Differentiation of Treatment-Related Effects from Glioma Recurrence Using Machine Learning Classifiers Based Upon Pre-and Post-Contrast T1WI and T2 FLAIR Subtraction Features: A Two-Center Study

The efficacy of using a multiparametric magnetic resonance imaging-based radiomics model to distinguish glioma recurrence from pseudoprogression

OBJECTIVE

The early differential diagnosis of the postoperative recurrence or pseudoprogression (psPD) of a glioma is of great guiding significance for individualized clinical treatment. This study aimed to evaluate the ability of a multiparametric magnetic resonance imaging (MRI)-based radiomics model to distinguish between the postoperative recurrence and psPD of a glioma early on and in a noninvasive manner.

METHODS

A total of 52 patients with gliomas who attended the Hainan Provincial People's Hospital between 2000 and 2021 and met the inclusion criteria were selected for this study. 1137 and 1137 radiomic features were extracted from T1 enhanced and T2WI/FLAIR sequence images, respectively.After clearing some invalid information and LASSO screening, a total of 9 and 10 characteristic radiological features were extracted and randomly divided into the training set and the test set according to 7:3 ratio. Select-Kbest and minimum Absolute contraction and selection operator (LASSO) were used for feature selection. Support vector machine and logistic regression were used to form a multi-parameter model for training and prediction. The optimal sequence and classifier were selected according to the area under the curve (AUC) and accuracy.

RESULTS

Radiomic models 1, 2 and 3 based on T1WI, T2FLAIR and T1WI + T2T2FLAIR sequences have better performance in the identification of postoperative recurrence and false progression of T1 glioma. The performance of model 2 is more stable, and the performance of support vector machine classifier is more stable. The multiparameter model based on CE-T1 + T2WI/FLAIR sequence showed the best performance (AUC:0.96, sensitivity: 0.87, specificity: 0.94, accuracy: 0.89,95% CI:0.93-1).

CONCLUSION

The use of multiparametric MRI-based radiomics provides a noninvasive, stable, and accurate method for differentiating between the postoperative recurrence and psPD of a glioma, which allows for timely individualized clinical treatment.

Post-treatment imaging of gliomas: challenging the existing dogmas

Gliomas are the commonest malignant central nervous system tumours in adults and imaging is the cornerstone of diagnosis, treatment, and post-treatment follow-up of these patients. With the ever-evolving treatment strategies post-treatment imaging and interpretation in glioma remains challenging, more so with the advent of anti-angiogenic drugs and immunotherapy, which can significantly alter the appearance in this setting, thus making interpretation of routine imaging findings such as contrast enhancement, oedema, and mass effect difficult to interpret. This review details the various methods of management of glioma including the upcoming novel therapies and their impact on imaging findings, with a comprehensive description of the imaging findings in conventional and advanced imaging techniques. A systematic appraisal for the existing and emerging techniques of imaging in these settings and their clinical application including various response assessment guidelines and artificial intelligence based response assessment will also be discussed.

Multimodality MRI Radiomics Based on Machine Learning for Identifying True Tumor Recurrence and Treatment-Related Effects in Patients with Postoperative Glioma

INTRODUCTION

Conventional magnetic resonance imaging (MRI) features have difficulty distinguishing glioma true tumor recurrence (TuR) from treatment-related effects (TrE). We aimed to develop a machine-learning model based on multimodality MRI radiomics to help improve the efficiency of identifying glioma TuR.

METHODS

A total of 131 patients were enrolled and randomly divided into the training set (n = 91) and the test set (n = 40). Radiomic features were extracted from the postoperative enhancement (PoE) region and edema (ED) region from four routine MRI sequences. After analyses of Spearman's rank correlation coefficient, and least absolute shrinkage and selection operator, the key radiomic features were selected to construct support vector machine (SVM) and k-nearest neighbor (KNN) models. Decision curve analysis (DCA) and receiver operating characteristic (ROC) curves were used to analyze the performance.

RESULTS

The PoE model had a significantly higher area under curve (AUC) than the ED model (p < 0.05). Among the models constructed with a single sequence, the model using PoE regional features from CE-T1WI was superior to other models, with an AUC of 0.905 for SVM and 0.899 for KNN. In multimodality models, the PoE model outperformed the ED model with an AUC of 0.931 for SVM and 0.896 for KNN. The multimodality model, which combined routine sequences and the whole regional features, showed a slightly better performance with an AUC of 0.965 for SVM and 0.955 for KNN. Decision curve analysis showed the good clinical utility of multimodal radiomics models.

CONCLUSIONS

Multimodality radiomics can identify glioma TuR and TrE, potentially aiding clinical decision-making for individualized treatment. And edematous regions may provide useful information for recognizing recurrence.

RETROSPECTIVELY REGISTERED

2021.04.15, No:2020039.

Contrast enhanced FLAIR versus contrast enhanced T1W images in evaluation of intraparenchymal brain lesions

Background

Patients with suspected brain lesions are usually evaluated by means of intravenous contrast materials. These lesions may demonstrate enhancement through different mechanisms. At most institutions, CE-T1WI is the preferred sequence. FLAIR is a sort of inversion recovery pulse sequence with a long TR, TE and T1 and hence effectually nulls signals from CSF. The long T1 causes mild T effect and this result in lesion enhancement on post-contrast study. Therefore, lesions demonstrating enhancement on CE-T1WI will also demonstrate enhancement on CE-FLAIR images. The purpose of this work was to assess the role of CE-FLAIR versus CE-T1WI in evaluation of different intraparenchymal brain lesions.

Results

Comparing CE-T1WI to CE-FLAIR in various brain pathologies, both observers found higher sensitivity and specificity for lesion to background contrast ratio on CE-FLAIR comparing to CE-T1WI. Observer 1 found that lesion to background contrast ratio on CE-FLAIR had sensitivity of 71.4%, specificity of 66.7% and AUC of 0.661 versus 63.3% sensitivity, 58.3% specificity and 0.634 AUC for CE-T1WI. Observer 2 found that lesion to background contrast ratio on CE-FLAIR had sensitivity of 77.6%, specificity of 66.7% and AUC of 0.719 versus 61.2% sensitivity, 50% specificity and 0.628 AUC for CE-T1WI.

Conclusion

On comparing CE-FLAIR to CE-T1WI, CE-FLAIR display better lesion detection and enhancement also better soft tissue contrast resolution.

Multimodal MRI-Based Radiomic Nomogram for the Early Differentiation of Recurrence and Pseudoprogression of High-Grade Glioma

Objective

To evaluate the diagnostic value of multimodal MRI radiomics based on T2-weighted fluid attenuated inversion recovery imaging (T2WI-FLAIR) combined with T1-weighted contrast enhanced imaging (T1WI-CE) in the early differentiation of high-grade glioma recurrence from pseudoprogression.

Methods

A total of one hundred eighteen patients with brain gliomas who were diagnosed from March 2014 to April 2020 were retrospectively analyzed. According to the clinical characteristics, the patients were randomly split into a training group (n = 83) and a test group (n = 35) at a 7 : 3 ratio. The region of interest (ROI) was delineated, and 2632 radiomic features were extracted. We used multiple logistic regression to establish a classification model, including the T1 model, T2 model, and T1 + T2 model, to differentiate recurrence from pseudoprogression. The diagnostic efficiency of the model was evaluated by calculating the area under the receiver operating characteristic curve (AUC) and accuracy (ACC) and by analyzing the calibration curve of the nomogram and decision curve.

Results

There were 75 cases of recurrence and 43 cases of pseudoprogression. The diagnostic efficacies of the multimodal MRI-based radiomic model were relatively high. The AUC values and ACC of the training group were 0.831 and 77.11%, respectively, and the AUC values and ACC of the test group were 0.829 and 88.57%, respectively. The calibration curve of the nomogram showed that the discrimination probability was consistent with the actual occurrence in the training group, and the discrimination probability was roughly the same as the actual occurrence in the test group. In the decision curve analysis, the T1 + T2 model showed greater overall net efficiency.

Conclusion

The multimodal MRI radiomic model has relatively high efficiency in the early differentiation of recurrence from pseudoprogression, and it could be helpful for clinicians in devising correct treatment plans so that patients can be treated in a timely and accurate manner.

Tumor Progression and Treatment-Related Changes: Radiological Diagnosis Challenges for the Evaluation of Post Treated Glioma

Simple Summary

Glioma is the most common primary malignant tumor of the adult central nervous system. Despite aggressive multimodal treatment, its prognosis remains poor. During follow-up, it remains challenging to distinguish treatment-related changes from tumor progression in treated patients with gliomas due to both share clinical symptoms and morphological imaging characteristics (with new and/or increasing enhancing mass lesions). The early effective identification of tumor progression and treatment-related changes is of great significance for the prognosis and treatment of gliomas. We believe that advanced neuroimaging techniques can provide additional information for distinguishing both at an early stage. In this article, we focus on the research of magnetic resonance imaging technology and artificial intelligence in tumor progression and treatment-related changes. Finally, it provides new ideas and insights for clinical diagnosis.

Abstract

As the most common neuro-epithelial tumors of the central nervous system in adults, gliomas are highly malignant and easy to recurrence, with a dismal prognosis. Imaging studies are indispensable for tracking tumor progression (TP) or treatment-related changes (TRCs). During follow-up, distinguishing TRCs from TP in treated patients with gliomas remains challenging as both share similar clinical symptoms and morphological imaging characteristics (with new and/or increasing enhancing mass lesions) and fulfill criteria for progression. Thus, the early identification of TP and TRCs is of great significance for determining the prognosis and treatment. Histopathological biopsy is currently the gold standard for TP and TRC diagnosis. However, the invasive nature of this technique limits its clinical application. Advanced imaging methods (e.g., diffusion magnetic resonance imaging (MRI), perfusion MRI, magnetic resonance spectroscopy (MRS), positron emission tomography (PET), amide proton transfer (APT) and artificial intelligence (AI)) provide a non-invasive and feasible technical means for identifying of TP and TRCs at an early stage, which have recently become research hotspots. This paper reviews the current research on using the abovementioned advanced imaging methods to identify TP and TRCs of gliomas. First, the review focuses on the pathological changes of the two entities to establish a theoretical basis for imaging identification. Then, it elaborates on the application of different imaging techniques and AI in identifying the two entities. Finally, the current challenges and future prospects of these techniques and methods are discussed.

Machine learning imaging applications in the differentiation of true tumour progression from treatment‐related effects in brain tumours: A systematic review and meta‐analysis

Introduction

Chemotherapy and radiotherapy can produce treatment‐related effects, which may mimic tumour progression. Advances in Artificial Intelligence (AI) offer the potential to provide a more consistent approach of diagnosis with improved accuracy. The aim of this study was to determine the efficacy of machine learning models to differentiate treatment‐related effects (TRE), consisting of pseudoprogression (PsP) and radiation necrosis (RN), and true tumour progression (TTP).

Methods

The systematic review was conducted in accordance with PRISMA‐DTA guidelines. Searches were performed on PubMed, Scopus, Embase, Medline (Ovid) and ProQuest databases. Quality was assessed according to the PROBAST and CLAIM criteria. There were 25 original full‐text journal articles eligible for inclusion.

Results

For gliomas: PsP versus TTP (16 studies, highest AUC = 0.98), RN versus TTP (4 studies, highest AUC = 0.9988) and TRE versus TTP (3 studies, highest AUC = 0.94). For metastasis: RN vs. TTP (2 studies, highest AUC = 0.81). A meta‐analysis was performed on 9 studies in the gliomas PsP versus TTP group using STATA. The meta‐analysis reported a high sensitivity of 95.2% (95%CI: 86.6–98.4%) and specificity of 82.4% (95%CI: 67.0–91.6%).

Conclusion

TRE can be distinguished from TTP with good performance using machine learning‐based imaging models. There remain issues with the quality of articles and the integration of models into clinical practice. Future studies should focus on the external validation of models and utilize standardized criteria such as CLAIM to allow for consistency in reporting.

Treatment Response and Prognosis Evaluation in High-Grade Glioma: An Imaging Review Based on MRI

In recent years, the development of advanced magnetic resonance imaging (MRI) technology and machine learning (ML) have created new tools for evaluating treatment response and prognosis of patients with high-grade gliomas (HGG); however, patient prognosis has not improved significantly. This is mainly due to the heterogeneity between and within HGG tumors, resulting in standard treatment methods not benefitting all patients. Moreover, the survival of patients with HGG is not only related to tumor cells, but also to noncancer cells in the tumor microenvironment (TME). Therefore, during preoperative diagnosis and follow-up treatment of patients with HGG, noninvasive imaging markers are needed to characterize intratumoral heterogeneity, and then to evaluate treatment response and predict prognosis, timeously adjust treatment strategies, and achieve individualized diagnosis and treatment. In this review, we summarize the research progress of conventional MRI, advanced MRI technology, and ML in evaluation of treatment response and prognosis of patients with HGG. We further discuss the significance of the TME in the prognosis of HGG patients, associate imaging features with the TME, indirectly reflecting the heterogeneity within the tumor, and shifting treatment strategies from tumor cells alone to systemic therapy of the TME, which may be a major development direction in the future. LEVEL OF EVIDENCE: 5 TECHNICAL EFFICACY STAGE: 4.

Imaging Biomarkers of Glioblastoma Treatment Response: A Systematic Review and Meta-Analysis of Recent Machine Learning Studies

OBJECTIVE

Monitoring biomarkers using machine learning (ML) may determine glioblastoma treatment response. We systematically reviewed quality and performance accuracy of recently published studies.

METHODS

Following Preferred Reporting Items for Systematic Reviews and Meta-Analysis: Diagnostic Test Accuracy, we extracted articles from MEDLINE, EMBASE and Cochrane Register between 09/2018-01/2021. Included study participants were adults with glioblastoma having undergone standard treatment (maximal resection, radiotherapy with concomitant and adjuvant temozolomide), and follow-up imaging to determine treatment response status (specifically, distinguishing progression/recurrence from progression/recurrence mimics, the target condition). Using Quality Assessment of Diagnostic Accuracy Studies Two/Checklist for Artificial Intelligence in Medical Imaging, we assessed bias risk and applicability concerns. We determined test set performance accuracy (sensitivity, specificity, precision, F1-score, balanced accuracy). We used a bivariate random-effect model to determine pooled sensitivity, specificity, area-under the receiver operator characteristic curve (ROC-AUC). Pooled measures of balanced accuracy, positive/negative likelihood ratios (PLR/NLR) and diagnostic odds ratio (DOR) were calculated. PROSPERO registered (CRD42021261965).

RESULTS

Eighteen studies were included (1335/384 patients for training/testing respectively). Small patient numbers, high bias risk, applicability concerns (particularly confounding in reference standard and patient selection) and low level of evidence, allow limited conclusions from studies. Ten studies (10/18, 56%) included in meta-analysis gave 0.769 (0.649-0.858) sensitivity [pooled (95% CI)]; 0.648 (0.749-0.532) specificity; 0.706 (0.623-0.779) balanced accuracy; 2.220 (1.560-3.140) PLR; 0.366 (0.213-0.572) NLR; 6.670 (2.800-13.500) DOR; 0.765 ROC-AUC.

CONCLUSION

ML models using MRI features to distinguish between progression and mimics appear to demonstrate good diagnostic performance. However, study quality and design require improvement.

Research Progress of Gliomas in Machine Learning

In the field of gliomas research, the broad availability of genetic and image information originated by computer technologies and the booming of biomedical publications has led to the advent of the big-data era. Machine learning methods were applied as possible approaches to speed up the data mining processes. In this article, we reviewed the present situation and future orientations of machine learning application in gliomas within the context of workflows to integrate analysis for precision cancer care. Publicly available tools or algorithms for key machine learning technologies in the literature mining for glioma clinical research were reviewed and compared. Further, the existing solutions of machine learning methods and their limitations in glioma prediction and diagnostics, such as overfitting and class imbalanced, were critically analyzed.

Longitudinal structural and perfusion MRI enhanced by machine learning outperforms standalone modalities and radiological expertise in high-grade glioma surveillance

PURPOSE

Surveillance of patients with high-grade glioma (HGG) and identification of disease progression remain a major challenge in neurooncology. This study aimed to develop a support vector machine (SVM) classifier, employing combined longitudinal structural and perfusion MRI studies, to classify between stable disease, pseudoprogression and progressive disease (3-class problem).

METHODS

Study participants were separated into two groups: group I (total cohort: 64 patients) with a single DSC time point and group II (19 patients) with longitudinal DSC time points (2-3). We retrospectively analysed 269 structural MRI and 92 dynamic susceptibility contrast perfusion (DSC) MRI scans. The SVM classifier was trained using all available MRI studies for each group. Classification accuracy was assessed for different feature dataset and time point combinations and compared to radiologists' classifications.

RESULTS

SVM classification based on combined perfusion and structural features outperformed radiologists' classification across all groups. For the identification of progressive disease, use of combined features and longitudinal DSC time points improved classification performance (lowest error rate 1.6%). Optimal performance was observed in group II (multiple time points) with SVM sensitivity/specificity/accuracy of 100/91.67/94.7% (first time point analysis) and 85.71/100/94.7% (longitudinal analysis), compared to 60/78/68% and 70/90/84.2% for the respective radiologist classifications. In group I (single time point), the SVM classifier also outperformed radiologists' classifications with sensitivity/specificity/accuracy of 86.49/75.00/81.53% (SVM) compared to 75.7/68.9/73.84% (radiologists).

CONCLUSION

Our results indicate that utilisation of a machine learning (SVM) classifier based on analysis of longitudinal perfusion time points and combined structural and perfusion features significantly enhances classification outcome (p value= 0.0001).

Machine learning-based radiomic evaluation of treatment response prediction in glioblastoma

AIM

To investigate machine learning based models combining clinical, radiomic, and molecular information to distinguish between early true progression (tPD) and pseudoprogression (psPD) in patients with glioblastoma.

MATERIALS AND METHODS

A retrospective analysis was undertaken of 76 patients (46 tPD, 30 psPD) with early enhancing disease following chemoradiotherapy for glioblastoma. Outcome was determined on follow-up until 6 months post-chemoradiotherapy. Models comprised clinical characteristics, O⁶-methylguanine-DNA methyltransferase (MGMT) promoter methylation status, and 307 quantitative imaging features extracted from enhancing disease and perilesional oedema masks on early post-chemoradiotherapy contrast-enhanced T1-weighted imaging, T2-weighted imaging (T2WI), and apparent diffusion coefficient (ADC) maps. Feature selection was performed within bootstrapped cross-validated recursive feature elimination with a random forest algorithm. Naive Bayes five-fold cross-validation was used to validate the final model.

RESULTS

Top selected features included age, MGMT promoter methylation status, two shape-based features from the enhancing disease mask, three radiomic features from the enhancing disease mask on ADC, and one radiomic feature from the perilesional oedema mask on T2WI. The final model had an area under the receiver operating characteristics curve (AUC) of 0.80, sensitivity 78.2%, specificity 66.7%, and accuracy of 73.7%.

CONCLUSION

Incorporating a machine learning-based approach using quantitative radiomic features from standard-of-care magnetic resonance imaging (MRI), in combination with clinical characteristics and MGMT promoter methylation status has a complementary effect and improves model performance for early prediction of glioblastoma treatment response.

Meningiomas: Preoperative predictive histopathological grading based on radiomics of MRI

PURPOSE

We aimed to develop a radiomics model to predict the histopathological grading of meningiomas by magnetic resonance imaging (MRI) before surgery.

METHODS

We recruited 131 patients with pathological diagnosis of meningiomas. All the patients had undergone MRI before surgery on a 3.0 T MRI scanner to obtain T1 fluid- attenuated inversion recovery (T1 FLAIR) images, T2-weighted images (T2WI) and T1 FLAIR with contrast enhancement (CE-T1 FLAIR) images covering the whole brain. The removing features with low variance, univariate feature selection, and least absolute shrinkage and selection operator (LASSO) were used to select radiomics features. Six classifiers were used to train the models (logistic regression (LR), k-nearest neighbor (KNN), decision tree (DT), support vector machine (SVM), random forests (RF), and XGBoost), and then 24 models were established using a random verification method to differentiate low-grade from high-grade meningiomas. The performance was assessed by receiver-operating characteristic (ROC) analysis, the f1-score, sensitivity, and specificity.

RESULTS

The radiomics features were significantly associated with the histopathological grading. Quantitative imaging features (n = 1409) were extracted, and nine features were selected to predict the grades of meningiomas. The best performance of the radiomics model for the degree of differentiation was obtained by SVM (area under the curve (AUC), 0.956; 95% confidence interval (CI), 0.83-1.00; sensitivity, 0.87; specificity, 0.92; f1-score, 0.90).

CONCLUSION

The radiomics models are of great value in predicting the histopathological grades of meningiomas, and have broad prospects in radiology and clinics.

Subtraction Maps Derived from Longitudinal Magnetic Resonance Imaging in Patients with Glioma Facilitate Early Detection of Tumor Progression

Progression of glioma is frequently characterized by increases or enhanced spread of a hyperintensity in fluid attenuated inversion recovery (FLAIR) sequences. However, changes in FLAIR signal over time can be subtle, and conventional (CONV) visual reading is time-consuming. The purpose of this monocentric, retrospective study was to compare CONV reading to reading of subtraction maps (SMs) for serial FLAIR imaging. FLAIR datasets of cranial 3-Tesla magnetic resonance imaging (MRI), acquired at two different time points (mean inter-scan interval: 5.4 ± 1.9 months), were considered per patient in a consecutive series of 100 patients (mean age: 49.0 ± 13.7 years) diagnosed with glioma (19 glioma World Health Organization [WHO] grade I and II, 81 glioma WHO grade III and IV). Two readers (R1 and R2) performed CONV and SM reading by assessing overall image quality and artifacts, alterations in tumor-associated FLAIR signal over time (stable/unchanged or progressive) including diagnostic confidence (1-very high to 5-very low diagnostic confidence), and time needed for reading. Gold-standard (GS) reading, including all available clinical and imaging information, was performed by a senior reader, revealing progressive FLAIR signal in 61 patients (tumor progression or recurrence in 38 patients, pseudoprogression in 10 patients, and unclear in the remaining 13 patients). SM reading used an officially certified and commercially available algorithm performing semi-automatic coregistration, intensity normalization, and color-coding to generate individual SMs. The approach of SM reading revealed FLAIR signal increases in a larger proportion of patients according to evaluations of both readers (R1: 61 patients/R2: 60 patients identified with FLAIR signal increase vs. R1: 45 patients/R2: 44 patients for CONV reading) with significantly higher diagnostic confidence (R1: 1.29 ± 0.48, R2: 1.26 ± 0.44 vs. R1: 1.73 ± 0.80, R2: 1.82 ± 0.85; p < 0.0001). This resulted in increased sensitivity (99.9% vs. 73.3%) with maintained high specificity (98.1% vs. 98.8%) for SM reading when compared to CONV reading. Furthermore, the time needed for SM reading was significantly lower compared to CONV assessments (p < 0.0001). In conclusion, SM reading may improve diagnostic accuracy and sensitivity while reducing reading time, thus potentially enabling earlier detection of disease progression.