Conventional magnetic resonance imaging in the differentiation between high and low-grade brain tumours in paediatric patients

Deciphering Machine Learning Decisions to Distinguish between Posterior Fossa Tumor Types Using MRI Features: What Do the Data Tell Us?

Machine learning (ML) models have become capable of making critical decisions on our behalf. Nevertheless, due to complexity of these models, interpreting their decisions can be challenging, and humans cannot always control them. This paper provides explanations of decisions made by ML models in diagnosing four types of posterior fossa tumors: medulloblastoma, ependymoma, pilocytic astrocytoma, and brainstem glioma. The proposed methodology involves data analysis using kernel density estimations with Gaussian distributions to examine individual MRI features, conducting an analysis on the relationships between these features, and performing a comprehensive analysis of ML model behavior. This approach offers a simple yet informative and reliable means of identifying and validating distinguishable MRI features for the diagnosis of pediatric brain tumors. By presenting a comprehensive analysis of the responses of the four pediatric tumor types to each other and to ML models in a single source, this study aims to bridge the knowledge gap in the existing literature concerning the relationship between ML and medical outcomes. The results highlight that employing a simplistic approach in the absence of very large datasets leads to significantly more pronounced and explainable outcomes, as expected. Additionally, the study also demonstrates that the pre-analysis results consistently align with the outputs of the ML models and the clinical findings reported in the existing literature.

5-ALA fluorescence in randomly selected pediatric brain tumors assessed by spectroscopy and surgical microscope

PURPOSE

Fluorescence-guided surgery applying 5-aminolevulinic acid (5-ALA) in high-grade gliomas is an established method in adults. In children, results have so far been ambiguous. The aim of this study was to investigate 5-ALA-induced fluorescence in pediatric brain tumors by using the surgical microscope and a spectroscopic hand-held probe.

METHODS

Fourteen randomly selected children (age 4-17) with newly MRI-verified brain tumors were included. No selection was based on the suspected diagnosis prior to surgery. All patients received 5-ALA (20 mg /kg) either orally or via a gastric tube prior to surgery. Intratumoral fluorescence was detected with the microscope and the probe. Moreover, fluorescence in the skin of the forearm was measured. Histopathology samples revealed seven low-grade gliomas, four medulloblastomas, one diffuse intrinsic pontine glioma, one glioblastoma and one atypical meningioma. Blood samples were analyzed, and potential clinical side effects were monitored.

RESULTS

Microscopically, vague fluorescence was visible in two patients. Intratumoral fluorescence could be detected in five patients with the probe, including the two patients with vague microscopic fluorescence. Three of the oldest children had PpIX fluorescence in the skin. Nine children did not show any fluorescence in the tumor or in the skin. No clinical side effects or laboratory adverse events were observed.

CONCLUSION

Fluorescence could not be used to guide surgery in this study, neither with the surgical microscope nor with the hand-held probe. In nine children, no fluorescence was discerned and children with noticeable fluorescence were all older than nine years. 5-ALA was considered safe to apply in children.

Added value of magnetic resonance spectroscopy for diagnosing childhood cerebellar tumours

¹ H-magnetic resonance spectroscopy (MRS) provides noninvasive metabolite profiles with the potential to aid the diagnosis of brain tumours. Prospective studies of diagnostic accuracy and comparisons with conventional MRI are lacking. The aim of the current study was to evaluate, prospectively, the diagnostic accuracy of a previously established classifier for diagnosing the three major childhood cerebellar tumours, and to determine added value compared with standard reporting of conventional imaging. Single-voxel MRS (1.5 T, PRESS, TE 30 ms, TR 1500 ms, spectral resolution 1 Hz/point) was acquired prospectively on 39 consecutive cerebellar tumours with histopathological diagnoses of pilocytic astrocytoma, ependymoma or medulloblastoma. Spectra were analysed with LCModel and predefined quality control criteria were applied, leaving 33 cases in the analysis. The MRS diagnostic classifier was applied to this dataset. A retrospective analysis was subsequently undertaken by three radiologists, blind to histopathological diagnosis, to determine the change in diagnostic certainty when sequentially viewing conventional imaging, MRS and a decision support tool, based on the classifier. The overall classifier accuracy, evaluated prospectively, was 91%. Incorrectly classified cases, two anaplastic ependymomas, and a rare histological variant of medulloblastoma, were not well represented in the original training set. On retrospective review of conventional MRI, MRS and the classifier result, all radiologists showed a significant increase (Wilcoxon signed rank test, p < 0.001) in their certainty of the correct diagnosis, between viewing the conventional imaging and MRS with the decision support system. It was concluded that MRS can aid the noninvasive diagnosis of posterior fossa tumours in children, and that a decision support classifier helps in MRS interpretation.

Differentiation of paediatric posterior fossa tumours by the multiregional and multiparametric MRI radiomics approach: a study on the selection of optimal multiple sequences and multiregions

OBJECTIVE

To evaluate the diagnostic performance of a radiomics model based on multiregional and multiparametric MRI to classify paediatric posterior fossa tumours (PPFTs), explore the contribution of different MR sequences and tumour subregions in tumour classification, and examine whether contrast-enhanced T₁ weighted (T1C) images have irreplaceable added value.

METHODS

This retrospective study of 136 PPFTs extracted 11,958 multiregional (enhanced, non-enhanced, and total tumour) features from multiparametric MRI (T₁- and T₂ weighted, T1C, fluid-attenuated inversion recovery, and diffusion-weighted images). These features were subjected to fast correlation-based feature selection and classified by a support vector machine based on different tasks. Diagnostic performances of multiregional and multiparametric MRI features, different sequences, and different tumoral regions were evaluated using multiclass and one-vs-rest strategies.

RESULTS

The established model achieved an overall area under the curve (AUC) of 0.977 in the validation cohort. The performance of PPFTs significantly improved after replacing T1C with apparent diffusion coefficient maps added into the plain scan sequences (AUC from 0.812 to 0.917). When oedema features were added to contrast-enhancing tumour volume, the performance did not significantly improve.

CONCLUSION

The radiomics model built by multiregional and multiparametric MRI features allows for the excellent distinction of different PPFTs and provides valuable references for the rational adoption of MR sequences.

ADVANCES IN KNOWLEDGE

This study emphasized that T1C has limited added value in predicting PPFTs and should be cautiously adopted. Selecting optimal MR sequences may help guide clinicians to better allocate acquisition sequences and reduce medical costs.

Perfusion-weighted techniques in MRI grading of pediatric cerebral tumors: efficiency of dynamic susceptibility contrast and arterial spin labeling

PURPOSE

Dynamic susceptibility contrast (DSC) and arterial spin labeling (ASL) perfusion MRI are applied in pediatric brain tumor grading, but their value for clinical daily practice remains unclear. We explored the ability of ASL and DSC to distinguish low- and high-grade lesions, in an unselected cohort of pediatric cerebral tumors.

METHODS

We retrospectively compared standard perfusion outcomes including blood volume, blood flow, and time parameters from DSC and ASL at 1.5T or 3T MRI scanners of 46 treatment-naive patients by drawing ROI via consensus by two neuroradiologists on the solid portions of every tumor. The discriminant abilities of perfusion parameters were evaluated by receiver operating characteristic (ROC) over the entire cohort and depending on the tumor location and the magnetic field.

RESULTS

ASL and DSC parameters showed overall low to moderate performances to distinguish low- and high-grade tumors (area under the curve: between 0.548 and 0.697). Discriminant abilities were better for tumors located supratentorially (AUC between 0.777 and 0.810) than infratentorially, where none of the metrics reached significance. We observed a better differentiation between low- and high-grade cancers at 3T than at 1.5-T. For infratentorial tumors, time parameters from DSC performed better than the commonly used metrics (AUC ≥ 0.8).

CONCLUSION

DSC and ASL show moderate abilities to distinguish low- and high-grade brain tumors in an unselected cohort. Absolute value of K2, TMAX, tMIP, and normalized value of TMAX of the DSC appear as an alternative to conventional parameters for infratentorial tumors. Three Tesla evaluation should be favored over 1.5-Tesla.

The effect of semi-quantitative T1-perfusion parameters for the differentiation between pediatric medulloblastoma and ependymoma

Background

The differentiation between medulloblastomas and ependymomas plays an important role in treatment planning and prognosis for children. This study aims to investigate the role of T1-perfusion parameters during the differentiation between medulloblastomas and ependymomas in children. The institutional review board approved this prospective study. The brain magnetic resonance imaging (MRI) protocol, including axial T1-perfusion, was assessed in 26 patients, divided into a medulloblastoma group (group 1, n = 22) and an ependymoma group (group 2, n = 4). The quantified region of interest (ROI) values for tumors and the tumor to parenchyma ratios were collected and compared between the two groups. Receiver operating characteristic (ROC) curve analysis and the Youden index were utilized to identify the best cut-off, sensitivity, specificity, and area under the curve (AUC) values for the independent T1-perfusion parameters.

Results

The relative enhancement, maximum enhancement, maximum relative enhancement, time to peak, and AUC values for medulloblastomas were significantly higher than those for ependymomas (p < 0.05). Furthermore, the maximum enhancement and maximum relative enhancement for medulloblastoma to parenchyma ratios were also significantly higher than those for ependymomas. A cut-off maximum enhancement value of 100.25 was identified as sufficient to discriminate between medulloblastoma and ependymoma and resulted in a sensitivity of 90.9%, a specificity of 100%, and an AUC of 94.3%.

Conclusion

A cut-off maximum enhancement value of 100.25 derived from T1-perfusion was able to discriminate between medulloblastoma and ependymoma, with high sensitivity, specificity, and accuracy values.

Factors associated with seizure and cognitive outcomes after epilepsy surgery for low-grade epilepsy-associated neuroepithelial tumors in children

Low-grade epilepsy-associated neuroepithelial tumors (LEATs) are responsible for drug-resistant chronic focal epilepsy, and are the second-most common reason for epilepsy surgery in children. LEATs are extremely responsive to surgical treatment, and therefore epilepsy surgery should be considered as a treatment option for LEATs. However, the optimal time for surgery remains controversial, and surgeries are often delayed. In this review, we reviewed published article on the factors associated with seizure and cognitive outcomes after epilepsy surgery for LEATs in children to help clinicians in their decision whether to pursue epilepsy surgery for LEATs. The achievement of gross total resection may be the most important prognostic factor for seizure freedom. A shorter duration of epilepsy, a younger age at surgery, and extended resection of temporal lobe tumors have also been suggested as favorable prognostic factors in terms of seizure control. Poor cognitive function in children with LEATs is associated with a longer duration of epilepsy and a younger age at seizure onset.

Magnetic resonance imaging based functional imaging in paediatric oncology

Imaging is central to management of solid tumours in children. Conventional magnetic resonance imaging (MRI) is the standard imaging modality for tumours of the central nervous system (CNS) and limbs and is increasingly used in the abdomen. It provides excellent structural detail, but imparts limited information about tumour type, aggressiveness, metastatic potential or early treatment response. MRI based functional imaging techniques, such as magnetic resonance spectroscopy, diffusion and perfusion weighted imaging, probe tissue properties to provide clinically important information about metabolites, structure and blood flow. This review describes the role of and evidence behind these functional imaging techniques in paediatric oncology and implications for integrating them into routine clinical practice.

The diagnostic accuracy of multiparametric MRI to determine pediatric brain tumor grades and types

Childhood brain tumors show great histological variability. The goal of this retrospective study was to assess the diagnostic accuracy of multimodal MR imaging (diffusion, perfusion, MR spectroscopy) in the distinction of pediatric brain tumor grades and types. Seventy-six patients (range 1 month to 18 years) with brain tumors underwent multimodal MR imaging. Tumors were categorized by grade (I-IV) and by histological type (A-H). Multivariate statistical analysis was performed to evaluate the diagnostic accuracy of single and combined MR modalities, and of single imaging parameters to distinguish the different groups. The highest diagnostic accuracy for tumor grading was obtained with diffusion-perfusion (73.24%) and for tumor typing with diffusion-perfusion-MR spectroscopy (55.76%). The best diagnostic accuracy was obtained for tumor grading in I and IV and for tumor typing in embryonal tumor and pilocytic astrocytoma. Poor accuracy was seen in other grades and types. ADC and rADC were the best parameters for tumor grading and typing followed by choline level with an intermediate echo time, CBV for grading and Tmax for typing. Multiparametric MR imaging can be accurate in determining tumor grades (primarily grades I and IV) and types (mainly pilocytic astrocytomas and embryonal tumors) in children.

Analysis of T2 signal intensity helps in the differentiation between high and low-grade brain tumours in paediatric patients

PURPOSE

Previous studies hypothesized that the analysis of magnetic resonance intensity of the solid portion in paediatric tumours can provide pre-surgical information about the histopathology. Classically, high signal-intensity in T2weighted (T2w) images identifies low-grade tumours, while anaplasia is characterized by T2 hypointensity. We aimed to investigate if T2w signal intensities can pre-operatively distinguish between low-grade and high-grade brain tumours in paediatric patients.

METHODS

Two raters, blinded to the histological diagnosis, rated the signal intensity of MR images (T2w) from 36 children with newly diagnosed brain tumours, 17 children with low-grade brain tumours and 19 children with high-grade brain tumours were included in this study. Relative T2 values were obtained by dividing the T2w values of the solid portion of the tumour by the T2w values of the vitreous humour.

RESULTS

The best cut-off point to distinguish low and high-grade paediatric brain tumours was 0.8. If the signal intensity was less than or equal to 0.8 the tumour was expected to be a high-grade tumour with a sensitivity of 100%. Prediction of a low-grade tumour was more uncertain with a sensitivity of 70.5%. Overall, 86% of the tumours would have been predicted correctly.

CONCLUSION

Our data suggest that T2w signal intensities of the solid portion of brain tumours in paediatrics can pre-operatively differentiate between low-grade and high-grade tumours. In addition, T2 hypointensity may be helpful in targeting stereotactic biopsy.

Differentiating low- and high-grade pediatric brain tumors using a continuous-time random-walk diffusion model at high b-values

PURPOSE

To demonstrate that a continuous-time random-walk (CTRW) diffusion model can improve diagnostic accuracy of differentiating low- and high-grade pediatric brain tumors.

METHODS

Fifty-four children with histopathologically confirmed brain tumors underwent diffusion MRI scans at 3Twith 12 b-values (0-4000 s/mm(2) ). The diffusion imageswere fit to a simplified CTRW model to extract anomalous diffusion coefficient, Dm , and temporal and spatial heterogeneity parameters, α and β, respectively. Using histopathology results as reference, a k-means clustering algorithm and a receiver operating characteristic (ROC) analysis were employed to determine the sensitivity, specificity, and diagnostic accuracy of the CTRW parameters in differentiating tumor grades.

RESULTS

Significant differences between the low- and high-grade tumors were observed in the CTRW parameters (p-values<0.001). The k-means analysis showed that the combination of three CTRW parameters produced higher diagnostic accuracy (85% vs. 75%) and specificity (83% vs. 54%) than the apparent diffusion coefficient (ADC) from a mono-exponential model. The ROC analysis revealed that any combination of the CTRW parameters gave a larger area under the curve (0.90-0.96) than using ADC (0.80).

CONCLUSION

With its sensitivity to intravoxel heterogeneity, the simplified CTRW model is useful for non-invasive grading of pediatric brain tumors, particularly when surgical biopsy is not feasible. Magn Reson Med 76:1149-1157, 2016. © 2015 Wiley Periodicals, Inc.