Volumetric segmentation of glioblastoma progression compared to bidimensional products and clinical radiological reports

RANO 2.0 criteria: concepts applicable to the neuroradiologist's clinical practice

PURPOSE OF REVIEW

The Response Assessment in Neuro-Oncology (RANO) 2.0 criteria aim at improving the standardization and reliability of treatment response assessment in clinical trials studying central nervous system (CNS) gliomas. This review presents the evidence supporting RANO 2.0 updates and discusses which concepts can be applicable to the clinical practice, particularly in the clinical radiographic reads.

RECENT FINDINGS

Updates in RANO 2.0 were supported by recent retrospective analyses of multicenter data from recent clinical trials. As proposed in RANO 2.0, in tumors receiving radiation therapy, the post-RT MRI scan should be used as a reference baseline for the following scans, as opposed to the pre-RT scan, and radiographic findings suggesting progression within three months after radiation therapy completion should be verified with confirmatory scans. Volumetric assessments should be considered, when available, especially for low-grade gliomas, and the evaluation of nonenhancing disease should have a marginal role in glioblastoma. However, the radiographic reads in the clinical setting also benefit from aspects that lie outside RANO 2.0 criteria, such as qualitative evaluations, patient-specific clinical considerations, and advanced imaging.

SUMMARY

While RANO 2.0 criteria are meant for the standardization of the response assessment in clinical trials, some concepts have the potential to improve patients' management in the clinical practice.

The peritumoral edema index and related mechanisms influence the prognosis of GBM patients

Background

Peritumoral brain edema (PTBE) represents a characteristic phenotype of intracranial gliomas. However, there is a lack of consensus regarding the prognosis and mechanism of PTBE. In this study, clinical imaging data, along with publicly available imaging data, were utilized to assess the prognosis of PTBE in glioblastoma (GBM) patients, and the associated mechanisms were preliminarily analyzed.

Methods

We investigated relevant imaging features, including edema, in GBM patients using ITK-SNAP imaging segmentation software. Risk factors affecting progression-free survival (PFS) and overall survival (OS) were assessed using a Cox proportional hazard regression model. In addition, the impact of PTBE on PFS and OS was analyzed in clinical GBM patients using the Kaplan-Meier survival analysis method, and the results further validated by combining data from The Cancer Imaging Archive (TCIA) and The Cancer Genome Atlas (TCGA). Finally, functional enrichment analysis based on TCIA and TCGA datasets identified several pathways potentially involved in the mechanism of edema formation.

Results

This study included a total of 32 clinical GBM patients and 132 GBM patients from public databases. Univariate and multivariate analyses indicated that age and edema index (EI) are independent risk factors for PFS, but not for OS. Kaplan-Meier curves revealed consistent survival analysis results between IE groups among both clinical patients and TCIA and TCGA patients, suggesting a significant effect of PTBE on PFS but not on OS. Furthermore, functional enrichment analysis predicted the involvement of several pathways related mainly to cellular bioenergetics and vasculogenic processes in the mechanism of PTBE formation. While these novel results warrant confirmation in a larger patient cohort, they support good prognostic value for PTBE assessment in GBM.

Conclusions

Our results indicate that a low EI positively impacts disease control in GBM patients, but this does not entirely translate into an improvement in OS. Multiple genes, signaling pathways, and biological processes may contribute to the formation of peritumoral edema in GBM through cytotoxic and vascular mechanisms.

Not without Context-A Multiple Methods Study on Evaluation and Correction of Automated Brain Tumor Segmentations by Experts

RATIONALE AND OBJECTIVES

Brain tumor segmentations are integral to the clinical management of patients with glioblastoma, the deadliest primary brain tumor in adults. The manual delineation of tumors is time-consuming and highly provider-dependent. These two problems must be addressed by introducing automated, deep-learning-based segmentation tools. This study aimed to identify criteria experts use to evaluate the quality of automatically generated segmentations and their thought processes as they correct them.

MATERIALS AND METHODS

Multiple methods were used to develop a detailed understanding of the complex factors that shape experts' perception of segmentation quality and their thought processes in correcting proposed segmentations. Data from a questionnaire and semistructured interview with neuro-oncologists and neuroradiologists were collected between August and December 2021 and analyzed using a combined deductive and inductive approach.

RESULTS

Brain tumors are highly complex and ambiguous segmentation targets. Therefore, physicians rely heavily on the given context related to the patient and clinical context in evaluating the quality and need to correct brain tumor segmentation. Most importantly, the intended clinical application determines the segmentation quality criteria and editing decisions. Physicians' personal beliefs and preferences about the capabilities of AI algorithms and whether questionable areas should not be included are additional criteria influencing the perception of segmentation quality and appearance of an edited segmentation.

CONCLUSION

Our findings on experts' perceptions of segmentation quality will allow the design of improved frameworks for expert-centered evaluation of brain tumor segmentation models. In particular, the knowledge presented here can inspire the development of brain tumor-specific metrics for segmentation model training and evaluation.

Standardized brain tumor imaging protocols for clinical trials: current recommendations and tips for integration

Standardized MRI acquisition protocols are crucial for reducing the measurement and interpretation variability associated with response assessment in brain tumor clinical trials. The main challenge is that standardized protocols should ensure high image quality while maximizing the number of institutions meeting the acquisition requirements. In recent years, extensive effort has been made by consensus groups to propose different "ideal" and "minimum requirements" brain tumor imaging protocols (BTIPs) for gliomas, brain metastases (BM), and primary central nervous system lymphomas (PCSNL). In clinical practice, BTIPs for clinical trials can be easily integrated with additional MRI sequences that may be desired for clinical patient management at individual sites. In this review, we summarize the general concepts behind the choice and timing of sequences included in the current recommended BTIPs, we provide a comparative overview, and discuss tips and caveats to integrate additional clinical or research sequences while preserving the recommended BTIPs. Finally, we also reflect on potential future directions for brain tumor imaging in clinical trials.

Segmentation of glioblastomas in early post-operative multi-modal MRI with deep neural networks

Extent of resection after surgery is one of the main prognostic factors for patients diagnosed with glioblastoma. To achieve this, accurate segmentation and classification of residual tumor from post-operative MR images is essential. The current standard method for estimating it is subject to high inter- and intra-rater variability, and an automated method for segmentation of residual tumor in early post-operative MRI could lead to a more accurate estimation of extent of resection. In this study, two state-of-the-art neural network architectures for pre-operative segmentation were trained for the task. The models were extensively validated on a multicenter dataset with nearly 1000 patients, from 12 hospitals in Europe and the United States. The best performance achieved was a 61% Dice score, and the best classification performance was about 80% balanced accuracy, with a demonstrated ability to generalize across hospitals. In addition, the segmentation performance of the best models was on par with human expert raters. The predicted segmentations can be used to accurately classify the patients into those with residual tumor, and those with gross total resection.

"A net for everyone": fully personalized and unsupervised neural networks trained with longitudinal data from a single patient

BACKGROUND

With the rise in importance of personalized medicine and deep learning, we combine the two to create personalized neural networks. The aim of the study is to show a proof of concept that data from just one patient can be used to train deep neural networks to detect tumor progression in longitudinal datasets.

METHODS

Two datasets with 64 scans from 32 patients with glioblastoma multiforme (GBM) were evaluated in this study. The contrast-enhanced T1w sequences of brain magnetic resonance imaging (MRI) images were used. We trained a neural network for each patient using just two scans from different timepoints to map the difference between the images. The change in tumor volume can be calculated with this map. The neural networks were a form of a Wasserstein-GAN (generative adversarial network), an unsupervised learning architecture. The combination of data augmentation and the network architecture allowed us to skip the co-registration of the images. Furthermore, no additional training data, pre-training of the networks or any (manual) annotations are necessary.

RESULTS

The model achieved an AUC-score of 0.87 for tumor change. We also introduced a modified RANO criteria, for which an accuracy of 66% can be achieved.

CONCLUSIONS

We show a novel approach to deep learning in using data from just one patient to train deep neural networks to monitor tumor change. Using two different datasets to evaluate the results shows the potential to generalize the method.

Deep learning-based algorithm for postoperative glioblastoma MRI segmentation: a promising new tool for tumor burden assessment

OBJECTIVE

Clinical and surgical decisions for glioblastoma patients depend on a tumor imaging-based evaluation. Artificial Intelligence (AI) can be applied to magnetic resonance imaging (MRI) assessment to support clinical practice, surgery planning and prognostic predictions. In a real-world context, the current obstacles for AI are low-quality imaging and postoperative reliability. The aim of this study is to train an automatic algorithm for glioblastoma segmentation on a clinical MRI dataset and to obtain reliable results both pre- and post-operatively.

METHODS

The dataset used for this study comprises 237 (71 preoperative and 166 postoperative) MRIs from 71 patients affected by a histologically confirmed Grade IV Glioma. The implemented U-Net architecture was trained by transfer learning to perform the segmentation task on postoperative MRIs. The training was carried out first on BraTS2021 dataset for preoperative segmentation. Performance is evaluated using DICE score (DS) and Hausdorff 95% (H95).

RESULTS

In preoperative scenario, overall DS is 91.09 (± 0.60) and H95 is 8.35 (± 1.12), considering tumor core, enhancing tumor and whole tumor (ET and edema). In postoperative context, overall DS is 72.31 (± 2.88) and H95 is 23.43 (± 7.24), considering resection cavity (RC), gross tumor volume (GTV) and whole tumor (WT). Remarkably, the RC segmentation obtained a mean DS of 63.52 (± 8.90) in postoperative MRIs.

CONCLUSIONS

The performances achieved by the algorithm are consistent with previous literature for both pre-operative and post-operative glioblastoma's MRI evaluation. Through the proposed algorithm, it is possible to reduce the impact of low-quality images and missing sequences.

Evolution and implementation of radiographic response criteria in neuro-oncology

Radiographic response assessment in neuro-oncology is critical in clinical practice and trials. Conventional criteria, such as the MacDonald and response assessment in neuro-oncology (RANO) criteria, rely on bidimensional (2D) measurements of a single tumor cross-section. Although RANO criteria are established for response assessment in clinical trials, there is a critical need to address the complexity of brain tumor treatment response with multiple new approaches being proposed. These include volumetric analysis of tumor compartments, structured MRI reporting systems like the Brain Tumor Reporting and Data System, and standardized approaches to advanced imaging techniques to distinguish tumor response from treatment effects. In this review, we discuss the strengths and limitations of different neuro-oncology response criteria and summarize current research findings on the role of novel response methods in neuro-oncology clinical trials and practice.

Preoperative Brain Tumor Imaging: Models and Software for Segmentation and Standardized Reporting

For patients suffering from brain tumor, prognosis estimation and treatment decisions are made by a multidisciplinary team based on a set of preoperative MR scans. Currently, the lack of standardized and automatic methods for tumor detection and generation of clinical reports, incorporating a wide range of tumor characteristics, represents a major hurdle. In this study, we investigate the most occurring brain tumor types: glioblastomas, lower grade gliomas, meningiomas, and metastases, through four cohorts of up to 4,000 patients. Tumor segmentation models were trained using the AGU-Net architecture with different preprocessing steps and protocols. Segmentation performances were assessed in-depth using a wide-range of voxel and patient-wise metrics covering volume, distance, and probabilistic aspects. Finally, two software solutions have been developed, enabling an easy use of the trained models and standardized generation of clinical reports: Raidionics and Raidionics-Slicer. Segmentation performances were quite homogeneous across the four different brain tumor types, with an average true positive Dice ranging between 80 and 90%, patient-wise recall between 88 and 98%, and patient-wise precision around 95%. In conjunction to Dice, the identified most relevant other metrics were the relative absolute volume difference, the variation of information, and the Hausdorff, Mahalanobis, and object average symmetric surface distances. With our Raidionics software, running on a desktop computer with CPU support, tumor segmentation can be performed in 16-54 s depending on the dimensions of the MRI volume. For the generation of a standardized clinical report, including the tumor segmentation and features computation, 5-15 min are necessary. All trained models have been made open-access together with the source code for both software solutions and validation metrics computation. In the future, a method to convert results from a set of metrics into a final single score would be highly desirable for easier ranking across trained models. In addition, an automatic classification of the brain tumor type would be necessary to replace manual user input. Finally, the inclusion of post-operative segmentation in both software solutions will be key for generating complete post-operative standardized clinical reports.

Ellipsoid calculations versus manual tumor delineations for glioblastoma tumor volume evaluation

In glioblastoma, the response to treatment assessment is essentially based on the 2D tumor size evolution but remains disputable. Volumetric approaches were evaluated for a more accurate estimation of tumor size. This study included 57 patients and compared two volume measurement methods to determine the size of different glioblastoma regions of interest: the contrast-enhancing area, the necrotic area, the gross target volume and the volume of the edema area. The two methods, the ellipsoid formula (the calculated method) and the manual delineation (the measured method) showed a high correlation to determine glioblastoma volume and a high agreement to classify patients assessment response to treatment according to RANO criteria. This study revealed that calculated and measured methods could be used in clinical practice to estimate glioblastoma volume size and to evaluate tumor size evolution.

The impact of different volumetric thresholds to determine progressive disease in patients with recurrent glioblastoma treated with bevacizumab

Background

The optimal volumetric threshold for determining progressive disease (PD) in recurrent glioblastoma is yet to be determined. We investigated a range of thresholds in association with overall survival (OS).

Methods

First recurrent glioblastoma patients treated with bevacizumab and/or lomustine were included from the phase II BELOB and phase III EORTC26101 trials. Enhancing and nonenhancing tumor volumes were measured at baseline, first (6 weeks), and second (12 weeks) follow-up. Hazard ratios (HRs) for the appearance of new lesions and several thresholds for tumor volume increase were calculated using cox regression analysis. Results were corrected in a multivariate analysis for well-established prognostic factors.

Results

At first and second follow-up, 138 and 94 patients respectively, were deemed eligible for analysis of enhancing volumes, while 89 patients were included in the analysis of nonenhancing volumes at first follow-up. New lesions were associated with a significantly worse OS (3.2 versus 11.2 months, HR = 7.03, P < .001). At first follow-up a threshold of enhancing volume increase of ≥20% provided the highest HR (5.55, p = .001. At second follow-up, any increase in enhancing volume (≥0%) provided the highest HR (9.00, p < .001). When measuring nonenhancing volume at first follow-up, only 6 additional patients were scored as PD with the highest HR of ≥25% increase in volume (HR=3.25, p = .008).

Conclusion

Early appearing new lesions were associated with poor OS. Lowering the volumetric threshold for PD at both first and second follow-up improved survival prediction. However, the additional number of patients categorized as PD by lowering the threshold was very low. The per-RANO added change in nonenhancing volumes to the analyses was of limited value.

Prediction of Malignant Transformation of WHO II Astrocytoma Using Mathematical Models Incorporating Apparent Diffusion Coefficient and Contrast Enhancement

Using only increasing contrast enhancement as a marker of malignant transformation (MT) in gliomas has low specificity and may affect interpretation of clinical outcomes. Therefore we developed a mathematical model to predict MT of low-grade gliomas (LGGs) by considering areas of reduced apparent diffusion coefficient (ADC) with increased contrast enhancement. Patients with contrast-enhancing LGGs who had contemporaneous ADC and histopathology were retrospectively analyzed. Multiple clinical factors and imaging factors (contrast-enhancement size, whole-tumor size, and ADC) were assessed for association with MT. Patients were split into training and validation groups for the development of a predictive model using logistic regression which was assessed with receiver operating characteristic analysis. Among 132 patients, (median age 46.5 years), 106 patients (64 MT) were assigned to the training group and 26 (20 MT) to the validation group. The predictive model comprised age (P = 0.110), radiotherapy (P = 0.168), contrast-enhancement size (P = 0.015), and ADC (P < 0.001). The predictive model (area-under-the-curve [AUC] 0.87) outperformed ADC (AUC 0.85) and contrast-enhancement size (AUC 0.67). The model had an accuracy of 84% for the training group and 85% respectively for the validation group. Our model incorporating ADC and contrast-enhancement size predicted MT in contrast-enhancing LGGs.

Meningioma Segmentation in T1-Weighted MRI Leveraging Global Context and Attention Mechanisms

Purpose: Meningiomas are the most common type of primary brain tumor, accounting for ~30% of all brain tumors. A substantial number of these tumors are never surgically removed but rather monitored over time. Automatic and precise meningioma segmentation is, therefore, beneficial to enable reliable growth estimation and patient-specific treatment planning. Methods: In this study, we propose the inclusion of attention mechanisms on top of a U-Net architecture used as backbone: (i) Attention-gated U-Net (AGUNet) and (ii) Dual Attention U-Net (DAUNet), using a three-dimensional (3D) magnetic resonance imaging (MRI) volume as input. Attention has the potential to leverage the global context and identify features' relationships across the entire volume. To limit spatial resolution degradation and loss of detail inherent to encoder-decoder architectures, we studied the impact of multi-scale input and deep supervision components. The proposed architectures are trainable end-to-end and each concept can be seamlessly disabled for ablation studies. Results: The validation studies were performed using a five-fold cross-validation over 600 T1-weighted MRI volumes from St. Olavs Hospital, Trondheim University Hospital, Norway. Models were evaluated based on segmentation, detection, and speed performances, and results are reported patient-wise after averaging across all folds. For the best-performing architecture, an average Dice score of 81.6% was reached for an F1-score of 95.6%. With an almost perfect precision of 98%, meningiomas smaller than 3 ml were occasionally missed hence reaching an overall recall of 93%. Conclusion: Leveraging global context from a 3D MRI volume provided the best performances, even if the native volume resolution could not be processed directly due to current GPU memory limitations. Overall, near-perfect detection was achieved for meningiomas larger than 3 ml, which is relevant for clinical use. In the future, the use of multi-scale designs and refinement networks should be further investigated. A larger number of cases with meningiomas below 3 ml might also be needed to improve the performance for the smallest tumors.

Glioblastoma Surgery Imaging-Reporting and Data System: Validation and Performance of the Automated Segmentation Task

For patients with presumed glioblastoma, essential tumor characteristics are determined from preoperative MR images to optimize the treatment strategy. This procedure is time-consuming and subjective, if performed by crude eyeballing or manually. The standardized GSI-RADS aims to provide neurosurgeons with automatic tumor segmentations to extract tumor features rapidly and objectively. In this study, we improved automatic tumor segmentation and compared the agreement with manual raters, describe the technical details of the different components of GSI-RADS, and determined their speed. Two recent neural network architectures were considered for the segmentation task: nnU-Net and AGU-Net. Two preprocessing schemes were introduced to investigate the tradeoff between performance and processing speed. A summarized description of the tumor feature extraction and standardized reporting process is included. The trained architectures for automatic segmentation and the code for computing the standardized report are distributed as open-source and as open-access software. Validation studies were performed on a dataset of 1594 gadolinium-enhanced T1-weighted MRI volumes from 13 hospitals and 293 T1-weighted MRI volumes from the BraTS challenge. The glioblastoma tumor core segmentation reached a Dice score slightly below 90%, a patientwise F1-score close to 99%, and a 95th percentile Hausdorff distance slightly below 4.0 mm on average with either architecture and the heavy preprocessing scheme. A patient MRI volume can be segmented in less than one minute, and a standardized report can be generated in up to five minutes. The proposed GSI-RADS software showed robust performance on a large collection of MRI volumes from various hospitals and generated results within a reasonable runtime.

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.

Spinal Meningiomas: Influence of Cord Compression and Radiological Features on Preoperative Functional Status and Outcome

Simple Summary

Patients affected by spinal meningioma globally exhibit an excellent postoperative functional outcome despite tumor size and severity of spinal cord compression. This work aims to analyze the impact of tumor size and other radiological findings on clinical functional preoperative status and postoperative outcome. In this multicentric retrospective study involving 90 adult patients, we found that spinal cord and tumor occupancy as well as cord compression and tumor volume are correlated with low preoperative functional status. Higher tumor occupancy and tumor-canal volume ratio corresponded to lower postoperative neurological recovery. Cord re-expansion did not show any correlation with postoperative outcome, while preoperative signs of cord myelopathy were predictors of worse postoperative outcome. These findings suggest that these radiological features should be taken into consideration during preoperative counselling.

Abstract

Background: Radiological parameters predicting the postoperative neurological outcome after resection of a spinal meningioma (SM) are poorly studied, with controversial results. Methods: Observational multicenter cohort (2011–2018) of adult patients undergoing surgery for resection of SM. Tumor-canal volume ratio (TCR), the areas related to the cord and tumor occupancy at maximum compression, the presence of dural tail, calcifications, signs of myelopathy, and postoperative cord expansion were compared with the modified McCormick scale (mMCS) preoperative and at follow-up. Results: In the cohort (n = 90 patients), cord and tumor occupancy as well as cord compression and tumor volume showed a correlation with preoperative mMCS (p < 0.05, R −0.23; p < 0.001, R 0.35; p < 0.005, R −0.29; p < 0.001, R 0.42). Cord occupancy had a strong correlation with cord compression (p < 0.001, R 0.72). Tumor occupancy and TCR were correlated with relative outcome at follow-up (p < 0.005 R 0.3; p < 0.005 R 0.29). No correlation was found between cord re-expansion and clinical outcome at follow-up. Finally, a correlation was shown between preoperative signs of cord myelopathy and mMCS (p < 0.05 R 0.21) at follow-up. Conclusions: Larger tumors showed lower preoperative functional status and a worse clinical outcome. Moreover, preoperative T2 cord signal changes are correlated with a poorer outcome.

Glioblastoma Surgery Imaging-Reporting and Data System: Standardized Reporting of Tumor Volume, Location, and Resectability Based on Automated Segmentations

Treatment decisions for patients with presumed glioblastoma are based on tumor characteristics available from a preoperative MR scan. Tumor characteristics, including volume, location, and resectability, are often estimated or manually delineated. This process is time consuming and subjective. Hence, comparison across cohorts, trials, or registries are subject to assessment bias. In this study, we propose a standardized Glioblastoma Surgery Imaging Reporting and Data System (GSI-RADS) based on an automated method of tumor segmentation that provides standard reports on tumor features that are potentially relevant for glioblastoma surgery. As clinical validation, we determine the agreement in extracted tumor features between the automated method and the current standard of manual segmentations from routine clinical MR scans before treatment. In an observational consecutive cohort of 1596 adult patients with a first time surgery of a glioblastoma from 13 institutions, we segmented gadolinium-enhanced tumor parts both by a human rater and by an automated algorithm. Tumor features were extracted from segmentations of both methods and compared to assess differences, concordance, and equivalence. The laterality, contralateral infiltration, and the laterality indices were in excellent agreement. The native and normalized tumor volumes had excellent agreement, consistency, and equivalence. Multifocality, but not the number of foci, had good agreement and equivalence. The location profiles of cortical and subcortical structures were in excellent agreement. The expected residual tumor volumes and resectability indices had excellent agreement, consistency, and equivalence. Tumor probability maps were in good agreement. In conclusion, automated segmentations are in excellent agreement with manual segmentations and practically equivalent regarding tumor features that are potentially relevant for neurosurgical purposes. Standard GSI-RADS reports can be generated by open access software.

Fast meningioma segmentation in T1-weighted magnetic resonance imaging volumes using a lightweight 3D deep learning architecture

Purpose: Automatic and consistent meningioma segmentation in T1-weighted magnetic resonance (MR) imaging volumes and corresponding volumetric assessment is of use for diagnosis, treatment planning, and tumor growth evaluation. We optimized the segmentation and processing speed performances using a large number of both surgically treated meningiomas and untreated meningiomas followed at the outpatient clinic. Approach: We studied two different three-dimensional (3D) neural network architectures: (i) a simple encoder-decoder similar to a 3D U-Net, and (ii) a lightweight multi-scale architecture [Pulmonary Lobe Segmentation Network (PLS-Net)]. In addition, we studied the impact of different training schemes. For the validation studies, we used 698 T1-weighted MR volumes from St. Olav University Hospital, Trondheim, Norway. The models were evaluated in terms of detection accuracy, segmentation accuracy, and training/inference speed. Results: While both architectures reached a similar Dice score of 70% on average, the PLS-Net was more accurate with an F 1 -score of up to 88%. The highest accuracy was achieved for the largest meningiomas. Speed-wise, the PLS-Net architecture tended to converge in about 50 h while 130 h were necessary for U-Net. Inference with PLS-Net takes less than a second on GPU and about 15 s on CPU. Conclusions: Overall, with the use of mixed precision training, it was possible to train competitive segmentation models in a relatively short amount of time using the lightweight PLS-Net architecture. In the future, the focus should be brought toward the segmentation of small meningiomas ( < 2    ml ) to improve clinical relevance for automatic and early diagnosis and speed of growth estimates.

Extent of radiological response does not reflect survival in primary central nervous system lymphoma

Background

In primary central nervous system lymphoma (PCNSL), small enhancing lesions can persist after treatment. It is unknown whether a difference in response category (complete response [CR], complete response unconfirmed [CRu], or partial response [PR]) reflects survival. We aimed to determine the value of a central radiology review on response assessment and whether the extent of response influenced progression-free and/or overall survival.

Methods

All patients in the HOVON 105/ALLG NHL 24 study with at least a baseline MRI and one MRI made for response evaluation available for central review were included. Tumor measurements were done by 2 independent central reviewers, disagreements were adjudicated by a third reviewer. Crude agreement and interobserver agreement (Cohen's kappa) were calculated. Differences in progression-free and overall survival between different categories of response at the end-of-protocol-treatment were assessed by the log-rank test in a landmark survival-analysis.

Results

Agreement between the central reviewers was 61.7% and between local and central response assessment was 63.0%. Cohen's kappa's, which corrects for expected agreement, were 0.44 and 0.46 (moderate), respectively. Progression agreement or not was 93.3% (kappa 0.87) between local and central response assessment. There were no significant differences in progression-free and overall survival between patients with CR, CRu, or PR at the end-of-protocol-treatment, according to both local and central response assessment.

Conclusions

Reliability of response assessment (CR/CRu/PR) is moderate even by central radiology review and these response categories do not reliably predict survival. Therefore, primary outcome in PCNSL studies should be survival rather than CR or CR/CRu-rate.

Advancements in predicting outcomes in patients with glioma: a surgical perspective

Introduction: Diffuse glioma is a challenging neurosurgical entity. Although surgery does not provide a cure, it may greatly influence survival, brain function, and quality of life. Surgical treatment is by nature highly personalized and outcome prediction is very complex. To engage and succeed in this balancing act it is important to make best use of the information available to the neurosurgeon.Areas covered: This narrative review provides an update on advancements in predicting outcomes in patients with glioma that are relevant to neurosurgeons.Expert opinion: The classical 'gut feeling' is notoriously unreliable and better prediction strategies for patients with glioma are warranted. There are numerous tools readily available for the neurosurgeon in predicting tumor biology and survival. Predicting extent of resection, functional outcome, and quality of life remains difficult. Although machine-learning approaches are currently not readily available in daily clinical practice, there are several ongoing efforts with the use of big data sets that are likely to create new prediction models and refine the existing models.