Discriminating pseudoprogression and true progression in diffuse infiltrating glioma using multi-parametric MRI data through deep learning

Deep Learning-Based Techniques in Glioma Brain Tumor Segmentation Using Multi-Parametric MRI: A Review on Clinical Applications and Future Outlooks

This comprehensive review explores the role of deep learning (DL) in glioma segmentation using multiparametric magnetic resonance imaging (MRI) data. The study surveys advanced techniques such as multiparametric MRI for capturing the complex nature of gliomas. It delves into the integration of DL with MRI, focusing on convolutional neural networks (CNNs) and their remarkable capabilities in tumor segmentation. Clinical applications of DL-based segmentation are highlighted, including treatment planning, monitoring treatment response, and distinguishing between tumor progression and pseudo-progression. Furthermore, the review examines the evolution of DL-based segmentation studies, from early CNN models to recent advancements such as attention mechanisms and transformer models. Challenges in data quality, gradient vanishing, and model interpretability are discussed. The review concludes with insights into future research directions, emphasizing the importance of addressing tumor heterogeneity, integrating genomic data, and ensuring responsible deployment of DL-driven healthcare technologies. EVIDENCE LEVEL: N/A TECHNICAL EFFICACY: Stage 2.

CFINet: Cross-Modality MRI Feature Interaction Network for Pseudoprogression Prediction of Glioblastoma

Pseudoprogression (PSP) is a related reaction of glioblastoma treatment, and misdiagnosis can lead to unnecessary intervention. Magnetic resonance imaging (MRI) provides cross-modality images for PSP prediction studies. However, how to effectively use the complementary information between the cross-modality MRI to improve PSP prediction is still a challenging task. To address this challenge, we propose a cross-modality feature interaction network for PSP prediction. Firstly, we propose a triple-branch multi-scale module to extract low-order feature representations and a skip-connection multi-scale module to extract high-order feature representations. Then, a cross-modality interaction module based on attention mechanism is designed to make the complementary information between cross-modality MRI fully interact. Finally, the high-order cross-modality interaction information is fed into a multi-layer perceptron to achieve the PSP prediction task. We evaluate the proposed network on a private dataset with 52 subjects from Hunan Cancer Hospital and validate it on a private dataset with 30 subjects from Xiangya Hospital. The accuracy of our proposed network on the datasets is 0.954 and 0.929, respectively, which is better than most typical convolutional neural network and interaction methods.

Post-ictal changes presenting as late pseudoprogression on MRI and PET in a patient with diffuse glioma: Case report and brief literature review

Following completion of adjuvant radiation and chemotherapy imaging surveillance forms a major role in the management of diffuse gliomas. The primary role of imaging is to detect recurrences earlier than clinical symptomatology. Magnetic resonance imaging (MRI) is considered the gold standard in follow-up protocols owing to better soft tissue delineation and multiparametric nature. True recurrence can often mimic treatment-related changes, it is of paramount importance to differentiate between the two entities as the clinical course is divergent. Addition of functional sequences like perfusion, spectroscopy and metabolic imaging can provide further details into the microenvironment. In equivocal cases, a follow-up short interval imaging might be obtained to settle the diagnostic dilemma. Here, we present a patient with diagnosis of recurrent oligodendroglioma treated with adjuvant chemoradiation, presenting with seizures five years post-completion of chemotherapy for recurrence. On MRI, subtle new onset gyral thickening of the left frontal region with mild increase in perfusion and patchy areas of raised choline. FET-PET (fluoro-ethyltyrosine) showed an increased tumour-to-white matter (T/Wm) ratio favouring tumour recurrence. Based on discussion in a multi-disciplinary joint clinic, short interval follow-up MRI was undertaken at two months showing decrease in gyral thickening and resolution of enhancing areas in left frontal lobe. Repeat imaging one year later demonstrated stable disease status without further new imaging findings. Given the changes resolving completely without any anti-tumoral intervention, we conclude this to be peri-ictal pseudoprogression, being the second such case described in India.

Incidence and outcome of pseudoprogression after radiation therapy in glioblastoma patients: A cohort study

Background

Differentiating post-radiation MRI changes from progressive disease (PD) in glioblastoma (GBM) patients represents a major challenge. The clinical problem is two-sided; avoid termination of effective therapy in case of pseudoprogression (PsP) and continuation of ineffective therapy in case of PD. We retrospectively assessed the incidence, management, and prognostic impact of PsP and analyzed factors associated with PsP in a GBM patient cohort.

Methods

Consecutive GBM patients diagnosed in the South-Eastern Norway Health Region from 2015 to 2018 who had received RT and follow-up MRI were included. Tumor, patient, and treatment characteristics were analyzed in relationship to re-evaluated MRI examinations at 3 and 6 months post-radiation using Response Assessment in Neuro-Oncology criteria.

Results

A total of 284 patients were included in the study. PsP incidence 3 and 6 months post-radiation was 19.4% and 7.0%, respectively. In adjusted analyses, methylated O6-methylguanine-DNA methyltransferase (MGMT) promoter and the absence of neurological deterioration were associated with PsP at both 3 (p < .001 and p = .029, respectively) and 6 months (p = .045 and p = .034, respectively) post-radiation. For patients retrospectively assessed as PD 3 months post-radiation, there was no survival benefit of treatment change (p = .838).

Conclusions

PsP incidence was similar to previous reports. In addition to the previously described correlation of methylated MGMT promoter with PsP, we also found that absence of neurological deterioration significantly correlated with PsP. Continuation of temozolomide courses did not seem to compromise survival for patients with PD at 3 months post-radiation; therefore, we recommend continuing adjuvant temozolomide courses in case of inconclusive MRI findings.

Convolutional Neural Networks for Glioma Segmentation and Prognosis: A Systematic Review

Deep learning (DL) is poised to redefine the way medical images are processed and analyzed. Convolutional neural networks (CNNs), a specific type of DL architecture, are exceptional for high-throughput processing, allowing for the effective extraction of relevant diagnostic patterns from large volumes of complex visual data. This technology has garnered substantial interest in the field of neuro-oncology as a promising tool to enhance medical imaging throughput and analysis. A multitude of methods harnessing MRI-based CNNs have been proposed for brain tumor segmentation, classification, and prognosis prediction. They are often applied to gliomas, the most common primary brain cancer, to classify subtypes with the goal of guiding therapy decisions. Additionally, the difficulty of repeating brain biopsies to evaluate treatment response in the setting of often confusing imaging findings provides a unique niche for CNNs to help distinguish the treatment response to gliomas. For example, glioblastoma, the most aggressive type of brain cancer, can grow due to poor treatment response, can appear to grow acutely due to treatment-related inflammation as the tumor dies (pseudo-progression), or falsely appear to be regrowing after treatment as a result of brain damage from radiation (radiation necrosis). CNNs are being applied to separate this diagnostic dilemma. This review provides a detailed synthesis of recent DL methods and applications for intratumor segmentation, glioma classification, and prognosis prediction. Furthermore, this review discusses the future direction of MRI-based CNN in the field of neuro-oncology and challenges in model interpretability, data availability, and computation efficiency.

Artificial intelligence in neuroimaging of brain tumors: reality or still promise?

PURPOSE OF REVIEW

To provide an updated overview of artificial intelligence (AI) applications in neuro-oncologic imaging and discuss current barriers to wider clinical adoption.

RECENT FINDINGS

A wide variety of AI applications in neuro-oncologic imaging have been developed and researched, spanning tasks from pretreatment brain tumor classification and segmentation, preoperative planning, radiogenomics, prognostication and survival prediction, posttreatment surveillance, and differentiating between pseudoprogression and true disease progression. While earlier studies were largely based on data from a single institution, more recent studies have demonstrated that the performance of these algorithms are also effective on external data from other institutions. Nevertheless, most of these algorithms have yet to see widespread clinical adoption, given the lack of prospective studies demonstrating their efficacy and the logistical difficulties involved in clinical implementation.

SUMMARY

While there has been significant progress in AI and neuro-oncologic imaging, clinical utility remains to be demonstrated. The next wave of progress in this area will be driven by prospective studies measuring outcomes relevant to clinical practice and go beyond retrospective studies which primarily aim to demonstrate high performance.

Automated Prediction of Early Recurrence in Advanced Sinonasal Squamous Cell Carcinoma With Deep Learning and Multi-parametric MRI-based Radiomics Nomogram

RATIONALE AND OBJECTIVES

Preoperative prediction of the recurrence risk in patients with advanced sinonasal squamous cell carcinoma (SNSCC) is critical for individualized treatment. To evaluate the predictive ability of radiomics signature (RS) based on deep learning and multiparametric MRI for the risk of 2-year recurrence in advanced SNSCC.

MATERIALS AND METHODS

Preoperative MRI datasets were retrospectively collected from 265 SNSCC patients (145 recurrences) who underwent preoperative MRI, including T2-weighted (T2W), contrast-enhanced T1-weighted (T1c) sequences and diffusion-weighted (DW). All patients were divided into 165 training cohort and 70 test cohort. A deep learning segmentation model based on VB-Net was used to segment regions of interest (ROIs) for preoperative MRI and radiomics features were extracted from automatically segmented ROIs. Least absolute shrinkage and selection operator (LASSO) and logistic regression (LR) were applied for feature selection and radiomics score construction. Combined with meaningful clinicopathological predictors, a nomogram was developed and its performance was evaluated. In addition, X-title software was used to divide patients into high-risk or low-risk early relapse (ER) subgroups. Recurrence-free survival probability (RFS) was assessed for each subgroup.

RESULTS

The radiomics score, T stage, histological grade and Ki-67 predictors were independent predictors. The segmentation models of T2WI, T1c, and apparent diffusion coefficient (ADC) sequences achieved Dice coefficients of 0.720, 0.727, and 0.756, respectively, in the test cohort. RS-T2, RS-T1c and RS-ADC were derived from single-parameter MRI. RS-Combined (combined with T2WI, T1c, and ADC features) was derived from multiparametric MRI and reached area under curve (AUC) and accuracy of 0.854 (0.749-0.927) and 74.3% (0.624-0.840), respectively, in the test cohort. The calibration curve and decision curve analysis (DCA) illustrate its value in clinical practice. Kaplan-Meier analysis showed that the 2-year RFS rate for low-risk patients was significantly greater than that for high-risk patients in both the training and testing cohorts (p < 0.001).

CONCLUSION

Automated nomograms based on multi-sequence MRI help to predict ER in SNSCC patients preoperatively.

Pseudoprogression versus true progression in glioblastoma: what neurosurgeons need to know

Management of patients with glioblastoma (GBM) is complex and involves implementing standard therapies including resection, radiation therapy, and chemotherapy, as well as novel immunotherapies and targeted small-molecule inhibitors through clinical trials and precision medicine approaches. As treatments have advanced, the radiological and clinical assessment of patients with GBM has become even more challenging and nuanced. Advances in spatial resolution and both anatomical and physiological information that can be derived from MRI have greatly improved the noninvasive assessment of GBM before, during, and after therapy. Identification of pseudoprogression (PsP), defined as changes concerning for tumor progression that are, in fact, transient and related to treatment response, is critical for successful patient management. These temporary changes can produce new clinical symptoms due to mass effect and edema. Differentiating this entity from true tumor progression is a major decision point in the patient's management and prognosis. Providers may choose to start an alternative therapy, transition to a clinical trial, consider repeat resection, or continue with the current therapy in hopes of resolution. In this review, the authors describe the invasive and noninvasive techniques neurosurgeons need to be aware of to identify PsP and facilitate surgical decision-making.

Taming Glioblastoma in "Real Time": Integrating Multimodal Advanced Neuroimaging/AI Tools Towards Creating a Robust and Therapy Agnostic Model for Response Assessment in Neuro-Oncology

The highly aggressive nature of glioblastoma carries a dismal prognosis despite aggressive multimodal therapy. Alternative treatment regimens, such as immunotherapies, are known to intensify the inflammatory response in the treatment field. Follow-up imaging in these scenarios often mimics disease progression on conventional MRI, making accurate evaluation extremely challenging. To this end, revised criteria for assessment of treatment response in high-grade gliomas were successfully proposed by the RANO Working Group to distinguish pseudoprogression from true progression, with intrinsic constraints related to the postcontrast T1-weighted MRI sequence. To address these existing limitations, our group proposes a more objective and quantifiable "treatment agnostic" model, integrating into the RANO criteria advanced multimodal neuroimaging techniques, such as diffusion tensor imaging (DTI), dynamic susceptibility contrast-perfusion weighted imaging (DSC-PWI), dynamic contrast enhanced (DCE)-MRI, MR spectroscopy, and amino acid-based positron emission tomography (PET) imaging tracers, along with artificial intelligence (AI) tools (radiomics, radiogenomics, and radiopathomics) and molecular information to address this complex issue of treatment-related changes versus tumor progression in "real-time", particularly in the early posttreatment window. Our perspective delineates the potential of incorporating multimodal neuroimaging techniques to improve consistency and automation for the assessment of early treatment response in neuro-oncology.

Low-parameter supervised learning models can discriminate pseudoprogression and true progression in non-perfusion-based MRI

Discrimination of pseudoprogression and true progression is one challenge to the treatment of malignant gliomas. Although some techniques such as circulating tumor DNA (ctDNA) and perfusion-weighted imaging (PWI) demonstrate promise in distinguishing PsP from TP, we investigate robust and replicable alternatives to distinguish the two entities based on more widely-available media. In this study, we use low-parametric supervised learning techniques based on geographically-weighted regression (GWR) to investigate the utility of both conventional MRI sequences as well as a diffusion-weighted sequence (apparent diffusion coefficient or ADC) in the discrimination of PsP v TP. GWR applied to MRI modality pairs is a unique approach for small sample sizes and is a novel approach in this arena. From our analysis, all modality pairs involving ADC maps, and those involving post-contrast T1 regressed onto T2 showed potential promise. This work on ADC data adds to a growing body of research suggesting the predictive benefits of ADC, and suggests further research on the relationships between post-contrast T1 and T2.Clinical relevance- Few studies have investigated predictive potential of conventional MRI and ADC to detect PsP. Our study adds to the growing research on the topic and presents a new perspective to research by exploiting the utility of ADC in PsP v TP distinction. In addition, our GWR methodology for low-parametric supervised computer vision models demonstrates a unique approach for image processing of small sample sizes.

Performance of deep learning algorithms to distinguish high-grade glioma from low-grade glioma: A systematic review and meta-analysis

This study aims to evaluate deep learning (DL) performance in differentiating low- and high-grade glioma. Search online database for studies continuously published from 1st January 2015 until 16th August 2022. The random-effects model was used for synthesis, based on pooled sensitivity (SE), specificity (SP), and area under the curve (AUC). Heterogeneity was estimated using the Higgins inconsistency index (I²). 33 were ultimately included in the meta-analysis. The overall pooled SE and SP were 94% and 93%, with an AUC of 0.98. There was great heterogeneity in this field. Our evidence-based study shows DL achieves high accuracy in glioma grading. Subgroup analysis reveals several limitations in this field: 1) Diagnostic trials require standard method for data merging for AI; 2) small sample size; 3) poor-quality image preprocessing; 4) not standard algorithm development; 5) not standard data report; 6) different definition of HGG and LGG; and 7) poor extrapolation.

Emerging role of artificial intelligence in diagnosis, classification and clinical management of glioma

Glioma represents a dominant primary intracranial malignancy in the central nervous system. Artificial intelligence that mainly includes machine learning, and deep learning computational approaches, presents a unique opportunity to enhance clinical management of glioma through improving tumor segmentation, diagnosis, differentiation, grading, treatment, prediction of clinical outcomes (prognosis, and recurrence), molecular features, clinical classification, characterization of the tumor microenvironment, and drug discovery. A growing body of recent studies apply artificial intelligence-based models to disparate data sources of glioma, covering imaging modalities, digital pathology, high-throughput multi-omics data (especially emerging single-cell RNA sequencing and spatial transcriptome), etc. While these early findings are promising, future studies are required to normalize artificial intelligence-based models to improve the generalizability and interpretability of the results. Despite prominent issues, targeted clinical application of artificial intelligence approaches in glioma will facilitate the development of precision medicine of this field. If these challenges can be overcome, artificial intelligence has the potential to profoundly change the way patients with or at risk of glioma are provided with more rational care.

Diagnosis of Glioblastoma Multiforme Progression via Interpretable Structure-Constrained Graph Neural Networks

Glioblastoma multiforme (GBM) is the most common type of brain tumors with high recurrence and mortality rates. After chemotherapy treatment, GBM patients still show a high rate of differentiating pseudoprogression (PsP), which is often confused as true tumor progression (TTP) due to high phenotypical similarities. Thus, it is crucial to construct an automated diagnosis model for differentiating between these two types of glioma progression. However, attaining this goal is impeded by the limited data availability and the high demand for interpretability in clinical settings. In this work, we propose an interpretable structure-constrained graph neural network (ISGNN) with enhanced features to automatically discriminate between PsP and TTP. This network employs a metric-based meta-learning strategy to aggregate class-specific graph nodes, focus on meta-tasks associated with various small graphs, thus improving the classification performance on small-scale datasets. Specifically, a node feature enhancement module is proposed to account for the relative importance of node features and enhance their distinguishability through inductive learning. A graph generation constraint module enables learning reasonable graph structures to improve the efficiency of information diffusion while avoiding propagation errors. Furthermore, model interpretability can be naturally enhanced based on the learned node features and graph structures that are closely related to the classification results. Comprehensive experimental evaluation of our method demonstrated excellent interpretable results in the diagnosis of glioma progression. In general, our work provides a novel systematic GNN approach for dealing with data scarcity and enhancing decision interpretability. Our source codes will be released at https://github.com/SJTUBME-QianLab/GBM-GNN.

Radiomic Analysis to Predict Histopathologically Confirmed Pseudoprogression in Glioblastoma Patients

Purpose

Pseudoprogression mimicking recurrent glioblastoma remains a diagnostic challenge that may adversely confound or delay appropriate treatment or clinical trial enrollment. We sought to build a radiomic classifier to predict pseudoprogression in patients with primary isocitrate dehydrogenase wild type glioblastoma.

Methods and Materials

We retrospectively examined a training cohort of 74 patients with isocitrate dehydrogenase wild type glioblastomas with brain magnetic resonance imaging including dynamic contrast enhanced T1 perfusion before resection of an enhancing lesion indeterminate for recurrent tumor or pseudoprogression. A recursive feature elimination random forest classifier was built using nested cross-validation without and with O⁶-methylguanine-DNA methyltransferase status to predict pseudoprogression.

Results

A classifier constructed with cross-validation on the training cohort achieved an area under the receiver operating curve of 81% for predicting pseudoprogression. This was further improved to 89% with the addition of O⁶-methylguanine-DNA methyltransferase status into the classifier.

Conclusions

Our results suggest that radiomic analysis of contrast T1-weighted images and magnetic resonance imaging perfusion images can assist the prompt diagnosis of pseudoprogression. Validation on external and independent data sets is necessary to verify these advanced analyses, which can be performed on routinely acquired clinical images and may help inform clinical treatment decisions.

Multiparametric magnetic resonance imaging-derived deep learning network to determine ferroptosis-related gene signatures in gliomas

Introduction

Ferroptosis-related gene (FRG) signature is important for assessing novel therapeutic approaches and prognosis in glioma. We trained a deep learning network for determining FRG signatures using multiparametric magnetic resonance imaging (MRI).

Methods

FRGs of patients with glioma were acquired from public databases. FRG-related risk score stratifying prognosis was developed from The Cancer Genome Atlas (TCGA) and validated using the Chinese Glioma Genome Atlas. Multiparametric MRI-derived glioma images and the corresponding genomic information were obtained for 122 cases from TCGA and The Cancer Imaging Archive. The deep learning network was trained using 3D-Resnet, and threefold cross-validation was performed to evaluate the predictive performance.

Results

The FRG-related risk score was associated with poor clinicopathological features and had a high predictive value for glioma prognosis. Based on the FRG-related risk score, patients with glioma were successfully classified into two subgroups (28 and 94 in the high- and low-risk groups, respectively). The deep learning networks TC (enhancing tumor and non-enhancing portion of the tumor core) mask achieved an average cross-validation accuracy of 0.842 and an average AUC of 0.781, while the deep learning networks WT (whole tumor and peritumoral edema) mask achieved an average cross-validation accuracy of 0.825 and an average AUC of 0.781.

Discussion

Our findings indicate that FRG signature is a prognostic indicator of glioma. In addition, we developed a deep learning network that has high classification accuracy in automatically determining FRG signatures, which may be an important step toward the clinical translation of novel therapeutic approaches and prognosis of glioma.

Monitoring Therapeutic Responses to Silicified Cancer Cell Immunotherapy Using PET/MRI in a Mouse Model of Disseminated Ovarian Cancer

Current imaging approaches used to monitor tumor progression can lack the ability to distinguish true progression from pseudoprogression. Simultaneous metabolic 2-deoxy-2-[¹⁸F]fluoro-D-glucose ([¹⁸F]FDG) positron emission tomography (PET) and magnetic resonance imaging (MRI) offers new opportunities to overcome this challenge by refining tumor identification and monitoring therapeutic responses to cancer immunotherapy. In the current work, spatial and quantitative analysis of tumor burden were performed using simultaneous [¹⁸F]FDG-PET/MRI to monitor therapeutic responses to a novel silicified cancer cell immunotherapy in a mouse model of disseminated serous epithelial ovarian cancer. Tumor progression was validated by bioluminescence imaging of luciferase expressing tumor cells, flow cytometric analysis of immune cells in the tumor microenvironment, and histopathology. While PET demonstrated the presence of metabolically active cancer cells through [¹⁸F]FDG uptake, MRI confirmed cancer-related accumulation of ascites and tissue anatomy. This approach provides complementary information on disease status without a confounding signal from treatment-induced inflammation. This work provides a possible roadmap to facilitate accurate monitoring of therapeutic responses to cancer immunotherapies.

Artificial intelligence in the radiomic analysis of glioblastomas: A review, taxonomy, and perspective

Radiological imaging techniques, including magnetic resonance imaging (MRI) and positron emission tomography (PET), are the standard-of-care non-invasive diagnostic approaches widely applied in neuro-oncology. Unfortunately, accurate interpretation of radiological imaging data is constantly challenged by the indistinguishable radiological image features shared by different pathological changes associated with tumor progression and/or various therapeutic interventions. In recent years, machine learning (ML)-based artificial intelligence (AI) technology has been widely applied in medical image processing and bioinformatics due to its advantages in implicit image feature extraction and integrative data analysis. Despite its recent rapid development, ML technology still faces many hurdles for its broader applications in neuro-oncological radiomic analysis, such as lack of large accessible standardized real patient radiomic brain tumor data of all kinds and reliable predictions on tumor response upon various treatments. Therefore, understanding ML-based AI technologies is critically important to help us address the skyrocketing demands of neuro-oncology clinical deployments. Here, we provide an overview on the latest advancements in ML techniques for brain tumor radiomic analysis, emphasizing proprietary and public dataset preparation and state-of-the-art ML models for brain tumor diagnosis, classifications (e.g., primary and secondary tumors), discriminations between treatment effects (pseudoprogression, radiation necrosis) and true progression, survival prediction, inflammation, and identification of brain tumor biomarkers. We also compare the key features of ML models in the realm of neuroradiology with ML models employed in other medical imaging fields and discuss open research challenges and directions for future work in this nascent precision medicine area.

Advanced Neuroimaging Approaches to Pediatric Brain Tumors

Simple Summary

After leukemias, brain tumors are the most common cancers in children, and early, accurate diagnosis is critical to improve patient outcomes. Beyond the conventional imaging methods of computed tomography (CT) and magnetic resonance imaging (MRI), advanced neuroimaging techniques capable of both structural and functional imaging are moving to the forefront to improve the early detection and differential diagnosis of tumors of the central nervous system. Here, we review recent developments in neuroimaging techniques for pediatric brain tumors.

Abstract

Central nervous system tumors are the most common pediatric solid tumors; they are also the most lethal. Unlike adults, childhood brain tumors are mostly primary in origin and differ in type, location and molecular signature. Tumor characteristics (incidence, location, and type) vary with age. Children present with a variety of symptoms, making early accurate diagnosis challenging. Neuroimaging is key in the initial diagnosis and monitoring of pediatric brain tumors. Conventional anatomic imaging approaches (computed tomography (CT) and magnetic resonance imaging (MRI)) are useful for tumor detection but have limited utility differentiating tumor types and grades. Advanced MRI techniques (diffusion-weighed imaging, diffusion tensor imaging, functional MRI, arterial spin labeling perfusion imaging, MR spectroscopy, and MR elastography) provide additional and improved structural and functional information. Combined with positron emission tomography (PET) and single-photon emission CT (SPECT), advanced techniques provide functional information on tumor metabolism and physiology through the use of radiotracer probes. Radiomics and radiogenomics offer promising insight into the prediction of tumor subtype, post-treatment response to treatment, and prognostication. In this paper, a brief review of pediatric brain cancers, by type, is provided with a comprehensive description of advanced imaging techniques including clinical applications that are currently utilized for the assessment and evaluation of pediatric brain tumors.

Long Noncoding RNA RP11-732M18.3 Promotes Glioma Angiogenesis by Upregulating VEGFA

Gliomas are the most aggressive and common type of malignant brain tumor, with limited treatment options and a dismal prognosis. Angiogenesis, a hallmarks of cancer, is one of two critical events in the progression of gliomas. Accumulating evidence has demonstrated that in glioma dysregulated molecules like long noncoding RNAs (lncRNAs), are closely linked to tumorigenesis and prognosis. However, the effects of and mechanisms of action of lncRNAs during tumor angiogenesis are poorly understood. The effect of lncRNA RP11-732M18.3 on angiogenesis was elucidated through an intracranial orthotopic glioma model, immunohistochemistry, and an in vitro angiogenesis assay. Co-culture experiments and cell migration assays were performed to investigate the function of lncRNA RP11-732M18.3 in vitro. lncRNA RP11-732M18.3 increased CD31⁺ microvessel density, and overexpression of lncRNA RP11-732M18.3 resulted in poor mouse survival. lncRNA RP11-732M18.3 promoted endothelial cell migration and tube formation. Nomogram and Kaplan-Meier survival analyses indicated that higher VEGFA is correlated with a poor prognosis. Mechanistically, lncRNA RP11-732M18.3 promotes angiogenesis by increasing the nuclear level of EP300 and facilitating the transcription and secretion of VEGFA. Our study contributes to the latest understanding of glioma angiogenesis and prognosis. lncRNA RP11-732M18.3 may be a potential treatment target in glioma.

Multi-parametric MRI-based radiomics signature for preoperative prediction of Ki-67 proliferation status in sinonasal malignancies: a two-centre study

OBJECTIVE

To assess the predictive ability of a multi-parametric MRI-based radiomics signature (RS) for the preoperative evaluation of Ki-67 proliferation status in sinonasal malignancies.

METHODS

A total of 128 patients with sinonasal malignancies that underwent multi-parametric MRIs at two medical centres were retrospectively analysed. Data from one medical centre (n = 77) were used to develop the predictive models and data from the other medical centre (n = 51) constitute the test dataset. Clinical data and conventional MRI findings were reviewed to identify significant predictors. Radiomics features were determined using maximum relevance minimum redundancy and least absolute shrinkage and selection operator algorithms. Subsequently, RSs were established using a logistic regression (LR) algorithm. The predictive performance of RSs was assessed using calibration, decision curve analysis (DCA), accuracy, and AUC.

RESULTS

No independent predictors of high Ki-67 proliferation were observed based on clinical data and conventional MRI findings. RS-T1, RS-T2, and RS-T1c (contrast enhancement T1WI) were established based on a single-parametric MRI. RS-Combined (combining T1WI, FS-T2WI, and T1c features) was developed based on multi-parametric MRI and achieved an AUC and accuracy of 0.852 (0.733-0.971) and 86.3%, respectively, on the test dataset. The calibration curve and DCA demonstrated an improved fitness and benefits in clinical practice.

CONCLUSIONS

A multi-parametric MRI-based RS may be used as a non-invasive, dependable, and accurate tool for preoperative evaluation of the Ki-67 proliferation status to overcome the sampling bias in sinonasal malignancies.

KEY POINTS

• Multi-parametric MRI-based radiomics signatures (RSs) are used to preoperatively evaluate the proliferation status of Ki-67 in sinonasal malignancies. • Radiomics features are determined using maximum relevance minimum redundancy (mRMR) and least absolute shrinkage and selection operator (LASSO) algorithms. • RSs are established using a logistic regression (LR) algorithm.

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.

The efficacy of deep learning models in the diagnosis of endometrial cancer using MRI: a comparison with radiologists

Purpose

To compare the diagnostic performance of deep learning models using convolutional neural networks (CNN) with that of radiologists in diagnosing endometrial cancer and to verify suitable imaging conditions.

Methods

This retrospective study included patients with endometrial cancer or non-cancerous lesions who underwent MRI between 2015 and 2020. In Experiment 1, single and combined image sets of several sequences from 204 patients with cancer and 184 patients with non-cancerous lesions were used to train CNNs. Subsequently, testing was performed using 97 images from 51 patients with cancer and 46 patients with non-cancerous lesions. The test image sets were independently interpreted by three blinded radiologists. Experiment 2 investigated whether the addition of different types of images for training using the single image sets improved the diagnostic performance of CNNs.

Results

The AUC of the CNNs pertaining to the single and combined image sets were 0.88–0.95 and 0.87–0.93, respectively, indicating non-inferior diagnostic performance than the radiologists. The AUC of the CNNs trained with the addition of other types of single images to the single image sets was 0.88–0.95.

Conclusion

CNNs demonstrated high diagnostic performance for the diagnosis of endometrial cancer using MRI. Although there were no significant differences, adding other types of images improved the diagnostic performance for some single image sets.

Alternations and Applications of the Structural and Functional Connectome in Gliomas: A Mini-Review

In the central nervous system, gliomas are the most common, but complex primary tumors. Genome-based molecular and clinical studies have revealed different classifications and subtypes of gliomas. Neuroradiological approaches have non-invasively provided a macroscopic view for surgical resection and therapeutic effects. The connectome is a structural map of a physical object, the brain, which raises issues of spatial scale and definition, and it is calculated through diffusion magnetic resonance imaging (MRI) and functional MRI. In this study, we reviewed the basic principles and attributes of the structural and functional connectome, followed by the alternations of connectomes and their influences on glioma. To extend the applications of connectome, we demonstrated that a series of multi-center projects still need to be conducted to systemically investigate the connectome and the structural-functional coupling of glioma. Additionally, the brain-computer interface based on accurate connectome could provide more precise structural and functional data, which are significant for surgery and postoperative recovery. Besides, integrating the data from different sources, including connectome and other omics information, and their processing with artificial intelligence, together with validated biological and clinical findings will be significant for the development of a personalized surgical strategy.

Unsupervised machine learning for identifying important visual features through bag-of-words using histopathology data from chronic kidney disease

Pathologists use visual classification to assess patient kidney biopsy samples when diagnosing the underlying cause of kidney disease. However, the assessment is qualitative, or semi-quantitative at best, and reproducibility is challenging. To discover previously unknown features which predict patient outcomes and overcome substantial interobserver variability, we developed an unsupervised bag-of-words model. Our study applied to the C-PROBE cohort of patients with chronic kidney disease (CKD). 107,471 histopathology images were obtained from 161 biopsy cores and identified important morphological features in biopsy tissue that are highly predictive of the presence of CKD both at the time of biopsy and in one year. To evaluate the performance of our model, we estimated the AUC and its 95% confidence interval. We show that this method is reliable and reproducible and can achieve 0.93 AUC at predicting glomerular filtration rate at the time of biopsy as well as predicting a loss of function at one year. Additionally, with this method, we ranked the identified morphological features according to their importance as diagnostic markers for chronic kidney disease. In this study, we have demonstrated the feasibility of using an unsupervised machine learning method without human input in order to predict the level of kidney function in CKD. The results from our study indicate that the visual dictionary, or visual image pattern, obtained from unsupervised machine learning can predict outcomes using machine-derived values that correspond to both known and unknown clinically relevant features.

Regorafenib in Glioblastoma Recurrence: How to Deal With MR Imaging Treatments Changes

The treatment of recurrent high-grade gliomas remains a major challenge of daily neuro-oncology practice, and imaging findings of new therapies may be challenging. Regorafenib is a multi-kinase inhibitor that has recently been introduced into clinical practice to treat recurrent glioblastoma, bringing with it a novel panel of MRI imaging findings. On the basis of the few data in the literature and on our personal experience, we have identified the main MRI changes during regorafenib therapy, and then, we defined two different patterns, trying to create a simple summary line of the main changes of pathological tissue during therapy. We named these patterns, respectively, pattern A (less frequent, similar to classical progression disease) and pattern B (more frequent, with decreased diffusivity and decrease contrast-enhancement). We have also reported MR changes concerning signal intensity on T1-weighted and T2-weighted images, SWI, and perfusion imaging, derived from the literature (small series or case reports) and from our clinical experience. The clinical implication of these imaging modifications remains to be defined, taking into account that we are still at the dawn in the evaluation of such imaging modifications.

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.

Magnetic resonance spectroscopy outperforms perfusion in distinguishing between pseudoprogression and disease progression in patients with glioblastoma

Background

There is a need to establish biomarkers that distinguish between pseudoprogression (PsP) and true tumor progression in patients with glioblastoma (GBM) treated with chemoradiation.

Methods

We analyzed magnetic resonance spectroscopic imaging (MRSI) and dynamic susceptibility contrast (DSC) MR perfusion data in patients with GBM with PsP or disease progression after chemoradiation. MRSI metabolites of interest included intratumoral choline (Cho), myo-inositol (mI), glutamate + glutamine (Glx), lactate (Lac), and creatine on the contralateral hemisphere (c-Cr). Student T-tests and area under the ROC curve analyses were used to detect group differences in metabolic ratios and their ability to predict clinical status, respectively.

Results

28 subjects (63 ± 9 years, 19 men) were evaluated. Subjects with true progression (n = 20) had decreased enhancing region mI/c-Cr (P = .011), a marker for more aggressive tumors, compared to those with PsP, which predicted tumor progression (AUC: 0.84 [0.76, 0.92]). Those with true progression had elevated Lac/Glx (P = .0009), a proxy of the Warburg effect, compared to those with PsP which predicted tumor progression (AUC: 0.84 [0.75, 0.92]). Cho/c-Cr did not distinguish between PsP and true tumor progression. Despite rCBV (AUC: 0.70 [0.60, 0.80]) and rCBF (AUC: 0.75 [0.65, 0.84]) being individually predictive of tumor response, they added no additional predictive value when combined with MRSI metabolic markers.

Conclusions

Incorporating enhancing lesion MRSI measures of mI/c-Cr and Lac/Glx into brain tumor imaging protocols can distinguish between PsP and true progression and inform patient management decisions.

[Artificial intelligence technologies in clinical neurooncology]

Neurooncology in the 21st century is a complex discipline integrating achievements of fundamental and applied neurosciences. Complex processes and data in clinical neurooncology determine the necessity for advanced methods of mathematical modeling and predictive analytics to obtain new scientific knowledge. Such methods are currently being developed in computer science (artificial intelligence). This review is devoted to potential and range of possible applications of artificial intelligence technologies in neurooncology with a special emphasis on glial tumors. Our conclusions may be valid for other areas of clinical medicine.

Oncolytic Viro-Immunotherapy: An Emerging Option in the Treatment of Gliomas

The prognosis of malignant gliomas remains poor, with median survival fewer than 20 months and a 5-year survival rate merely 5%. Their primary location in the central nervous system (CNS) and its immunosuppressive environment with little T cell infiltration has rendered cancer therapies mostly ineffective, and breakthrough therapies such as immune checkpoint inhibitors (ICIs) have shown limited benefit. However, tumor immunotherapy is developing rapidly and can help overcome these obstacles. But for now, malignant gliomas remain fatal with short survival and limited therapeutic options. Oncolytic virotherapy (OVT) is a unique antitumor immunotherapy wherein viruses selectively or preferentially kill tumor cells, replicate and spread through tumors while inducing antitumor immune responses. OVTs can also recondition the tumor microenvironment and improve the efficacy of other immunotherapies by escalating the infiltration of immune cells into tumors. Some OVTs can penetrate the blood-brain barrier (BBB) and possess tropism for the CNS, enabling intravenous delivery. Despite the therapeutic potential displayed by oncolytic viruses (OVs), optimizing OVT has proved challenging in clinical development, and marketing approvals for OVTs have been rare. In June 2021 however, as a genetically engineered OV based on herpes simplex virus-1 (G47Δ), teserpaturev got conditional and time-limited approval for the treatment of malignant gliomas in Japan. In this review, we summarize the current state of OVT, the synergistic effect of OVT in combination with other immunotherapies as well as the hurdles to successful clinical use. We also provide some suggestions to overcome the challenges in treating of gliomas.

Dynamic Susceptibility Perfusion Imaging for Differentiating Progressive Disease from Pseudoprogression in Diffuse Glioma Molecular Subtypes

Rationale and Objectives: Advanced adjuvant therapy of diffuse gliomas can result in equivocal findings in follow-up imaging. We aimed to assess the additional value of dynamic susceptibility perfusion imaging in the differentiation of progressive disease (PD) from pseudoprogression (PsP) in different molecular glioma subtypes. Materials and Methods: 89 patients with treated diffuse glioma with different molecular subtypes (IDH wild type (Astro-IDH^wt), IDH mutant astrocytomas (Astro-IDH^mut) and oligodendrogliomas), and tumor-suspect lesions on post-treatment follow-up imaging were classified into two outcome groups (PD or PsP) retrospectively by histopathology or clinical follow-up. The relative cerebral blood volume (rCBV) was assessed in the tumor-suspect FLAIR and contrast-enhancing (CE) lesions. We analyzed how a multilevel classification using a molecular subtype, the presence of a CE lesion, and two rCBV histogram parameters performed for PD prediction compared with a decision tree model (DTM) using additional rCBV parameters. Results: The PD rate was 69% in the whole cohort, 86% in Astro-IDH^wt, 52% in Astro-IDH^mut, and 55% in oligodendrogliomas. In the presence of a CE lesion, the PD rate was higher with 82%, 94%, 59%, and 88%, respectively; if there was no CE lesion, however, the PD rate was only 44%, 60%, 40%, and 33%, respectively. The additional use of the rCBV parameters in the DTM yielded a prediction accuracy for PD of 99%, 100%, 93%, and 95%, respectively. Conclusion: Utilizing combined information about the molecular tumor type, the presence or absence of CE lesions and rCBV parameters increases PD prediction accuracy in diffuse glioma.