Fully automatic GBM segmentation in the TCGA-GBM dataset: Prognosis and correlation with VASARI features

Training and Comparison of nnU-Net and DeepMedic Methods for Autosegmentation of Pediatric Brain Tumors

BACKGROUND AND PURPOSE

Tumor segmentation is essential in surgical and treatment planning and response assessment and monitoring in pediatric brain tumors, the leading cause of cancer-related death among children. However, manual segmentation is time-consuming and has high interoperator variability, underscoring the need for more efficient methods. After training, we compared 2 deep-learning-based 3D segmentation models, DeepMedic and nnU-Net, with pediatric-specific multi-institutional brain tumor data based on multiparametric MR images.

MATERIALS AND METHODS

Multiparametric preoperative MR imaging scans of 339 pediatric patients (n = 293 internal and n = 46 external cohorts) with a variety of tumor subtypes were preprocessed and manually segmented into 4 tumor subregions, ie, enhancing tumor, nonenhancing tumor, cystic components, and peritumoral edema. After training, performances of the 2 models on internal and external test sets were evaluated with reference to ground truth manual segmentations. Additionally, concordance was assessed by comparing the volume of the subregions as a percentage of the whole tumor between model predictions and ground truth segmentations using the Pearson or Spearman correlation coefficients and the Bland-Altman method.

RESULTS

The mean Dice score for nnU-Net internal test set was 0.9 (SD, 0.07) (median, 0.94) for whole tumor; 0.77 (SD, 0.29) for enhancing tumor; 0.66 (SD, 0.32) for nonenhancing tumor; 0.71 (SD, 0.33) for cystic components, and 0.71 (SD, 0.40) for peritumoral edema, respectively. For DeepMedic, the mean Dice scores were 0.82 (SD, 0.16) for whole tumor; 0.66 (SD, 0.32) for enhancing tumor; 0.48 (SD, 0.27) for nonenhancing tumor; 0.48 (SD, 0.36) for cystic components, and 0.19 (SD, 0.33) for peritumoral edema, respectively. Dice scores were significantly higher for nnU-Net (P ≤ .01). Correlation coefficients for tumor subregion percentage volumes were higher (0.98 versus 0.91 for enhancing tumor, 0.97 versus 0.75 for nonenhancing tumor, 0.98 versus 0.80 for cystic components, 0.95 versus 0.33 for peritumoral edema in the internal test set). Bland-Altman plots were better for nnU-Net compared with DeepMedic. External validation of the trained nnU-Net model on the multi-institutional Brain Tumor Segmentation Challenge in Pediatrics (BraTS-PEDs) 2023 data set revealed high generalization capability in the segmentation of whole tumor, tumor core (a combination of enhancing tumor, nonenhancing tumor, and cystic components), and enhancing tumor with mean Dice scores of 0.87 (SD, 0.13) (median, 0.91), 0.83 (SD, 0.18) (median, 0.89), and 0.48 (SD, 0.38) (median, 0.58), respectively.

CONCLUSIONS

The pediatric-specific data-trained nnU-Net model is superior to DeepMedic for whole tumor and subregion segmentation of pediatric brain tumors.

Application of novel PACS-based informatics platform to identify imaging based predictors of CDKN2A allelic status in glioblastomas

Gliomas with CDKN2A mutations are known to have worse prognosis but imaging features of these gliomas are unknown. Our goal is to identify CDKN2A specific qualitative imaging biomarkers in glioblastomas using a new informatics workflow that enables rapid analysis of qualitative imaging features with Visually AcceSAble Rembrandtr Images (VASARI) for large datasets in PACS. Sixty nine patients undergoing GBM resection with CDKN2A status determined by whole-exome sequencing were included. GBMs on magnetic resonance images were automatically 3D segmented using deep learning algorithms incorporated within PACS. VASARI features were assessed using FHIR forms integrated within PACS. GBMs without CDKN2A alterations were significantly larger (64 vs. 30%, p = 0.007) compared to tumors with homozygous deletion (HOMDEL) and heterozygous loss (HETLOSS). Lesions larger than 8 cm were four times more likely to have no CDKN2A alteration (OR: 4.3; 95% CI 1.5-12.1; p < 0.001). We developed a novel integrated PACS informatics platform for the assessment of GBM molecular subtypes and show that tumors with HOMDEL are more likely to have radiographic evidence of pial invasion and less likely to have deep white matter invasion or subependymal invasion. These imaging features may allow noninvasive identification of CDKN2A allele status.

A review on brain tumor segmentation based on deep learning methods with federated learning techniques

Brain tumors have become a severe medical complication in recent years due to their high fatality rate. Radiologists segment the tumor manually, which is time-consuming, error-prone, and expensive. In recent years, automated segmentation based on deep learning has demonstrated promising results in solving computer vision problems such as image classification and segmentation. Brain tumor segmentation has recently become a prevalent task in medical imaging to determine the tumor location, size, and shape using automated methods. Many researchers have worked on various machine and deep learning approaches to determine the most optimal solution using the convolutional methodology. In this review paper, we discuss the most effective segmentation techniques based on the datasets that are widely used and publicly available. We also proposed a survey of federated learning methodologies to enhance global segmentation performance and ensure privacy. A comprehensive literature review is suggested after studying more than 100 papers to generalize the most recent techniques in segmentation and multi-modality information. Finally, we concentrated on unsolved problems in brain tumor segmentation and a client-based federated model training strategy. Based on this review, future researchers will understand the optimal solution path to solve these issues.

Multi-class glioma segmentation on real-world data with missing MRI sequences: comparison of three deep learning algorithms

This study tests the generalisability of three Brain Tumor Segmentation (BraTS) challenge models using a multi-center dataset of varying image quality and incomplete MRI datasets. In this retrospective study, DeepMedic, no-new-Unet (nn-Unet), and NVIDIA-net (nv-Net) were trained and tested using manual segmentations from preoperative MRI of glioblastoma (GBM) and low-grade gliomas (LGG) from the BraTS 2021 dataset (1251 in total), in addition to 275 GBM and 205 LGG acquired clinically across 12 hospitals worldwide. Data was split into 80% training, 5% validation, and 15% internal test data. An additional external test-set of 158 GBM and 69 LGG was used to assess generalisability to other hospitals' data. All models' median Dice similarity coefficient (DSC) for both test sets were within, or higher than, previously reported human inter-rater agreement (range of 0.74-0.85). For both test sets, nn-Unet achieved the highest DSC (internal = 0.86, external = 0.93) and the lowest Hausdorff distances (10.07, 13.87 mm, respectively) for all tumor classes (p < 0.001). By applying Sparsified training, missing MRI sequences did not statistically affect the performance. nn-Unet achieves accurate segmentations in clinical settings even in the presence of incomplete MRI datasets. This facilitates future clinical adoption of automated glioma segmentation, which could help inform treatment planning and glioma monitoring.

Evaluation of the HD-GLIO Deep Learning Algorithm for Brain Tumour Segmentation on Postoperative MRI

In the context of brain tumour response assessment, deep learning-based three-dimensional (3D) tumour segmentation has shown potential to enter the routine radiological workflow. The purpose of the present study was to perform an external evaluation of a state-of-the-art deep learning 3D brain tumour segmentation algorithm (HD-GLIO) on an independent cohort of consecutive, post-operative patients. For 66 consecutive magnetic resonance imaging examinations, we compared delineations of contrast-enhancing (CE) tumour lesions and non-enhancing T2/FLAIR hyperintense abnormality (NE) lesions by the HD-GLIO algorithm and radiologists using Dice similarity coefficients (Dice). Volume agreement was assessed using concordance correlation coefficients (CCCs) and Bland-Altman plots. The algorithm performed very well regarding the segmentation of NE volumes (median Dice = 0.79) and CE tumour volumes larger than 1.0 cm³ (median Dice = 0.86). If considering all cases with CE tumour lesions, the performance dropped significantly (median Dice = 0.40). Volume agreement was excellent with CCCs of 0.997 (CE tumour volumes) and 0.922 (NE volumes). The findings have implications for the application of the HD-GLIO algorithm in the routine radiological workflow where small contrast-enhancing tumours will constitute a considerable share of the follow-up cases. Our study underlines that independent validations on clinical datasets are key to asserting the robustness of deep learning algorithms.

Multiparametric Magnetic Resonance Imaging Correlates of Isocitrate Dehydrogenase Mutation in WHO high-Grade Astrocytomas

Purpose and background: Isocitrate dehydrogenase (IDH) mutation and O-6 methyl guanine methyl transferase (MGMT) methylation are surrogate biomarkers of improved survival in gliomas. This study aims at studying the ability of semantic magnetic resonance imaging (MRI) features to predict the IDH mutation status confirmed by the gold standard molecular tests. Methods: The MRI of 148 patients were reviewed for various imaging parameters based on the Visually AcceSAble Rembrandt Images (VASARI) study. Their IDH status was determined using immunohistochemistry (IHC). Fisher’s exact or chi-square tests for univariate and logistic regression for multivariate analysis were used. Results: Parameters such as mild and patchy enhancement, minimal edema, necrosis < 25%, presence of cysts, and less rCBV (relative cerebral blood volume) correlated with IDH mutation. The median age of IDH-mutant and IDH-wild patients were 34 years (IQR: 29−43) and 52 years (IQR: 45−59), respectively. Mild to moderate enhancement was observed in 15/19 IDH-mutant patients (79%), while 99/129 IDH-wildtype (77%) had severe enhancement (p-value <0.001). The volume of edema with respect to tumor volume distinguished IDH-mutants from wild phenotypes (peritumoral edema volume < tumor volume was associated with higher IDH-mutant phenotypes; p-value < 0.025). IDH-mutant patients had a median rCBV value of 1.8 (IQR: 1.4−2.0), while for IDH-wild phenotypes, it was 2.6 (IQR: 1.9−3.5) {p-value = 0.001}. On multivariate analysis, a cut-off of 25% necrosis was able to differentiate IDH-mutant from IDH-wildtype (p-value < 0.001), and a cut-off rCBV of 2.0 could differentiate IDH-mutant from IDH-wild phenotypes (p-value < 0.007). Conclusion: Semantic imaging features could reliably predict the IDH mutation status in high-grade gliomas. Presurgical prediction of IDH mutation status could help the treating oncologist to tailor the adjuvant therapy or use novel IDH inhibitors.

A Review of the Multi-Systemic Complications of a Ketogenic Diet in Children and Infants with Epilepsy

Ketogenic diets (KDs) are highly effective in the treatment of epilepsy. However, numerous complications have been reported. During the initiation phase of the diet, common side effects include vomiting, hypoglycemia, metabolic acidosis and refusal of the diet. While on the diet, the side effects involve the following systems: gastrointestinal, hepatic, cardiovascular, renal, dermatological, hematologic and bone. Many of the common side effects can be tackled easily with careful monitoring including blood counts, liver enzymes, renal function tests, urinalysis, vitamin levels, mineral levels, lipid profiles, and serum carnitine levels. Some rare and serious side effects reported in the literature include pancreatitis, protein-losing enteropathy, prolonged QT interval, cardiomyopathy and changes in the basal ganglia. These serious complications may need more advanced work-up and immediate cessation of the diet. With appropriate monitoring and close follow-up to minimize adverse effects, KDs can be effective for patients with intractable epilepsy.

The University of Pennsylvania glioblastoma (UPenn-GBM) cohort: advanced MRI, clinical, genomics, & radiomics

Glioblastoma is the most common aggressive adult brain tumor. Numerous studies have reported results from either private institutional data or publicly available datasets. However, current public datasets are limited in terms of: a) number of subjects, b) lack of consistent acquisition protocol, c) data quality, or d) accompanying clinical, demographic, and molecular information. Toward alleviating these limitations, we contribute the "University of Pennsylvania Glioblastoma Imaging, Genomics, and Radiomics" (UPenn-GBM) dataset, which describes the currently largest publicly available comprehensive collection of 630 patients diagnosed with de novo glioblastoma. The UPenn-GBM dataset includes (a) advanced multi-parametric magnetic resonance imaging scans acquired during routine clinical practice, at the University of Pennsylvania Health System, (b) accompanying clinical, demographic, and molecular information, (d) perfusion and diffusion derivative volumes, (e) computationally-derived and manually-revised expert annotations of tumor sub-regions, as well as (f) quantitative imaging (also known as radiomic) features corresponding to each of these regions. This collection describes our contribution towards repeatable, reproducible, and comparative quantitative studies leading to new predictive, prognostic, and diagnostic assessments.

Conventional and Advanced Imaging Techniques in Post-treatment Glioma Imaging

Despite decades of advancement in the diagnosis and therapy of gliomas, the most malignant primary brain tumors, the overall survival rate is still dismal, and their post-treatment imaging appearance remains very challenging to interpret. Since the limitations of conventional magnetic resonance imaging (MRI) in the distinction between recurrence and treatment effect have been recognized, a variety of advanced MR and functional imaging techniques including diffusion-weighted imaging (DWI), diffusion tensor imaging (DTI), perfusion-weighted imaging (PWI), MR spectroscopy (MRS), as well as a variety of radiotracers for single photon emission computed tomography (SPECT) and positron emission tomography (PET) have been investigated for this indication along with voxel-based and more quantitative analytical methods in recent years. Machine learning and radiomics approaches in recent years have shown promise in distinguishing between recurrence and treatment effect as well as improving prognostication in a malignancy with a very short life expectancy. This review provides a comprehensive overview of the conventional and advanced imaging techniques with the potential to differentiate recurrence from treatment effect and includes updates in the state-of-the-art in advanced imaging with a brief overview of emerging experimental techniques. A series of representative cases are provided to illustrate the synthesis of conventional and advanced imaging with the clinical context which informs the radiologic evaluation of gliomas in the post-treatment setting.

Machine Learning Tools for Image-Based Glioma Grading and the Quality of Their Reporting: Challenges and Opportunities

Technological innovation has enabled the development of machine learning (ML) tools that aim to improve the practice of radiologists. In the last decade, ML applications to neuro-oncology have expanded significantly, with the pre-operative prediction of glioma grade using medical imaging as a specific area of interest. We introduce the subject of ML models for glioma grade prediction by remarking upon the models reported in the literature as well as by describing their characteristic developmental workflow and widely used classifier algorithms. The challenges facing these models-including data sources, external validation, and glioma grade classification methods -are highlighted. We also discuss the quality of how these models are reported, explore the present and future of reporting guidelines and risk of bias tools, and provide suggestions for the reporting of prospective works. Finally, this review offers insights into next steps that the field of ML glioma grade prediction can take to facilitate clinical implementation.

Magnetic resonance image-based brain tumour segmentation methods: A systematic review

Background

Image segmentation is an essential step in the analysis and subsequent characterisation of brain tumours through magnetic resonance imaging. In the literature, segmentation methods are empowered by open-access magnetic resonance imaging datasets, such as the brain tumour segmentation dataset. Moreover, with the increased use of artificial intelligence methods in medical imaging, access to larger data repositories has become vital in method development.

Purpose

To determine what automated brain tumour segmentation techniques can medical imaging specialists and clinicians use to identify tumour components, compared to manual segmentation.

Methods

We conducted a systematic review of 572 brain tumour segmentation studies during 2015-2020. We reviewed segmentation techniques using T1-weighted, T2-weighted, gadolinium-enhanced T1-weighted, fluid-attenuated inversion recovery, diffusion-weighted and perfusion-weighted magnetic resonance imaging sequences. Moreover, we assessed physics or mathematics-based methods, deep learning methods, and software-based or semi-automatic methods, as applied to magnetic resonance imaging techniques. Particularly, we synthesised each method as per the utilised magnetic resonance imaging sequences, study population, technical approach (such as deep learning) and performance score measures (such as Dice score).

Statistical tests

We compared median Dice score in segmenting the whole tumour, tumour core and enhanced tumour.

Results

We found that T1-weighted, gadolinium-enhanced T1-weighted, T2-weighted and fluid-attenuated inversion recovery magnetic resonance imaging are used the most in various segmentation algorithms. However, there is limited use of perfusion-weighted and diffusion-weighted magnetic resonance imaging. Moreover, we found that the U-Net deep learning technology is cited the most, and has high accuracy (Dice score 0.9) for magnetic resonance imaging-based brain tumour segmentation.

Conclusion

U-Net is a promising deep learning technology for magnetic resonance imaging-based brain tumour segmentation. The community should be encouraged to contribute open-access datasets so training, testing and validation of deep learning algorithms can be improved, particularly for diffusion- and perfusion-weighted magnetic resonance imaging, where there are limited datasets available.

Magnetic Resonance Imaging in the Contemporary Management of Medulloblastoma: Current and Emerging Applications

Medulloblastoma, the most common malignant primary brain tumor in children, is now considered to comprise of four distinct molecular subgroups-wingless (WNT), sonic hedgehog (SHH), Group 3, and Group 4 medulloblastoma, each associated with distinct developmental origins, unique transcriptional profiles, diverse phenotypes, and variable clinical behavior. Due to its exquisite anatomic resolution, multiparametric nature, and ability to image the entire craniospinal axis, magnetic resonance imaging (MRI) is the preferred and recommended first-line imaging modality for suspected brain tumors including medulloblastoma. Preoperative MRI can reliably differentiate medulloblastoma from other common childhood posterior fossa masses such as ependymoma, pilocytic astrocytoma, and brainstem glioma. On T1-weighted images, medulloblastoma is generally iso- to hypointense, while on T2-weighted images, the densely packed cellular component of the tumor is significantly hypointense and displays restricted diffusion on diffusion-weighted imaging. Following intravenous gadolinium, medulloblastoma shows significant but variable and heterogeneous contrast enhancement. Given the propensity of neuraxial spread in medulloblastoma, sagittal fat-suppressed T1-postcontrast spinal MRI is recommended to rule out leptomeningeal metastases for accurate staging. Following neurosurgical excision, postoperative MRI done within 24-48 h confirms the extent of resection, accurately quantifying residual tumor burden imperative for risk assignment. Post-treatment MRI is needed to assess response and effectiveness of adjuvant radiotherapy and systemic chemotherapy. After completion of planned therapy, surveillance MRI is recommended periodically on follow-up for early detection of recurrence for timely institution of salvage therapy, as well as for monitoring treatment-related late complications. Recent studies suggest that preoperative MRI can reliably identify SHH and Group 4 medulloblastoma but has suboptimal predictive accuracy for WNT and Group 3 tumors. In this review, we focus on the role of MRI in the diagnosis, staging, and quantifying residual disease; post-treatment response assessment; and periodic surveillance, and provide a brief summary on radiogenomics in the contemporary management of medulloblastoma.

Segmentation of multicorrelated images with copula models and conditionally random fields

Purpose: Multisource images are interesting in medical imaging. Indeed, multisource images enable the use of complementary information of different sources such as for T1 and T2 modalities in MRI imaging. However, such multisource data can also be subject to redundancy and correlation. The question is how to efficiently fuse the multisource information without reinforcing the redundancy. We propose a method for segmenting multisource images that are statistically correlated. Approach: The method that we propose is the continuation of a prior work in which we introduce the copula model in hidden Markov fields (HMF). To achieve the multisource segmentations, we use a functional measure of dependency called "copula." This copula is incorporated in the conditionally random fields (CRF). Contrary to HMF, where we consider a prior knowledge on the hidden states modeled by an HMF, in CRF, there is no prior information and only the distribution of the hidden states conditionally to the observations can be known. This conditional distribution depends on the data and can be modeled by an energy function composed of two terms. The first one groups the voxels having similar intensities in the same class. As for the second term, it encourages a pair of voxels to be in the same class if the difference between their intensities is not too big. Results: A comparison between HMF and CRF is performed via theory and experimentations using both simulated and real data from BRATS 2013. Moreover, our method is compared with different state-of-the-art methods, which include supervised (convolutional neural networks) and unsupervised (hierarchical MRF). Our unsupervised method gives similar results as decision trees for synthetic images and as convolutional neural networks for real images; both methods are supervised. Conclusions: We compare two statistical methods using the copula: HMF and CRF to deal with multicorrelated images. We demonstrate the interest of using copula. In both models, the copula considerably improves the results compared with individual segmentations.

A Review of Radiomics and Deep Predictive Modeling in Glioma Characterization

Recent developments in glioma categorization based on biological genotypes and application of computational machine learning or deep learning based predictive models using multi-modal MRI biomarkers to assess these genotypes provides potential assurance for optimal and personalized treatment plans and efficacy. Artificial intelligence based quantified assessment of glioma using MRI derived hand-crafted or auto-extracted features have become crucial as genomic alterations can be associated with MRI based phenotypes. This survey integrates all the recent work carried out in state-of-the-art radiomics, and Artificial Intelligence based learning solutions related to molecular diagnosis, prognosis, and treatment monitoring with the aim to create a structured resource on radiogenomic analysis of glioma. Challenges such as inter-scanner variability, requirement of benchmark datasets, prospective validations for clinical applicability are discussed with further scope for designing optimal solutions for glioma stratification with immediate recommendations for further diagnostic decisions and personalized treatment plans for glioma patients.

MRI-based machine learning for determining quantitative and qualitative characteristics affecting the survival of glioblastoma multiforme

PURPOSE

Our current study aims to consider the image biomarkers extracted from the MRI images for exploring their effects on glioblastoma multiforme (GBM) patients' survival. Determining its biomarker helps better manage the disease and evaluate treatments. It has been proven that imaging features could be used as a biomarker. The purpose of this study is to investigate the features in MRI and clinical features as the biomarker association of survival of GBM.

METHODS

55 patients were considered with five clinical features, 10 qualities pre-operative MRI image features, and six quantitative features obtained using BraTumIA software. It was run ANN, C5, Bayesian, and Cox models in two phases for determining important variables. In the first phase, we selected the quality features that occur at least in three models and quantitative in two models. In the second phase, models were run with the extracted features, and then the probability value of variables in each model was calculated.

RESULTS

The mean of accuracy, sensitivity, specificity, and area under curve (AUC) after running four machine learning techniques were 80.47, 82.54, 79.78, and 0.85, respectively. In the second step, the mean of accuracy, sensitivity, specificity, and AUC were 79.55, 78.71, 79.83, and 0.87, respectively.

CONCLUSION

We found the largest size of the width, the largest size of length, radiotherapy, volume of enhancement, volume of nCET, satellites, enhancing margin, and age feature are important features.

Predictors for time to awake in patients undergoing awake craniotomies

OBJECTIVE

Awake craniotomies are often characterized by alternating asleep-awake-asleep periods. Preceding the awake phase, patients are weaned from anesthesia and mechanical ventilation. Although clinicians aim to minimize the time to awake for patient safety and operating room efficiency, in some patients, the time to awake exceeds 20 minutes. The goal of this study was to determine the average time to awake and the factors associated with prolonged time to awake (> 20 minutes) in patients undergoing awake craniotomy.

METHODS

Records of patients who underwent awake craniotomy between 2003 and 2020 were evaluated. Time to awake was defined as the time between discontinuation of propofol and remifentanil infusion and the time of extubation. Patient and perioperative characteristics were explored as predictors for time to awake using logistic regression analyses.

RESULTS

Data of 307 patients were analyzed. The median (IQR) time to awake was 13 (10-20) minutes and exceeded 20 minutes in 17% (95% CI 13%-21%) of the patients. In both univariate and multivariable analyses, increased age, nonsmoker status, and American Society of Anesthesiologists (ASA) class III versus II were associated with a time to awake exceeding 20 minutes. BMI, as well as the use of alcohol, drugs, dexamethasone, or antiepileptic agents, was not significantly associated with the time to awake.

CONCLUSIONS

While most patients undergoing awake craniotomy are awake within a reasonable time frame after discontinuation of propofol and remifentanil infusion, time to awake exceeded 20 minutes in 17% of the patients. Increasing age, nonsmoker status, and higher ASA classification were found to be associated with a prolonged time to awake.

Impact of postoperative dexamethasone on survival, steroid dependency, and infections in newly diagnosed glioblastoma patients

BACKGROUND

We examined the effect of dexamethasone prescribed in the initial 3 postoperative weeks on survival, steroid dependency, and infection in glioblastoma patients.

METHODS

In this single-center retrospective cohort analysis, we electronically retrieved inpatient administration and outpatient prescriptions of dexamethasone and laboratory values from the medical record of 360 glioblastoma patients. We correlated total dexamethasone prescribed from postoperative day (POD) 0 to 21 with survival, dexamethasone prescription from POD30 to POD90, and diagnosis of an infection by POD90. These analyses were adjusted for age, Karnofsky performance status score, tumor volume, extent of resection, IDH1/2 tumor mutation, tumor MGMT promoter methylation, temozolomide and radiotherapy initiation, and maximum blood glucose level.

RESULTS

Patients were prescribed a median of 159 mg [109-190] of dexamethasone cumulatively by POD21. Every 16-mg increment (4 mg every 6 hours/day) of total dexamethasone associated with a 4% increase in mortality (95% confidence interval [CI] 1%-7%, P < .01), 12% increase in the odds of being prescribed dexamethasone from POD30 to POD90 (95% CI 6%-19%, P < .01), and 10% increase in the odds of being diagnosed with an infection (95% CI, 4%-17%, P < .01). Of the 175 patients who had their absolute lymphocyte count measured in the preoperative week, 80 (45.7%) had a value indicative of lymphopenia. In the POD1-POD28 period, this proportion was 82/167 (49.1%).

CONCLUSIONS

Lower survival, steroid dependency, and higher infection rate in glioblastoma patients associated with higher dexamethasone administration in the initial 3 postoperative weeks. Nearly half of the glioblastoma patients are lymphopenic preoperatively and up to 1 month postoperatively.

Radiomics, machine learning, and artificial intelligence-what the neuroradiologist needs to know

PURPOSE

Artificial intelligence (AI) is playing an ever-increasing role in Neuroradiology.

METHODS

When designing AI-based research in neuroradiology and appreciating the literature, it is important to understand the fundamental principles of AI. Training, validation, and test datasets must be defined and set apart as priorities. External validation and testing datasets are preferable, when feasible. The specific type of learning process (supervised vs. unsupervised) and the machine learning model also require definition. Deep learning (DL) is an AI-based approach that is modelled on the structure of neurons of the brain; convolutional neural networks (CNN) are a commonly used example in neuroradiology.

RESULTS

Radiomics is a frequently used approach in which a multitude of imaging features are extracted from a region of interest and subsequently reduced and selected to convey diagnostic or prognostic information. Deep radiomics uses CNNs to directly extract features and obviate the need for predefined features.

CONCLUSION

Common limitations and pitfalls in AI-based research in neuroradiology are limited sample sizes ("small-n-large-p problem"), selection bias, as well as overfitting and underfitting.

Quantitative volumetric assessment of baseline enhancing tumor volume as an imaging biomarker predicts overall survival in patients with glioblastoma

BACKGROUND

Glioblastoma multiforme (GBM) is the commonest malignant primary brain tumor and still has one of the worst prognoses among cancers in general. There is a need for non-invasive methods to predict individual prognosis in patients with GBM.

PURPOSE

To evaluate quantitative volumetric tissue assessment of enhancing tumor volume on cranial magnetic resonance imaging (MRI) as an imaging biomarker for predicting overall survival (OS) in patients with GBM.

MATERIAL AND METHODS

MRI scans of 49 patients with histopathologically confirmed GBM were analyzed retrospectively. Baseline contrast-enhanced (CE) MRI sequences were transferred to a segmentation-based three-dimensional quantification tool, and the enhancing tumor component was analyzed. Based on a cut-off percentage of the enhancing tumor volume (PoETV) of >84.78%, samples were dichotomized, and the OS and intracranial progression-free survival (PFS) were evaluated. Univariable and multivariable analyses, including variables such as sex, Karnofsky Performance Status score, O⁶-methylguanine-DNA-methyltransferase status, age, and resection status, were performed using the Cox regression model.

RESULTS

The median OS and PFS were 16.9 and 7 months in the entire cohort, respectively. Patients with a CE tumor volume of >84.78% showed a significantly shortened OS (12.9 months) compared to those with a CE tumor volume of ≤84.78% (17.7 months) (hazard ratio [HR] 2.72; 95% confidence interval [CI] 1.22-6.03; P = 0.01). Multivariable analysis confirmed that PoETV had a significant prognostic role (HR 2.47; 95% CI 1.08-5.65; P = 0.03).

CONCLUSION

We observed a correlation between PoETV and OS. This imaging biomarker may help predict the OS of patients with GBM.

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.

Advanced MR techniques in glioblastoma imaging-upcoming challenges and how to face them

KEY POINTS

• The management of gliomas has changed dramatically since the presentation of the revised WHO Classification of Tumors of the Central Nervous System in 2016 emphasizing the tumor heterogeneity based on their molecular profile.• The need for a more noninvasive characterization of glioblastomas (GBM) by establishing reliable imaging biomarkers to predict patient outcome and improve therapy monitoring is bigger than ever.• Multiparametric MRI, including promising newer techniques like electrical property tomography and mapping, may have the potential to provide enough information for intelligent imaging postprocessing algorithms to face the challenge by decoding GBM heterogeneity noninvasively.

A Voxel-Based Radiographic Analysis Reveals the Biological Character of Proneural-Mesenchymal Transition in Glioblastoma

Introduction: Proneural and mesenchymal subtypes are the most distinct demarcated categories in classification scheme, and there is often a shift from proneural type to mesenchymal subtype in the progression of glioblastoma (GBM). The molecular characters are determined by specific genomic methods, however, the application of radiography in clinical practice remains to be further studied. Here, we studied the topography features of GBM in proneural subtype, and further demonstrated the survival characteristics and proneural-mesenchymal transition (PMT) progression of samples by combining with the imaging variables.

Methods: Data were acquired from The Cancer Imaging Archive (TCIA, http://cancerimagingarchive.net). The radiography image, clinical variables and transcriptome subtype from 223 samples were used in this study. Proneural and mesenchymal subtype on GBM topography based on overlay and Voxel-based lesion-symptom mapping (VLSM) analysis were revealed. Besides, we carried out the comparison of survival analysis and PMT progression in and outside the VLSM-determined area.

Results: The overlay of total GBM and separated image of proneural and mesenchymal subtype revealed a correlation of the two subtypes. By VLSM analysis, proneural subtype was confirmed to be related to left inferior temporal medulla, and no significant voxel was found for mesenchymal subtype. The subsequent comparison between samples in and outside the VLSM-determined area showed difference in overall survival (OS) time, tumor purity, epithelial-mesenchymal transition (EMT) score and clinical variables.

Conclusions: PMT progression was determined by radiography approach. GBM samples in the VLSM-determined area tended to harbor the signature of proneural subtype. This study provides a valuable VLSM-determined area related to the predilection site, prognosis and PMT progression by the association between GBM topography and molecular characters.

Clinical, molecular, and radiomic profile of gliomas with FGFR3-TACC3 fusions

BACKGROUND

Actionable fibroblast growth factor receptor 3 (FGFR3)-transforming acidic coiled-coil protein 3 fusions (F3T3) are found in approximately 3% of gliomas, but their characteristics and prognostic significance are still poorly defined. Our goal was to characterize the clinical, radiological, and molecular profile of F3T3 positive diffuse gliomas.

METHODS

We screened F3T3 fusion by real-time (RT)-PCR and FGFR3 immunohistochemistry in a large series of gliomas, characterized for main genetic alterations, histology, and clinical evolution. We performed a radiological and radiomic case control study, using an exploratory and a validation cohort.

RESULTS

We screened 1162 diffuse gliomas (951 unselected cases and 211 preselected for FGFR3 protein immunopositivity), identifying 80 F3T3 positive gliomas. F3T3 was mutually exclusive with IDH mutation (P < 0.001) and EGFR amplification (P = 0.01), defining a distinct molecular cluster associated with CDK4 (P = 0.04) and MDM2 amplification (P = 0.03). F3T3 fusion was associated with longer survival for the whole series and for glioblastomas (median overall survival was 31.1 vs 19.9 mo, P = 0.02) and was an independent predictor of better outcome on multivariate analysis.F3T3 positive gliomas had specific MRI features, affecting preferentially insula and temporal lobe, and with poorly defined tumor margins. F3T3 fusion was correctly predicted by radiomics analysis on both the exploratory (area under the curve [AUC] = 0.87) and the validation MRI (AUC = 0.75) cohort. Using Cox proportional hazards models, radiomics predicted survival with a high C-index (0.75, SD 0.04), while the model combining clinical, genetic, and radiomic data showed the highest C-index (0.81, SD 0.04).

CONCLUSION

F3T3 positive gliomas have distinct molecular and radiological features, and better outcome.

Prognostic and Predictive Value of Integrated Qualitative and Quantitative Magnetic Resonance Imaging Analysis in Glioblastoma

Simple Summary

Glioblastoma (GBM) is the most malignant primary brain tumor, for which improving patient outcome is limited by a substantial amount of tumor heterogeneity. Magnetic resonance imaging (MRI) in combination with machine learning offers the possibility to collect qualitative and quantitative imaging features which can be used to predict patient prognosis and relevant tumor markers which can aid in selecting the right treatment. This study showed that combining these MRI features with clinical features has the highest prognostic value for GBM patients; this model performed similarly in an independent GBM cohort, showing its reproducibility. The prediction of tumor markers showed promising results in the training set but not could be validated in the independent dataset. This study shows the potential of using MRI to predict prognosis and tumor markers, but further optimization and prospective studies are warranted.

Abstract

Glioblastoma (GBM) is the most malignant primary brain tumor for which no curative treatment options exist. Non-invasive qualitative (Visually Accessible Rembrandt Images (VASARI)) and quantitative (radiomics) imaging features to predict prognosis and clinically relevant markers for GBM patients are needed to guide clinicians. A retrospective analysis of GBM patients in two neuro-oncology centers was conducted. The multimodal Cox-regression model to predict overall survival (OS) was developed using clinical features with VASARI and radiomics features in isocitrate dehydrogenase (IDH)-wild type GBM. Predictive models for IDH-mutation, 06-methylguanine-DNA-methyltransferase (MGMT)-methylation and epidermal growth factor receptor (EGFR) amplification using imaging features were developed using machine learning. The performance of the prognostic model improved upon addition of clinical, VASARI and radiomics features, for which the combined model performed best. This could be reproduced after external validation (C-index 0.711 95% CI 0.64–0.78) and used to stratify Kaplan–Meijer curves in two survival groups (p-value < 0.001). The predictive models performed significantly in the external validation for EGFR amplification (area-under-the-curve (AUC) 0.707, 95% CI 0.582–8.25) and MGMT-methylation (AUC 0.667, 95% CI 0.522–0.82) but not for IDH-mutation (AUC 0.695, 95% CI 0.436–0.927). The integrated clinical and imaging prognostic model was shown to be robust and of potential clinical relevance. The prediction of molecular markers showed promising results in the training set but could not be validated after external validation in a clinically relevant manner. Overall, these results show the potential of combining clinical features with imaging features for prognostic and predictive models in GBM, but further optimization and larger prospective studies are warranted.

Automatic brain lesion segmentation on standard magnetic resonance images: a scoping review

OBJECTIVES

Medical image analysis practices face challenges that can potentially be addressed with algorithm-based segmentation tools. In this study, we map the field of automatic MR brain lesion segmentation to understand the clinical applicability of prevalent methods and study designs, as well as challenges and limitations in the field.

DESIGN

Scoping review.

SETTING

Three databases (PubMed, IEEE Xplore and Scopus) were searched with tailored queries. Studies were included based on predefined criteria. Emerging themes during consecutive title, abstract, methods and whole-text screening were identified. The full-text analysis focused on materials, preprocessing, performance evaluation and comparison.

RESULTS

Out of 2990 unique articles identified through the search, 441 articles met the eligibility criteria, with an estimated growth rate of 10% per year. We present a general overview and trends in the field with regard to publication sources, segmentation principles used and types of lesions. Algorithms are predominantly evaluated by measuring the agreement of segmentation results with a trusted reference. Few articles describe measures of clinical validity.

CONCLUSIONS

The observed reporting practices leave room for improvement with a view to studying replication, method comparison and clinical applicability. To promote this improvement, we propose a list of recommendations for future studies in the field.

Conventional and advanced imaging throughout the cycle of care of gliomas

Although imaging of gliomas has evolved tremendously over the last decades, published techniques and protocols are not always implemented into clinical practice. Furthermore, most of the published literature focuses on specific timepoints in glioma management. This article reviews the current literature on conventional and advanced imaging techniques and chronologically outlines their practical relevance for the clinical management of gliomas throughout the cycle of care. Relevant articles were located through the Pubmed/Medline database and included in this review. Interpretation of conventional and advanced imaging techniques is crucial along the entire process of glioma care, from diagnosis to follow-up. In addition to the described currently existing techniques, we expect deep learning or machine learning approaches to assist each step of glioma management through tumor segmentation, radiogenomics, prognostication, and characterization of pseudoprogression. Thorough knowledge of the specific performance, possibilities, and limitations of each imaging modality is key for their adequate use in glioma management.

Prognostic relevance of CSF and peri-tumoral edema volumes in glioblastoma

BACKGROUND

We conducted a segmental volumetric analysis of pre-operative brain magnetic resonance images (MRIs) of glioblastoma patients to identify brain- and tumor-related features that are prognostic of survival.

METHODS

Using a dataset of 210 single-institutional adult glioblastoma patients, total volumes of the following tumor- and brain-related features were quantified on pre-operative MRIs using a fully automated segmentation tool: tumor enhancement, tumor non-enhancement, tumor necrosis, peri-tumoral edema, grey matter, white matter, and cerebrospinal fluid (CSF). Their association with survival using Cox regression models, adjusting for the well-known predictors of glioblastoma survival. The findings were verified in a second dataset consisting of 96 glioblastoma patients from The Cancer Imaging Archive and The Cancer Genome Atlas (TCIA/TCGA).

RESULTS

CSF volume and edema were independently and consistently associated with overall survival of glioblastoma patients in both datasets. Greater edema was associated with increased hazard or decreased survival [adjusted hazard ratio (aHR) with 95% confidence interval (CI): 1.34 [1.08-1.67], p = 0.008 (institutional dataset); and, 1.44 [1.08-1.93], p = 0.013 (TCIA/TCGA dataset)]. Greater CSF volume also correlated with increased hazard or decreased survival [aHR 1.27 [1.02-1.59], p = 0.035 (institutional dataset), and 1.42 [1.03-1.95], p = 0.032 (TCIA/TCGA dataset)].

CONCLUSIONS

Higher brain CSF volume and higher edema levels at diagnosis are independently associated with decreased survival in glioblastoma patients. These results highlight the importance of a broader, quantitative brain-wide radiological analyses and invite investigations to understand tumor-related causes of increased edema and possibly increased CSF volume.

Exploratory Analysis of Qualitative MR Imaging Features for the Differentiation of Glioblastoma and Brain Metastases

OBJECTIVES

To identify qualitative VASARI (Visually AcceSIble Rembrandt Images) Magnetic Resonance (MR) Imaging features for differentiation of glioblastoma (GBM) and brain metastasis (BM) of different primary tumors.

MATERIALS AND METHODS

T1-weighted pre- and post-contrast, T2-weighted, and T2-weighted, fluid attenuated inversion recovery (FLAIR) MR images of a total of 239 lesions from 109 patients with either GBM or BM (breast cancer, non-small cell (NSCLC) adenocarcinoma, NSCLC squamous cell carcinoma, small-cell lung cancer (SCLC)) were included. A set of adapted, qualitative VASARI MR features describing tumor appearance and location was scored (binary; 1 = presence of feature, 0 = absence of feature). Exploratory data analysis was performed on binary scores using a combination of descriptive statistics (proportions with 95% binomial confidence intervals), unsupervised methods and supervised methods including multivariate feature ranking using either repeated fitting or recursive feature elimination with Support Vector Machines (SVMs).

RESULTS

GBMs were found to involve all lobes of the cerebrum with a fronto-occipital gradient, often affected the corpus callosum (32.4%, 95% CI 19.1-49.2), and showed a strong preference for the right hemisphere (79.4%, 95% CI 63.2-89.7). BMs occurred most frequently in the frontal lobe (35.1%, 95% CI 28.9-41.9) and cerebellum (28.3%, 95% CI 22.6-34.8). The appearance of GBMs was characterized by preference for well-defined non-enhancing tumor margin (100%, 89.8-100), ependymal extension (52.9%, 36.7-68.5) and substantially less enhancing foci than BMs (44.1%, 28.9-60.6 vs. 75.1%, 68.8-80.5). Unsupervised and supervised analyses showed that GBMs are distinctively different from BMs and that this difference is driven by definition of non-enhancing tumor margin, ependymal extension and features describing laterality. Differentiation of histological subtypes of BMs was driven by the presence of well-defined enhancing and non-enhancing tumor margins and localization in the vision center. SVM models with optimal hyperparameters led to weighted F1-score of 0.865 for differentiation of GBMs from BMs and weighted F1-score of 0.326 for differentiation of BM subtypes.

CONCLUSION

VASARI MR imaging features related to definition of non-enhancing margin, ependymal extension, and tumor localization may serve as potential imaging biomarkers to differentiate GBMs from BMs.

Introduction to radiomics and radiogenomics in neuro-oncology: implications and challenges

Neuro-oncology largely consists of malignancies of the brain and central nervous system including both primary as well as metastatic tumors. Currently, a significant clinical challenge in neuro-oncology is to tailor therapies for patients based on a priori knowledge of their survival outcome or treatment response to conventional or experimental therapies. Radiomics or the quantitative extraction of subvisual data from conventional radiographic imaging has recently emerged as a powerful data-driven approach to offer insights into clinically relevant questions related to diagnosis, prediction, prognosis, as well as assessing treatment response. Furthermore, radiogenomic approaches provide a mechanism to establish statistical correlations of radiomic features with point mutations and next-generation sequencing data to further leverage the potential of routine MRI scans to serve as "virtual biopsy" maps. In this review, we provide an introduction to radiomic and radiogenomic approaches in neuro-oncology, including a brief description of the workflow involving preprocessing, tumor segmentation, and extraction of "hand-crafted" features from the segmented region of interest, as well as identifying radiogenomic associations that could ultimately lead to the development of reliable prognostic and predictive models in neuro-oncology applications. Lastly, we discuss the promise of radiomics and radiogenomic approaches in personalizing treatment decisions in neuro-oncology, as well as the challenges with clinical adoption, which will rely heavily on their demonstrated resilience to nonstandardization in imaging protocols across sites and scanners, as well as in their ability to demonstrate reproducibility across large multi-institutional cohorts.

Deep learning for glioblastoma segmentation using preoperative magnetic resonance imaging identifies volumetric features associated with survival

BACKGROUND

Measurement of volumetric features is challenging in glioblastoma. We investigate whether volumetric features derived from preoperative MRI using a convolutional neural network-assisted segmentation is correlated with survival.

METHODS

Preoperative MRI of 120 patients were scored using Visually Accessible Rembrandt Images (VASARI) features. We trained and tested a multilayer, multi-scale convolutional neural network on multimodal brain tumour segmentation challenge (BRATS) data, prior to testing on our dataset. The automated labels were manually edited to generate ground truth segmentations. Network performance for our data and BRATS data was compared. Multivariable Cox regression analysis corrected for multiple testing using the false discovery rate was performed to correlate clinical and imaging variables with overall survival.

RESULTS

Median Dice coefficients in our sample were (1) whole tumour 0.94 (IQR, 0.82-0.98) compared to 0.91 (IQR, 0.83-0.94 p = 0.012), (2) FLAIR region 0.84 (IQR, 0.63-0.95) compared to 0.81 (IQR, 0.69-0.8 p = 0.170), (3) contrast-enhancing region 0.91 (IQR, 0.74-0.98) compared to 0.83 (IQR, 0.78-0.89 p = 0.003) and (4) necrosis region were 0.82 (IQR, 0.47-0.97) compared to 0.67 (IQR, 0.42-0.81 p = 0.005). Contrast-enhancing region/tumour core ratio (HR 4.73 [95% CI, 1.67-13.40], corrected p = 0.017) and necrotic core/tumour core ratio (HR 8.13 [95% CI, 2.06-32.12], corrected p = 0.011) were independently associated with overall survival.

CONCLUSION

Semi-automated segmentation of glioblastoma using a convolutional neural network trained on independent data is robust when applied to routine clinical data. The segmented volumes have prognostic significance.

Coactosin-Like Protein (COTL1) Promotes Glioblastoma (GBM) Growth in vitro and in vivo

Objective

To assess the expression levels of COTL1 in human GBM tissues and evaluate the potential involvement of COTL1 in cancer progression.

Methods

Bioinformation analysis was performed to evaluate COTL1 mRNA levels in GBM tissues and normal tissues, according to the TCGA database, and explore the effects on prognosis. Immunohistochemical (IHC) assays were performed to evaluate COTL1 expression in human GBM tissues and the clinical pathological analysis was performed. Colony formation and MTT assays were performed to evaluate the effects of COTL1 on GBM cell proliferation. Immunoblot assays were performed to detect the expression level of COTL1, Ki67, and PCNA. A xenograft model was developed in mice to assess the effects of COTL1 on tumor growth in vivo.

Results

We found COTL1 had an obvious high expression in human GBM tissues. The expression of COTL1 was related to recurrence (P=0.006**) and prognosis of patients with GBM. Our data further demonstrated COTL1 promoted cell proliferation in vitro and contributed to tumor growth of GBM cells in mice.

Conclusion

We therefore identified a novel and promising therapeutic target for the treatment of GBM.

How Reliable Are Volumetric Techniques for High-Grade Gliomas? A Comparison Study of Different Available Tools

BACKGROUND

Gliomas are the most common malignant primary brain tumors. Assessment of the tumor volume represents a crucial point in preoperative and postoperative evaluation.

OBJECTIVE

To compare pre- and postoperative tumor volumes obtained with an automated, semi-automatic, and manual segmentation tool. Mean processing time of each segmentation techniques was measured.

METHODS

Manual segmentation was performed on preoperative and postoperative magnetic resonance images with the open-source software Horos (Horos Project). "SmartBrush," a tool of the IPlan Cranial software (Brainlab, Feldkirchen, Germany), was used to carry out the semi-automatic segmentation. The open-source BraTumIA software (NeuroImaging Tools and Resources Collaboratory) was employed for the automated segmentation. Pearson correlation coefficient was used to assess volumetric comparison. Subsequently deviation/range and average discrepancy were determined. The Wilcoxon signed-rank test was used to assess statistical significance.

RESULTS

A total of 58 patients with a newly diagnosed high-grade glioma were enrolled. The comparison of the volumes calculated with Horos and IPlan showed a strong agreement both on preoperative and postoperative images (respectively: "enhancing" ρ = 0.99-0.78, "fluid-attenuated inversion recovery" ρ = 0.97-0.92, and "total tumor volume" ρ = 0.98-0.95). Agreement between BraTumIA and the other 2 techniques appeared to be strong for preoperative images, but showed a higher disagreement on postoperative images. Mean time expenditure for tumor segmentation was 27 min with manual segmentation, 17 min with semi-automated, and 8 min with automated software.

CONCLUSION

The considered segmentation tools showed high agreement in preoperative volumetric assessment. Both manual and semi-automated software appear adequate for the postoperative quantification of residual volume. The evaluated automated software is not yet reliable. Automated software considerably reduces the time expenditure.

Radiomics Features Predict CIC Mutation Status in Lower Grade Glioma

MRI in combination with genomic markers are critical in the management of gliomas. Radiomics and radiogenomics analysis facilitate the quantitative assessment of tumor properties which can be used to model both molecular subtype and predict disease progression. In this work, we report on the Drosophila gene capicua (CIC) mutation biomarker effects alongside radiomics features on the predictive ability of CIC mutation status in lower-grade gliomas (LGG). Genomic data of lower grade glioma (LGG) patients from The Cancer Genome Atlas (TCGA) (n = 509) and corresponding MR images from TCIA (n = 120) were utilized. Following tumor segmentation, radiomics features were extracted from T1, T2, T2 Flair, and T1 contrast enhanced (CE) images. Lasso feature reduction was used to obtain the most important MR image features and then logistic regression used to predict CIC mutation status. In our study, CIC mutation rarely occurred in Astrocytoma but has a high probability of occurrence in Oligodendroglioma. The presence of CIC mutation was found to be associated with better survival of glioma patients (p < 1e−4, HR: 0.2445), even with co-occurrence of IDH mutation and 1p/19q co-deletion (p = 0.0362, HR: 0.3674). An eleven-feature model achieved glioma prediction accuracy of 94.2% (95% CI, 94.03–94.38%), a six-feature model achieved oligodendroglioma prediction accuracy of 92.3% (95% CI, 91.70–92.92%). MR imaging and its derived image of gliomas with CIC mutation appears more complex and non-uniform but are associated with lower malignancy. Our study identified CIC as a potential prognostic factor in glioma which has close associations with survival. MRI radiomic features could predict CIC mutation, and reflect less malignant manifestations such as milder necrosis and larger tumor volume in MRI and its derived images that could help clinical judgment.

Overall survival prediction in glioblastoma patients using structural magnetic resonance imaging (MRI): advanced radiomic features may compensate for lack of advanced MRI modalities

Purpose: Glioblastoma, the most common and aggressive adult brain tumor, is considered noncurative at diagnosis, with 14 to 16 months median survival following treatment. There is increasing evidence that noninvasive integrative analysis of radiomic features can predict overall and progression-free survival, using advanced multiparametric magnetic resonance imaging (Adv-mpMRI). If successfully applicable, such noninvasive markers can considerably influence patient management. However, most patients prior to initiation of therapy typically undergo only basic structural mpMRI (Bas-mpMRI, i.e., T1, T1-Gd, T2, and T2-fluid-attenuated inversion recovery) preoperatively, rather than Adv-mpMRI that provides additional vascularization (dynamic susceptibility contrast-MRI) and cell-density (diffusion tensor imaging) related information. Approach: We assess a retrospective cohort of 101 glioblastoma patients with available Adv-mpMRI from a previous study, which has shown that an initial feature panel (IFP, i.e., intensity, volume, location, and growth model parameters) extracted from Adv-mpMRI can yield accurate overall survival stratification. We focus on demonstrating that equally accurate prediction models can be constructed using augmented radiomic feature panels (ARFPs, i.e., integrating morphology and textural descriptors) extracted solely from widely available Bas-mpMRI, obviating the need for using Adv-mpMRI. We extracted 1612 radiomic features from distinct tumor subregions to build multivariate models that stratified patients as long-, intermediate-, or short-survivors. Results: The classification accuracy of the model utilizing Adv-mpMRI protocols and the IFP was 72.77% and degraded to 60.89% when using only Bas-mpMRI. However, utilizing the ARFP on Bas-mpMRI improved the accuracy to 74.26%. Furthermore, Kaplan-Meier analysis demonstrated superior classification of subjects into short-, intermediate-, and long-survivor classes when using ARFP extracted from Bas-mpMRI. Conclusions: This quantitative evaluation indicates that accurate survival prediction in glioblastoma patients is feasible using solely Bas-mpMRI and integrative advanced radiomic features, which can compensate for the lack of Adv-mpMRI. Our finding holds promise for generalization across multiple institutions that may not have access to Adv-mpMRI and to better inform clinical decision-making about aggressive interventions and clinical trials.

Convolutional neural networks for brain tumour segmentation

The introduction of quantitative image analysis has given rise to fields such as radiomics which have been used to predict clinical sequelae. One growing area of interest for analysis is brain tumours, in particular glioblastoma multiforme (GBM). Tumour segmentation is an important step in the pipeline in the analysis of this pathology. Manual segmentation is often inconsistent as it varies between observers. Automated segmentation has been proposed to combat this issue. Methodologies such as convolutional neural networks (CNNs) which are machine learning pipelines modelled on the biological process of neurons (called nodes) and synapses (connections) have been of interest in the literature. We investigate the role of CNNs to segment brain tumours by firstly taking an educational look at CNNs and perform a literature search to determine an example pipeline for segmentation. We then investigate the future use of CNNs by exploring a novel field-radiomics. This examines quantitative features of brain tumours such as shape, texture, and signal intensity to predict clinical outcomes such as survival and response to therapy.

Integrated Biophysical Modeling and Image Analysis: Application to Neuro-Oncology

Central nervous system (CNS) tumors come with vastly heterogeneous histologic, molecular, and radiographic landscapes, rendering their precise characterization challenging. The rapidly growing fields of biophysical modeling and radiomics have shown promise in better characterizing the molecular, spatial, and temporal heterogeneity of tumors. Integrative analysis of CNS tumors, including clinically acquired multi-parametric magnetic resonance imaging (mpMRI) and the inverse problem of calibrating biophysical models to mpMRI data, assists in identifying macroscopic quantifiable tumor patterns of invasion and proliferation, potentially leading to improved (a) detection/segmentation of tumor subregions and (b) computer-aided diagnostic/prognostic/predictive modeling. This article presents a summary of (a) biophysical growth modeling and simulation,(b) inverse problems for model calibration, (c) these models' integration with imaging workflows, and (d) their application to clinically relevant studies. We anticipate that such quantitative integrative analysis may even be beneficial in a future revision of the World Health Organization (WHO) classification for CNS tumors, ultimately improving patient survival prospects.

Analysis of metabolic abnormalities in high-grade glioma using MRSI and convex NMF

Clinical use of MRSI is limited by the level of experience required to properly translate MRSI examinations into relevant clinical information. To solve this, several methods have been proposed to automatically recognize a predefined set of reference metabolic patterns. Given the variety of metabolic patterns seen in glioma patients, the decision on the optimal number of patterns that need to be used to describe the data is not trivial. In this paper, we propose a novel framework to (1) separate healthy from abnormal metabolic patterns and (2) retrieve an optimal number of reference patterns describing the most important types of abnormality. Using 41 MRSI examinations (1.5 T, PRESS, T_E 135 ms) from 22 glioma patients, four different patterns describing different types of abnormality were detected: edema, healthy without Glx, active tumor and necrosis. The identified patterns were then evaluated on 17 MRSI examinations from nine different glioma patients. The results were compared against BraTumIA, an automatic segmentation method trained to identify different tumor compartments on structural MRI data. Finally, the ability to predict future contrast enhancement using the proposed approach was also evaluated.

Optimizing Neuro-Oncology Imaging: A Review of Deep Learning Approaches for Glioma Imaging

Radiographic assessment with magnetic resonance imaging (MRI) is widely used to characterize gliomas, which represent 80% of all primary malignant brain tumors. Unfortunately, glioma biology is marked by heterogeneous angiogenesis, cellular proliferation, cellular invasion, and apoptosis. This translates into varying degrees of enhancement, edema, and necrosis, making reliable imaging assessment challenging. Deep learning, a subset of machine learning artificial intelligence, has gained traction as a method, which has seen effective employment in solving image-based problems, including those in medical imaging. This review seeks to summarize current deep learning applications used in the field of glioma detection and outcome prediction and will focus on (1) pre- and post-operative tumor segmentation, (2) genetic characterization of tissue, and (3) prognostication. We demonstrate that deep learning methods of segmenting, characterizing, grading, and predicting survival in gliomas are promising opportunities that may enhance both research and clinical activities.

Intergrating conventional MRI, texture analysis of dynamic contrast-enhanced MRI, and susceptibility weighted imaging for glioma grading

BACKGROUND

The application of conventional magnetic resonance imaging (MRI) in glioma grading is limited and non-specific.

PURPOSE

To investigate the application values of MRI, texture analysis (TA) of dynamic contrast-enhanced MRI (DCE-MRI) and intratumoral susceptibility signal (ITSS) on susceptibility weighted imaging (SWI), alone and in combination, for glioma grading.

MATERIAL AND METHODS

Fifty-two patients with pathologically confirmed gliomas who underwent DCE-MRI and SWI were enrolled in this retrospective study. Conventional MRIs were evaluated by the VASARI scoring system. TA of DCE-MRI-derived parameters and the degree of ITSS were compared between low-grade gliomas (LGGs) and high-grade gliomas (HGGs). The diagnostic ability of each parameter and their combination for glioma grading were analyzed.

RESULTS

Significant statistical differences in VASARI features were observed between LGGs and HGGs ( P < 0.05), of which the enhancement quality had the highest area under the curve (AUC) (0.873) with 93.3% sensitivity and 80% specificity. The TA of DCE-MRI derived parameters were significantly different between LGGs and HGGs ( P < 0.05), of which the uniformity of K^trans had the highest AUC (0.917) with 93.3% sensitivity and 90% specificity. The degree of ITSS was significantly different between LGGs and HGGs ( P < 0.001). The AUC of the ITSS was 0.925 with 93.3% sensitivity and 90% specificity. The best discriminative power was obtained from a combination of enhancement quality, K^trans- uniformity, and ITSS, resulting in 96.7% sensitivity, 100.0% specificity, and AUC of 0.993.

CONCLUSION

Combining conventional MRI, TA of DCE-MRI, and ITSS on SWI may help to improve the differentiation between LGGs and HGGs.

Correlation of post-contrast T1-weighted MRI surface regularity, tumor bulk, and necrotic volume with Ki67 and p53 in glioblastomas

PURPOSE

p53 and Ki67 status can be relevant to the management of glioblastoma. The goal of this study is to determine whether tumor morphology and bulk depicted on MRI correlate with p53 and Ki67 in glioblastoma.

METHODS

A retrospective review of 223 patients with glioblastoma and corresponding p53 or Ki67 status, along with T1-weighted post-contrast MR images was performed. Enhancing tumors were outlined for determining surface regularity, tumor bulk, and necrotic volume. The median value of 0.1 was chosen for p53 and 0.2 for Ki67 to separate each data set into two classes. T tests and receiver operating characteristic analysis were performed to determine the separation of the classes and the predicting power of each feature.

RESULTS

There were significant differences between tumor surface regularity (p = 0.01) and necrotic volume (p = 0.0429) according to Ki67 levels, although neither had statistically significant predictive power (AUC = 0.697, p = 0.0506 and AUC = 0.577, p = 0.164, respectively). There were also significant differences between tumor bulk (p = 0.0239) and necrotic volume (p = 0.0200) according to p53 levels, but again no significant predictive power was found using ROC analysis (AUC = 0.5882, p = 0.0894 and AUC = 0.567, p = 0.155, respectively).

CONCLUSION

Quantitative morphological tumor characteristics on post-contrast T1-weighted MRI can to a certain degree provide insights regarding Ki67 and p53 status in patients with glioblastoma.

Artificial intelligence in cancer imaging: Clinical challenges and applications

Judgement, as one of the core tenets of medicine, relies upon the integration of multilayered data with nuanced decision making. Cancer offers a unique context for medical decisions given not only its variegated forms with evolution of disease but also the need to take into account the individual condition of patients, their ability to receive treatment, and their responses to treatment. Challenges remain in the accurate detection, characterization, and monitoring of cancers despite improved technologies. Radiographic assessment of disease most commonly relies upon visual evaluations, the interpretations of which may be augmented by advanced computational analyses. In particular, artificial intelligence (AI) promises to make great strides in the qualitative interpretation of cancer imaging by expert clinicians, including volumetric delineation of tumors over time, extrapolation of the tumor genotype and biological course from its radiographic phenotype, prediction of clinical outcome, and assessment of the impact of disease and treatment on adjacent organs. AI may automate processes in the initial interpretation of images and shift the clinical workflow of radiographic detection, management decisions on whether or not to administer an intervention, and subsequent observation to a yet to be envisioned paradigm. Here, the authors review the current state of AI as applied to medical imaging of cancer and describe advances in 4 tumor types (lung, brain, breast, and prostate) to illustrate how common clinical problems are being addressed. Although most studies evaluating AI applications in oncology to date have not been vigorously validated for reproducibility and generalizability, the results do highlight increasingly concerted efforts in pushing AI technology to clinical use and to impact future directions in cancer care.

MRI Features Associated with TERT Promoter Mutation Status in Glioblastoma

BACKGROUND AND PURPOSE

Telomerase reverse transcriptase (TERT) promoter mutations are associated with worse prognosis in glioblastoma. The purpose of this study was to evaluate whether TERT mutation status was associated with specific morphologic and quantitative imaging features.

METHODS

Twenty-nine patients with isocitrate dehydrogenase 1/2-wildtype glioblastoma (13 TERT-wildtype, 16 TERT-mutated), who underwent preoperative magnetic resonance (MR) imaging were included in this retrospective study. Qualitative imaging phenotypes were evaluated using the Visually Accessible Rembrandt Images (VASARIs) feature set. Histogram analysis of apparent diffusion coefficient (ADC) and dynamic contrast-enhanced MR perfusion values were performed on enhancing tumor volumes-of-interest, and differences between TERT-wildtype and TERT-mutated tumors were assessed.

RESULTS

VASARI analysis demonstrated that the majority of morphologic features were not significantly different between TERT-wildtype and TERT-mutated tumors, although a higher proportion of TERT-wildtype tumors featured nonenhancing tumor crossing midline (P = .014). TERT-mutated tumors demonstrated lower median rate constant k_ep (.38 vs. .76, P = .03) and lower median volume transfer coefficient K^trans (.13 vs. .31, P = .02). There was no significant difference in median plasma volume v_p (P = .92) or ADC values (P = .66) between the two groups. We further found a significant interaction between median k_ep and K^trans and TERT status, respectively, suggesting greater risk of death with increasing blood-brain barrier dysfunction in TERT-mutated but not in TERT-wildtype tumors.

CONCLUSION

Our study demonstrates evidence of altered permeability metrics associated with TERT mutation in glioblastoma, laying the foundation for future prospective studies assessing implications for therapeutic management and clinical outcomes.

Radiogenomics for Precision Medicine With a Big Data Analytics Perspective

Precision medicine promises better healthcare delivery by improving clinical practice. Using evidence-based substratification of patients, the objective is to achieve better prognosis, diagnosis, and treatment that will transform existing clinical pathways toward optimizing care for the specific needs of each patient. The wealth of today's healthcare data, often characterized as big data, provides invaluable resources toward new knowledge discovery that has the potential to advance precision medicine. The latter requires interdisciplinary efforts that will capitalize the information, know-how, and medical data of newly formed groups fusing different backgrounds and expertise. The objective of this paper is to provide insights with respect to the state-of-the-art research in precision medicine. More specifically, our goal is to highlight the fundamental challenges in emerging fields of radiomics and radiogenomics by reviewing the case studies of Cancer and Alzheimer's disease, describe the computational challenges from a big data analytics perspective, and discuss standardization and open data initiatives that will facilitate the adoption of precision medicine methods and practices.

Preoperative Two-Dimensional Size of Glioblastoma is Associated with Patient Survival

BACKGROUND

Although tumor size affects survival of patients with lower-grade glioma, a prognostic effect on patients with glioblastoma remains to be established.

METHODS

We performed a retrospective analysis of 61 patients using volumetric data of tumor compartments of 61 patients obtained by preoperative magnetic resonance images using the visual ABC/2 method. Preoperative enhancing, nonenhancing, necrosis, and edema volume, the preoperative tumor area (TA) as a product of the 2 largest tumor diameters perpendicular to each other on axial T1-weighted postcontrast images, as well as postoperative enhancing residual volumes, were measured. Multivariable Cox proportional hazard models were used to associate these parameters with overall survival, adjusting for potential confounders.

RESULTS

The median preoperative enhancing tumor volume was 18.2 mL (interquartile range, 8.2-41.7 mL); the median remnant tumor volume was 1.3% (interquartile range, 0.0%-42.9%). During follow-up, 59 patients (92%) died; median survival time and median follow-up time were both 404 days. We found a statistically significant multiplicative effect of TA on survival: the hazard ratio (HR) was increased by 1.096 per unit increase of 200 mm² (95% confidence interval [CI], 1.027-1.170; P < 0.01). The effect of remnant tumor on HR increased multiplicatively by 1.013 (95% CI, 1.001-1.026; P = 0.04) per unit increase of 1 log (day) and 1% in tumor remnant. HR associated with age at surgery increased by 1.503 per 5 years of age (95% CI, 1.243-1.817; P < 0.01).

CONCLUSIONS

Preoperative TA proved to be the only glioblastoma size parameter that affects patient survival.

Experimental Texture Analysis in Glioblastoma: A Methodological Study

OBJECTIVES

Analysis of a single slice of a tumor to extract biomarkers for texture analysis may result in loss of information. We investigated correlation of fractional volumes to entire tumor volumes and introduced expanded regions of interest (ROIs) outside the visual tumor borders in glioblastoma.

MATERIALS AND METHODS

Retrospective slice-by-slice volumetric texture analysis on 46 brain magnetic resonance imaging subjects with histologically confirmed glioblastoma was performed. Fractional volumes were analyzed for correlation to total volume. Expanded ROIs were analyzed for significant differences to conservative ROIs.

RESULTS

As fractional tumor volumes increased, correlation with total volume values for mean, SD, mean of positive pixels, skewness, and kurtosis increased. Expanding ROI by 2 mm resulted in significant differences in all textural values.

CONCLUSIONS

Fractional volumes may provide an optimal trade-off for texture analysis in the clinical setting. All texture parameters proved significantly different with minimal expansion of the ROI, underlining the susceptibility of texture analysis to generating misrepresentative tumor information.

Identification of a gene expression signature predicting survival in oral cavity squamous cell carcinoma using Monte Carlo cross validation

OBJECTIVES

This study aims to identify a robust signature that performs well in predicting overall survival across tumor phenotypes and treatment strata, and validates the application of Monte Carlo cross validation (MCCV) as a means of identifying molecular signatures when utilizing small and highly heterogeneous datasets.

MATERIALS AND METHODS

RNA sequence gene expression data for 264 patient tumors were acquired from The Cancer Genome Atlas (TCGA). 100 iterations of Monte Carlo cross validation were applied to differential expression and Cox model validation. The association between the gene signature risk score and overall survival was measured using Kaplan-Meier survival curves, univariate, and multivariable Cox regression analyses.

RESULTS

Pathway analysis findings indicate that ligand-gated ion channel pathways are the most significantly enriched with the genes in the aggregated signature. The aggregated signature described in this study is predictive of overall survival in oral cancer patients across demographic and treatment strata.

CONCLUSION

This study reinforces previous findings supporting the role of ion channel gating, interleukin, calcitonin receptor, and keratinization pathways in tumor progression and treatment response in oral cancer. These results strengthen the argument that differential expression of genes within these pathways reduces tumor susceptibility to treatment. Conducting differential gene expression (DGE) with Monte Carlo cross validation, as this study describes, offers a potential solution to decreasing the variability in DGE results across future studies that are reliant upon highly heterogeneous datasets. This improves the ability of studies reliant upon similarly structured datasets to reach results that are reproducible.