ASSESSMENT

Two doctors manually labeled the volume of interest (VOI) on CE-T₁ W, then T₂ W and ASL were coregistered to CE-T₁ W. A total of 851 radiomics features were extracted on each VOI of three sequences. From the training cohort, all radiomics features with age and gender were processed by the Mann-Whitney U-test, Pearson test, and least absolute shrinkage and selection operator to obtain optimal feature groups to train support vector machine models. The accuracy and area under curve (AUC) of all models for classifying the IDH status were calculated on the test cohort. Two subtasks were performed to verify the efficiency of texture features and the Pearson test in IDH status classification, respectively.

STATISTICAL TESTS

The permutation test with Bonferroni correction; chi-square test.

RESULTS

The accuracy and AUC of the classifier, which combines the features of all three sequences, achieved 0.823 and 0.770 (P < 0.05), respectively. The best model established by texture features only had an AUC of 0.819 and an accuracy of 0.761. The best model established without the Pearson test got an AUC of 0.747 and an accuracy of 0.719.

DATA CONCLUSION

IDH genotypes of glioma can be identified by radiomics features from multiparameter MRI. The Pearson test improved the performance of the IDH classification models.

LEVEL OF EVIDENCE

4 TECHNICAL EFFICACY STAGE: 1.

Radiomics features of hippocampal regions in magnetic resonance imaging can differentiate medial temporal lobe epilepsy patients from healthy controls

To investigative whether radiomics features in bilateral hippocampi from MRI can identify temporal lobe epilepsy (TLE). A total of 131 subjects with MRI (66 TLE patients [35 right and 31 left TLE] and 65 healthy controls [HC]) were allocated to training (n = 90) and test (n = 41) sets. Radiomics features (n = 186) from the bilateral hippocampi were extracted from T1-weighted images. After feature selection, machine learning models were trained. The performance of the classifier was validated in the test set to differentiate TLE from HC and ipsilateral TLE from HC. Identical processes were performed to differentiate right TLE from HC (training set, n = 69; test set; n = 31) and left TLE from HC (training set, n = 66; test set, n = 30). The best-performing model for identifying TLE showed an AUC, accuracy, sensitivity, and specificity of 0.848, 84.8%, 76.2%, and 75.0% in the test set, respectively. The best-performing radiomics models for identifying right TLE and left TLE subgroups showed AUCs of 0.845 and 0.840 in the test set, respectively. In addition, multiple radiomics features significantly correlated with neuropsychological test scores (false discovery rate-corrected p-values < 0.05). The radiomics model from hippocampus can be a potential biomarker for identifying TLE.

Quality Reporting of Radiomics Analysis in Mild Cognitive Impairment and Alzheimer's Disease: A Roadmap for Moving Forward

Objective

To evaluate radiomics analysis in studies on mild cognitive impairment (MCI) and Alzheimer's disease (AD) using a radiomics quality score (RQS) system to establish a roadmap for further improvement in clinical use.

Materials and Methods

PubMed MEDLINE and EMBASE were searched using the terms ‘cognitive impairment’ or ‘Alzheimer’ or ‘dementia’ and ‘radiomic’ or ‘texture’ or ‘radiogenomic’ for articles published until March 2020. From 258 articles, 26 relevant original research articles were selected. Two neuroradiologists assessed the quality of the methodology according to the RQS. Adherence rates for the following six key domains were evaluated: image protocol and reproducibility, feature reduction and validation, biologic/clinical utility, performance index, high level of evidence, and open science.

Results

The hippocampus was the most frequently analyzed (46.2%) anatomical structure. Of the 26 studies, 16 (61.5%) used an open source database (14 from Alzheimer's Disease Neuroimaging Initiative and 2 from Open Access Series of Imaging Studies). The mean RQS was 3.6 out of 36 (9.9%), and the basic adherence rate was 27.6%. Only one study (3.8%) performed external validation. The adherence rate was relatively high for reporting the imaging protocol (96.2%), multiple segmentation (76.9%), discrimination statistics (69.2%), and open science and data (65.4%) but low for conducting test-retest analysis (7.7%) and biologic correlation (3.8%). None of the studies stated potential clinical utility, conducted a phantom study, performed cut-off analysis or calibration statistics, was a prospective study, or conducted cost-effectiveness analysis, resulting in a low level of evidence.

Conclusion

The quality of radiomics reporting in MCI and AD studies is suboptimal. Validation is necessary using external dataset, and improvements need to be made to feature reproducibility, feature selection, clinical utility, model performance index, and pursuits of a higher level of evidence.

Automated apparent diffusion coefficient analysis for genotype prediction in lower grade glioma: association with the T2-FLAIR mismatch sign

PURPOSE

The prognosis of lower grade glioma (LGG) patients depends (in large part) on both isocitrate dehydrogenase (IDH) gene mutation and chromosome 1p/19q codeletion status. IDH-mutant LGG without 1p/19q codeletion (IDHmut-Noncodel) often exhibit a unique imaging appearance that includes high apparent diffusion coefficient (ADC) values not observed in other subtypes. The purpose of this study was to develop an ADC analysis-based approach that can automatically identify IDHmut-Noncodel LGG.

METHODS

Whole-tumor ADC metrics, including fractional tumor volume with ADC > 1.5 × 10^-3mm²/s (V_ADC>1.5), were used to identify IDHmut-Noncodel LGG in a cohort of N = 134 patients. Optimal threshold values determined in this dataset were then validated using an external dataset containing N = 93 cases collected from The Cancer Imaging Archive. Classifications were also compared with radiologist-identified T2-FLAIR mismatch sign and evaluated concurrently to identify added value from a combined approach.

RESULTS

V_ADC>1.5 classified IDHmut-Noncodel LGG in the internal cohort with an area under the curve (AUC) of 0.80. An optimal threshold value of 0.35 led to sensitivity/specificity = 0.57/0.93. Classification performance was similar in the validation cohort, with V_ADC>1.5 ≥ 0.35 achieving sensitivity/specificity = 0.57/0.91 (AUC = 0.81). Across both groups, 37 cases exhibited positive T2-FLAIR mismatch sign-all of which were IDHmut-Noncodel. Of these, 32/37 (86%) also exhibited V_ADC>1.5 ≥ 0.35, as did 23 additional IDHmut-Noncodel cases which were negative for T2-FLAIR mismatch sign.

CONCLUSION

Tumor subregions with high ADC were a robust indicator of IDHmut-Noncodel LGG, with V_ADC>1.5 achieving > 90% classification specificity in both internal and validation cohorts. V_ADC>1.5 exhibited strong concordance with the T2-FLAIR mismatch sign and the combination of both parameters improved sensitivity in detecting IDHmut-Noncodel LGG.

Differentiating patients with schizophrenia from healthy controls by hippocampal subfields using radiomics

BACKGROUND

Accurately diagnosing schizophrenia is still challenging due to the lack of validated biomarkers. Here, we aimed to investigate whether radiomic features in bilateral hippocampal subfields from magnetic resonance images (MRIs) can differentiate patients with schizophrenia from healthy controls (HCs).

METHODS

A total of 152 participants with MRI (86 schizophrenia and 66 HCs) were allocated to training (n = 106) and test (n = 46) sets. Radiomic features (n = 642) from the bilateral hippocampal subfields processed with automatic segmentation techniques were extracted from T1-weighted MRIs. After feature selection, various combinations of classifiers (logistic regression, extra-trees, AdaBoost, XGBoost, or support vector machine) and subsampling were trained. The performance of the classifier was validated in the test set by determining the area under the curve (AUC). Furthermore, the association between selected radiomic features and clinical symptoms in schizophrenia was assessed.

RESULTS

Thirty radiomic features were identified to differentiate participants with schizophrenia from HCs. In the training set, the AUC exhibited poor to good performance (range: 0.683-0.861). The best performing radiomics model in the test set was achieved by the mutual information feature selection and logistic regression with an AUC, accuracy, sensitivity, and specificity of 0.821 (95% confidence interval 0.681-0.961), 82.1%, 76.9%, and 70%, respectively. Greater maximum values in the left cornu ammonis 1-3 subfield were associated with a higher severity of positive symptoms and general psychopathology in participants with schizophrenia.

CONCLUSION

Radiomic features from hippocampal subfields may be useful biomarkers for identifying schizophrenia.

Diffusion kurtosis imaging in evaluating gliomas: different region of interest selection methods on time efficiency, measurement repeatability, and diagnostic ability

OBJECTIVES

Comparing the diagnostic efficacy of diffusion kurtosis imaging (DKI) derived from different region of interest (ROI) methods in tumor parenchyma for grading and predicting IDH-1 mutation and 1p19q co-deletion status of glioma patients and correlating with their survival data.

METHODS

Sixty-six patients (29 females; median age, 45 years) with pathologically proved gliomas (low-grade gliomas, 36; high-grade gliomas, 30) were prospectively included, and their clinical data were collected. All patients underwent DKI examination. DKI maps of each metric were derived. Three groups of ROIs (ten spots, ROI-10s; three biggest tumor slices, ROI-3s; and whole-tumor parenchyma, ROI-whole) were manually drawn by two independent radiologists. The interobserver consistency, time spent, diagnostic efficacy, and survival analysis of DKI metrics based on these three ROI methods were analyzed.

RESULTS

The intraexaminer reliability for all parameters among these three ROI methods was good, and the time spent on ROI-10s was significantly less than that of the other two methods (p < 0.001). DKI based on ROI-10s demonstrated a slightly better diagnostic value than the other two ROI methods for grading and predicting the IDH-1 mutation status of glioma, whereas DKI metrics derived from ROI-10s performed much better than those of the ROI-3s and ROI-whole in identifying 1p19q co-deletion. In survival analysis, the model based on ROI-10s that included patient age and mean diffusivity showed the highest prediction value (C-index, 0.81).

CONCLUSIONS

Among the three ROI methods, the ROI-10s method had the least time spent and the best diagnostic value for a comprehensive evaluation of glioma. It is an effective way to process DKI data and has important application value in the clinical evaluation of glioma.

KEY POINTS

• The intraexaminer reliability for all DKI parameters among different ROI methods was good, and the time spent on ROI-10 spots was significantly less than the other two ROI methods. • DKI metrics derived from ROI-10 spots performed the best in ROI selection methods (ROI-10s, ten-spot ROIs; ROI-3s, three biggest tumor slices ROI; and ROI-whole, whole-tumor parenchyma ROI) for a comprehensive evaluation of glioma. • The ROI-10 spots method is an effective way to process DKI data and has important application value in the clinical evaluation of glioma.

Diffusion tensor and postcontrast T1-weighted imaging radiomics to differentiate the epidermal growth factor receptor mutation status of brain metastases from non-small cell lung cancer

PURPOSE

To assess whether the radiomic features of diffusion tensor imaging (DTI) and conventional postcontrast T1-weighted (T1C) images can differentiate the epidermal growth factor receptor (EGFR) mutation status in brain metastases from non-small cell lung cancer (NSCLC).

METHODS

A total of 99 brain metastases in 51 patients who underwent surgery or biopsy with underlying NSCLC and known EGFR mutation statuses (57 from EGFR wild type, 42 from EGFR mutant) were allocated to the training (57 lesions in 31 patients) and test (42 lesions in 20 patients) sets. Radiomic features (n = 526) were extracted from preoperative MR images including T1C and DTI. Radiomics classifiers were constructed by combinations of five feature selectors and four machine learning algorithms. The trained classifiers were validated on the test set, and the classifier performance was assessed by determining the area under the curve (AUC).

RESULTS

EGFR mutation status showed an overall discordance rate of 12% between the primary tumors and corresponding brain metastases. The best performing classifier was a combination of the tree-based feature selection and linear discriminant algorithm and 5 features were selected (1 from ADC, 2 from fractional anisotropy, and 2 from T1C images), resulting in an AUC, accuracy, sensitivity, and specificity of 0.73, 78.6%, 81.3%, and 76.9% in the test set, respectively.

CONCLUSIONS

Radiomics classifiers integrating multiparametric MRI parameters may have potential in differentiating the EGFR mutation status in brain metastases from NSCLC.

Extracting diffusion tensor fractional anisotropy and mean diffusivity from 3-direction DWI scans using deep learning

PURPOSE

To develop and evaluate machine-learning methods that reconstruct fractional anisotropy (FA) values and mean diffusivities (MD) from 3-direction diffusion MRI (dMRI) acquisitions.

METHODS

Two machine-learning models were implemented to map undersampled dMRI signals with high-quality FA and MD maps that were reconstructed from fully sampled DTI scans. The first model was a previously described multilayer perceptron (MLP), which maps signals and FA/MD values from a single voxel. The second was a convolutional neural network U-Net model, which maps dMRI slices to full FA/MD maps. Each method was trained on dMRI brain scans (N = 46), and reconstruction accuracies were compared with conventional linear-least-squares (LLS) reconstructions.

RESULTS

In an independent testing cohort (N = 20), 3-direction U-Net reconstructions had significantly lower absolute FA error than both 3-direction MLP (U-Net_3-dir : 0.06 ± 0.01 vs. MLP_3-dir : 0.08 ± 0.01, P < 1 × 10^-5 ) and 6-direction LLS (LLS_6-dir : 0.09 ± 0.03, P = 1 × 10^-5 ). The MD errors were not significantly different among 3-direction MLP (0.06 ± 0.01 × 10^-3 mm² /s), 3-direction U-Net (0.06 ± 0.01 × 10^-3 mm² /s), and 6-direction LLS (0.07 ± 0.02 × 10^-3 mm² /s, P > .1).

CONCLUSION

The proposed U-Net model reconstructed FA from 3-direction dMRI scans with improved accuracy compared with both a previously described MLP approach and LLS fitting from 6-direction scans. The MD reconstruction accuracies did not differ significantly between reconstructions.

Diffusion and perfusion MRI may predict EGFR amplification and the TERT promoter mutation status of IDH-wildtype lower-grade gliomas

OBJECTIVES

Epidermal growth factor receptor (EGFR) amplification and telomerase reverse transcriptase promoter (TERTp) mutation status of isocitrate dehydrogenase-wildtype (IDHwt) lower-grade gliomas (LGGs; grade II/III) are crucial for identifying IDHwt LGG with an aggressive clinical course. The purpose of this study was to assess whether parameters from diffusion tensor imaging, dynamic susceptibility contrast (DSC), and diffusion tensor imaging, dynamic contrast-enhanced imaging can predict the EGFR amplification and TERTp mutation status of IDHwt LGGs.

METHODS

A total of 49 patients with IDHwt LGGs with either known EGFR amplification (39 non-amplified, 10 amplified) or TERTp mutation (19 wildtype, 21 mutant) statuses underwent MRI. The mean ADC, fractional anisotropy (FA), normalized cerebral blood volume (nCBV), normalized cerebral blood flow (nCBF), volume transfer constant (K^trans), rate transfer coefficient (K_ep), extravascular extracellular volume fraction (V_e), and plasma volume fraction (V_p) values were assessed. Univariate and multivariate logistic regression models were constructed.

RESULTS

EGFR-amplified tumors showed lower mean ADC values than EGFR-non-amplified tumors (p = 0.019). Mean ADC was an independent predictor of EGFR amplification, with an AUC of 0.75. TERTp mutant tumors showed higher mean nCBV (p = 0.020), higher mean nCBF (p = 0.017), and higher mean V_p (p = 0.002) than TERTp wildtype tumors. With multivariate logistic regression, mean V_p was the independent predictor of TERTp mutation status, with an AUC of 0.85.

CONCLUSION

This exploratory pilot study shows that lower ADC values may be useful for prediction of EGFR amplification, whereas higher V_p values may be useful for prediction of the TERTp mutation status of IDHwt LGGs.

KEY POINTS

• EGFR amplification and TERTp mutation are key molecular markers that predict an aggressive clinical course of IDHwt LGGs. • EGFR-amplified tumors showed lower ADC values than EGFR-non-amplified tumors, suggesting higher cellularity. • TERTp mutant tumors showed a higher plasma volume fraction than TERTp wildtype tumors, suggesting higher vascular proliferation and tumor angiogenesis.

Precision Digital Oncology: Emerging Role of Radiomics-based Biomarkers and Artificial Intelligence for Advanced Imaging and Characterization of Brain Tumors

Advances in computerized image analysis and the use of artificial intelligence-based approaches for image-based analysis and construction of prediction algorithms represent a new era for noninvasive biomarker discovery. In recent literature, it has become apparent that radiologic images can serve as mineable databases that contain large amounts of quantitative features with potential clinical significance. Extraction and analysis of these quantitative features is commonly referred to as texture or radiomic analysis. Numerous studies have demonstrated applications for texture and radiomic characterization methods for assessing brain tumors to improve noninvasive predictions of tumor histologic characteristics, molecular profile, distinction of treatment-related changes, and prediction of patient survival. In this review, the current use and future potential of texture or radiomic-based approaches with machine learning for brain tumor image analysis and prediction algorithm construction will be discussed. This technology has the potential to advance the value of diagnostic imaging by extracting currently unused information on medical scans that enables more precise, personalized therapy; however, significant barriers must be overcome if this technology is to be successfully implemented on a wide scale for routine use in the clinical setting. Keywords: Adults and Pediatrics, Brain/Brain Stem, CNS, Computer Aided Diagnosis (CAD), Computer Applications-General (Informatics), Image Postprocessing, Informatics, Neural Networks, Neuro-Oncology, Oncology, Treatment Effects, Tumor Response Supplemental material is available for this article. © RSNA, 2020.

Radiomics MRI Phenotyping with Machine Learning to Predict the Grade of Lower-Grade Gliomas: A Study Focused on Nonenhancing Tumors

Objective

To assess whether radiomics features derived from multiparametric MRI can predict the tumor grade of lower-grade gliomas (LGGs; World Health Organization grade II and grade III) and the nonenhancing LGG subgroup.

Materials and Methods

Two-hundred four patients with LGGs from our institutional cohort were allocated to training (n = 136) and test (n = 68) sets. Postcontrast T1-weighted images, T2-weighted images, and fluid-attenuated inversion recovery images were analyzed to extract 250 radiomics features. Various machine learning classifiers were trained using the radiomics features to predict the glioma grade. The trained classifiers were internally validated on the institutional test set and externally validated on a separate cohort (n = 99) from The Cancer Genome Atlas (TCGA). Classifier performance was assessed by determining the area under the curve (AUC) from receiver operating characteristic curve analysis. An identical process was performed in the nonenhancing LGG subgroup (institutional training set, n = 73; institutional test set, n = 37; and TCGA cohort, n = 37) to predict the glioma grade.

Results

The performance of the best classifier was good in the internal validation set (AUC, 0.85) and fair in the external validation set (AUC, 0.72) to predict the LGG grade. For the nonenhancing LGG subgroup, the performance of the best classifier was good in the internal validation set (AUC, 0.82), but poor in the external validation set (AUC, 0.68).

Conclusion

Radiomics feature-based classifiers may be useful to predict LGG grades. However, radiomics classifiers may have a limited value when applied to the nonenhancing LGG subgroup in a TCGA cohort.

DTI histogram parameters correlate with the extent of myoinvasion and tumor type in endometrial carcinoma: a preliminary analysis

Background

Myoinvasion and tumor-type determines surgical planning in endometrial carcinoma.

Purpose

To evaluate whole tumor diffusion tensor imaging histogram texture parameters in evaluating myoinvasion and tumor type in endometrial carcinoma.

Material and Methods

Twenty-seven patients with endometrial carcinoma underwent diffusion tensor imaging on a 1.5-T MRI system using echo-planar imaging sequence with 0 and 700 s/mm2 b values. Whole tumor histogram parameters were obtained from fractional anisotropy, mean diffusivity maps. Mann–Whitney U test and receiver operating characteristic curve analyses were used

Results

The mean fractional anisotropy of tumors with no myoinvasion was significantly higher than tumors which underwent myoinvasion, suggesting higher anisotropy in tumors which did not invade the myometrium. Voxel-wise heterogeneity in distribution of fractional anisotropy and mean diffusivity was seen in the form of higher uniformity and lower entropy of tumors with superficial <50% myoinvasion versus >50% myoinvasion. Uniformity, entropy, and energy of voxel-wise fractional anisotropy distribution gave an area under the curve of 0.827, 0.821, and 0.796, respectively, in predicting the presence of deep myometrial invasion while energy, entropy, and uniformity of mean diffusivity distribution in tumor gave an area under the curve of 0.84, 0.815, and 0.809 respectively. Tumor type was predicted with an area under the curve of 0.747, 0.759, and 0.765 for the uniformity, energy, and entropy of voxel-wise fractional anisotropy distribution. A logistic regression combining all the important histogram parameters obtained 94% and 88% sensitivity and 88% and 80% specificity in predicting deep myoinvasion and tumor type, respectively.

Conclusion

Diffusion tensor histogram analysis can better characterize endometrial carcinomas and can be used as a quantitative marker of tumor behavior.

MR image phenotypes may add prognostic value to clinical features in IDH wild-type lower-grade gliomas

PURPOSE

To identify significant prognostic magnetic resonance imaging (MRI) features and their prognostic value when added to clinical features in patients with isocitrate dehydrogenase wild-type (IDHwt) lower-grade gliomas.

MATERIALS AND METHODS

Preoperative MR images of 158 patients (discovery set = 112, external validation set = 46) with IDHwt lower-grade gliomas (WHO grade II or III) were retrospectively analyzed using the Visually Accessible Rembrandt Images feature set. Radiologic risk scores (RRSs) for overall survival were derived from the least absolute shrinkage and selection operator and elastic net. Multivariable Cox regression analysis, including age, Karnofsky Performance score, extent of resection, WHO grade, and RRS, was performed. The added prognostic value of RRS was calculated by comparing the integrated area under the receiver operating characteristic curve (iAUC) between models with and without RRS.

RESULTS

The presence of cysts, pial invasion, and cortical involvement were favorable prognostic factors, while ependymal extension, multifocal or multicentric distribution, nonlobar location, proportion of necrosis > 33%, satellites, and eloquent cortex involvement were significantly associated with worse prognosis. RRS independently predicted survival and significantly enhanced model performance for survival prediction when integrated to clinical features (iAUC increased to 0.773-0.777 from 0.737), which was successfully validated on the validation set (iAUC increased to 0.805-0.830 from 0.735).

CONCLUSION

MRI features associated with prognosis in patients with IDHwt lower-grade gliomas were identified. RRSs derived from MRI features independently predicted survival and significantly improved performance of survival prediction models when integrated into clinical features.

KEY POINTS

• Comprehensive analysis of MRI features conveys prognostic information in patients with isocitrate dehydrogenase wild-type lower-grade gliomas. • Presence of cysts, pial invasion, and cortical involvement of the tumor were favorable prognostic factors. • Radiological phenotypes derived from MRI independently predict survival and have the potential to improve survival prediction when added to clinical features.

Comparison of Diagnostic Performance of Two-Dimensional and Three-Dimensional Fractal Dimension and Lacunarity Analyses for Predicting the Meningioma Grade

BACKGROUND

To compare the diagnostic performance of two-dimensional (2D) and three-dimensional (3D) fractal dimension (FD) and lacunarity features from MRI for predicting the meningioma grade.

METHODS

This retrospective study included 123 meningioma patients [90 World Health Organization (WHO) grade I, 33 WHO grade II/III] with preoperative MRI including post-contrast T1-weighted imaging. The 2D and 3D FD and lacunarity parameters from the contrast-enhancing portion of the tumor were calculated. Reproducibility was assessed with the intraclass correlation coefficient. Multivariable logistic regression analysis using 2D or 3D fractal features was performed to predict the meningioma grade. The diagnostic ability of the 2D and 3D fractal models were compared.

RESULTS

The reproducibility between observers was excellent, with intraclass correlation coefficients of 0.97, 0.95, 0.98, and 0.96 for 2D FD, 2D lacunarity, 3D FD, and 3D lacunarity, respectively. WHO grade II/III meningiomas had a higher 2D and 3D FD (p=0.003 and p<0.001, respectively) and higher 2D and 3D lacunarity (p=0.002 and p=0.006, respectively) than WHO grade I meningiomas. The 2D fractal model showed an area under the curve (AUC), accuracy, sensitivity, and specificity of 0.690 [95% confidence interval (CI) 0.581-0.799], 72.4%, 75.8%, and 64.4%, respectively. The 3D fractal model showed an AUC, accuracy, sensitivity, and specificity of 0.813 (95% CI 0.733-0.878), 82.9%, 81.8%, and 70.0%, respectively. The 3D fractal model exhibited significantly better diagnostic performance than the 2D fractal model (p<0.001).

CONCLUSION

The 3D fractal analysis proved superiority in diagnostic performance to 2D fractal analysis in grading meningioma.

Radiological differences between subtypes of WHO 2016 grade II-III gliomas: a systematic review and meta-analysis

BACKGROUND

Isocitrate dehydrogenase (IDH) mutation and 1p/19q-codeletion are oncogenetic alterations with a positive prognostic value for diffuse gliomas, especially grade II and III. Some studies have suggested differences in biological behavior as reflected by radiological characteristics. In this paper, the literature regarding radiological characteristics in grade II and III glioma subtypes was systematically evaluated and a meta-analysis was performed.

METHODS

Studies that addressed the relationship between conventional radiological characteristics and IDH mutations and/or 1p/19q-codeletions in newly diagnosed, grade II and III gliomas of adult patients were included. The "3-group analysis" compared radiological characteristics between the WHO 2016 glioma subtypes (IDH-mutant astrocytoma, IDH-wildtype astrocytoma, and oligodendroglioma), and the "2-group analysis" compared radiological characteristics between 1p/19q-codeleted gliomas and 1p/19q-intact gliomas.

RESULTS

Fourteen studies (3-group analysis: 670 cases, 2-group analysis: 1042 cases) were included. IDH-mutated astrocytomas showed more often sharp borders and less frequently contrast enhancement compared to IDH-wildtype astrocytomas. 1p/19q-codeleted gliomas had less frequently sharp borders, but showed a heterogeneous aspect, calcification, cysts, and edema more frequently. For the 1p/19q-codeleted gliomas, a sensitivity of 96% was found for heterogeneity and a specificity of 88.1% for calcification.

CONCLUSIONS

Significant differences in conventional radiological characteristics exist between the WHO 2016 glioma subtypes, which may reflect differences in biological behavior. However, the diagnostic value of the independent radiological characteristics is insufficient to reliably predict the molecular genetic subtype.

Diffusion- and perfusion-weighted MRI radiomics model may predict isocitrate dehydrogenase (IDH) mutation and tumor aggressiveness in diffuse lower grade glioma

OBJECTIVES

To determine whether diffusion- and perfusion-weighted MRI-based radiomics features can improve prediction of isocitrate dehydrogenase (IDH) mutation and tumor aggressiveness in lower grade gliomas (LGGs) METHODS: Radiomics features (n = 6472) were extracted from multiparametric MRI including conventional MRI, apparent diffusion coefficient (ADC), and normalized cerebral blood volume, acquired on 127 LGG patients with determined IDH mutation status and grade (WHO II or III). Radiomics models were constructed using machine learning-based feature selection and generalized linear model classifiers. Segmentation stability was calculated between two readers using concordance correlation coefficients (CCCs). Diagnostic performance to predict IDH mutation and tumor grade was compared between the multiparametric and conventional MRI radiomics models using the area under the receiver operating characteristics curve (AUC). The models were tested using a temporally independent validation set (n = 28).

RESULTS

The multiparametric MRI radiomics model was optimized with a random forest feature selector, with segmentation stability of a CCC threshold of 0.8. For IDH mutation, multiparametric MR radiomics showed similar performance (AUC 0.795) to the conventional radiomics model (AUC 0.729). In tumor grading, multiparametric model with ADC features showed higher performance (AUC 0.932) than the conventional model (AUC 0.555). The independent validation set showed the same trend with AUCs of 0.747 for IDH prediction and 0.819 for tumor grading with multiparametric MRI radiomics model.

CONCLUSION

Multiparametric MRI radiomics model showed improved diagnostic performance in tumor grading and comparable diagnostic performance in IDH mutation status, with ADC features playing a significant role.

KEY POINTS

• The multiparametric MRI radiomics model was comparable with conventional MRI radiomics model in predicting IDH mutation. • The multiparametric MRI radiomics model outperformed conventional MRI in glioma grading. • Apparent diffusion coefficient played an important role in glioma grading and predicting IDH mutation status using radiomics.