Classification of paediatric brain tumours by diffusion weighted imaging and machine learning

Texture analysis combined with machine learning in radiographs of the knee joint: potential to identify tibial plateau occult fractures

Background

Missed or delayed diagnosis of occult fractures of tibial plateau may cause adverse effects on patients. The objective of this study was to evaluate the diagnostic performance of texture analysis (TA) of knee joint radiographs combined with machine learning (ML) in identifying patients at risk of tibial plateau occult fractures.

Methods

A total of 169 patients with negative fracture on knee X-ray films from 2018 to 2022 who were diagnosed with occult tibial plateau fractures or no fractures by subsequent magnetic resonance imaging (MRI) examination were retrospectively enrolled. The X-ray images of the patient's knee joint were used for texture feature extraction. A total of 9 ML feature selection methods (including 6 mainstream methods and 3 methods provided by MaZda software) combined with 3 classification methods were used to build the best diagnostic model. The performance of each model was evaluated by accuracy, F1-value, and area under the curve (AUC).

Results

The least absolute shrinkage and selection operator (LASSO) method had the best performance of the 6 mainstream methods, with an accuracy of 0.81, an F1 value of 0.80, and an AUC of 0.920, all of which were higher than those of the other five methods (accuracy range: 0.65-0.80, F1 score range: 0.61-0.79, AUC range: 0.722-0.895). Among the three feature selection models in MaZda software, the most ideal method for accuracy measurement was the MI method, reaching 0.77. In the measurement of the F1 value and AUC, MaZda's best method was Fisher, reaching 0.78 and 0.888, respectively. All indicators were lower than those of the LASSO method. The combination of LASSO and support vector machine (SVM) yielded the best classification performance, while the performance of the combination of LASSO and logistic regression was slightly inferior, but the difference was not statistically significant.

Conclusions

TA of knee joint radiography combined with ML has achieved high performance in identifying patients at risk of occult fractures of the tibial plateau. Considering both the model performance and computational complexity, the LASSO feature selection method combined with the logistic regression classifier yielded the best classification performance in this process.

Advanced imaging techniques and non-invasive biomarkers in pediatric brain tumors: state of the art

In the pediatric age group, brain neoplasms are the second most common tumor category after leukemia, with an annual incidence of 6.13 per 100,000. Conventional MRI sequences, complemented by CT whenever necessary, are fundamental for the initial diagnosis and surgical planning as well as for post-operative evaluations, assessment of response to treatment, and surveillance; however, they have limitations, especially concerning histopathologic or biomolecular phenotyping and grading. In recent years, several advanced MRI sequences, including diffusion-weighted imaging, diffusion tensor imaging, arterial spin labelling (ASL) perfusion, and MR spectroscopy, have emerged as a powerful aid to diagnosis as well as prognostication; furthermore, other techniques such as diffusion kurtosis, amide proton transfer imaging, and MR elastography are being translated from the research environment to clinical practice. Molecular imaging, especially PET with amino-acid tracers, complement MRI in several aspects, including biopsy targeting and outcome prediction. Finally, radiomics with radiogenomics are opening entirely new perspectives for a quantitative approach aiming at identifying biomarkers that can be used for personalized, precision management strategies.

Enhancing neuro-oncology care through equity-driven applications of artificial intelligence

The disease course and clinical outcome for brain tumor patients depend not only on the molecular and histological features of the tumor but also on the patient's demographics and social determinants of health. While current investigations in neuro-oncology have broadly utilized artificial intelligence (AI) to enrich tumor diagnosis and more accurately predict treatment response, postoperative complications, and survival, equity-driven applications of AI have been limited. However, AI applications to advance health equity in the broader medical field have the potential to serve as practical blueprints to address known disparities in neuro-oncologic care. In this consensus review, we will describe current applications of AI in neuro-oncology, postulate viable AI solutions for the most pressing inequities in neuro-oncology based on broader literature, propose a framework for the effective integration of equity into AI-based neuro-oncology research, and close with the limitations of AI.

Diagnostic Utility of Diffusion-Weighted Imaging in Distinguishing Common Pediatric Posterior Fossa Tumors: A Single Center Retrospective Study

Objective

Pediatric posterior fossa tumors pose diagnostic challenges due to their diverse histopathological features and variable clinical presentations. Conventional magnetic resonance imaging (MRI) serves as the initial diagnostic tool; however, additional modalities, such as diffusion-weighted imaging (DWI), are essential for refining tumor classification. This retrospective single-center study aimed to evaluate the diagnostic utility of apparent diffusion coefficient (ADC) parameters in distinguishing between the most common pediatric posterior fossa tumors.

Materials and Methods

Fifty-nine patients under the age of 18 (27 females and 32 males) with histopathologically diagnosed primary posterior fossa tumors underwent pre-treatment conventional and diffusion MRI. Apparent diffusion coefficient values were measured from solid tumor regions and normal cerebellar parenchyma, with subsequent calculation of tumor/normal cerebellar ADC ratios.

Results

The median ADC values for pilocytic astrocytomas (PAs) were 1786.2 × 10-6 mm2 /s, ependymomas 1144.9 × 10-6 mm2 /s, and for medulloblastomas 666.1 × 10-6 mm2 /s were significantly different (P < .001 for all three). Similarly, the median ADC ratios demonstrated discriminatory potential, with PAs showing the highest ratio (2.46), followed by ependymomas (1.55) and medulloblastomas (0.89) (P < .001 for all three). Receiver operating characteristic analysis revealed distinct ADC cutoffs and ratios for differentiating all tumor types from each other.

Conclusion

Despite limitations, such as a small cohort size and different MRI protocols, our results show that ADC metrics are especially useful for distinguishing between the most common pediatric posterior fossa tumors. We recommend that future studies integrate advanced imaging techniques and larger cohorts to improve diagnostic accuracy.

Progress Achieved, Landmarks, and Future Concerns in Biomedical and Health Informatics

BACKGROUND

The biomedical and health informatics (BMHI) fields have been advancing rapidly, a trend particularly emphasised during the recent COVID-19 pandemic, introducing innovations in BMHI. Over nearly 50 years since its establishment as a scientific discipline, BMHI has encountered several challenges, such as mishaps, delays, failures, and moments of enthusiastic expectations and notable successes. This paper focuses on reviewing the progress made in the BMHI discipline, evaluating key milestones, and discussing future challenges.

METHODS

To, Structured, step-by-step qualitative methodology was developed and applied, centred on gathering expert opinions and analysing trends from the literature to provide a comprehensive assessment. Experts and pioneers in the BMHI field were assigned thematic tasks based on the research question, providing critical inputs for the thematic analysis. This led to the identification of five key dimensions used to present the findings in the paper: informatics in biomedicine and healthcare, health data in Informatics, nurses in informatics, education and accreditation in health informatics, and ethical, legal, social, and security issues.

RESULTS

Each dimension is examined through recently emerging innovations, linking them directly to the future of healthcare, like the role of artificial intelligence, innovative digital health tools, the expansion of telemedicine, and the use of mobile health apps and wearable devices. The new approach of BMHI covers newly introduced clinical needs and approaches like patient-centric, remote monitoring, and precision medicine clinical approaches.

CONCLUSIONS

These insights offer clear recommendations for improving education and developing experts to advance future innovations. Notably, this narrative review presents a body of knowledge essential for a deep understanding of the BMHI field from a human-centric perspective and, as such, could serve as a reference point for prospective analysis and innovation development.

Radiomics and artificial intelligence applications in pediatric brain tumors

BACKGROUND

The study of central nervous system (CNS) tumors is particularly relevant in the pediatric population because of their relatively high frequency in this demographic and the significant impact on disease- and treatment-related morbidity and mortality. While both morphological and non-morphological magnetic resonance imaging techniques can give important information concerning tumor characterization, grading, and patient prognosis, increasing evidence in recent years has highlighted the need for personalized treatment and the development of quantitative imaging parameters that can predict the nature of the lesion and its possible evolution. For this purpose, radiomics and the use of artificial intelligence software, aimed at obtaining valuable data from images beyond mere visual observation, are gaining increasing importance. This brief review illustrates the current state of the art of this new imaging approach and its contributions to understanding CNS tumors in children.

DATA SOURCES

We searched the PubMed, Scopus, and Web of Science databases using the following key search terms: ("radiomics" AND/OR "artificial intelligence") AND ("pediatric AND brain tumors"). Basic and clinical research literature related to the above key research terms, i.e., studies assessing the key factors, challenges, or problems of using radiomics and artificial intelligence in pediatric brain tumors management, was collected.

RESULTS

A total of 63 articles were included. The included ones were published between 2008 and 2024. Central nervous tumors are crucial in pediatrics due to their high frequency and impact on disease and treatment. MRI serves as the cornerstone of neuroimaging, providing cellular, vascular, and functional information in addition to morphological features for brain malignancies. Radiomics can provide a quantitative approach to medical imaging analysis, aimed at increasing the information obtainable from the pixels/voxel grey-level values and their interrelationships. The "radiomic workflow" involves a series of iterative steps for reproducible and consistent extraction of imaging data. These steps include image acquisition for tumor segmentation, feature extraction, and feature selection. Finally, the selected features, via training predictive model (CNN), are used to test the final model.

CONCLUSIONS

In the field of personalized medicine, the application of radiomics and artificial intelligence (AI) algorithms brings up new and significant possibilities. Neuroimaging yields enormous amounts of data that are significantly more than what can be gained from visual studies that radiologists can undertake on their own. Thus, new partnerships with other specialized experts, such as big data analysts and AI specialists, are desperately needed. We believe that radiomics and AI algorithms have the potential to move beyond their restricted use in research to clinical applications in the diagnosis, treatment, and follow-up of pediatric patients with brain tumors, despite the limitations set out.

Machine Learning Analysis in Diffusion Kurtosis Imaging for Discriminating Pediatric Posterior Fossa Tumors: A Repeatability and Accuracy Pilot Study

Background and purpose: Differentiating pediatric posterior fossa (PF) tumors such as medulloblastoma (MB), ependymoma (EP), and pilocytic astrocytoma (PA) remains relevant, because of important treatment and prognostic implications. Diffusion kurtosis imaging (DKI) has not yet been investigated for discrimination of pediatric PF tumors. Estimating diffusion values from whole-tumor-based (VOI) segmentations may improve diffusion measurement repeatability compared to conventional region-of-interest (ROI) approaches. Our purpose was to compare repeatability between ROI and VOI DKI-derived diffusion measurements and assess DKI accuracy in discriminating among pediatric PF tumors. Materials and methods: We retrospectively analyzed 34 children (M, F, mean age 7.48 years) with PF tumors who underwent preoperative examination on a 3 Tesla magnet, including DKI. For each patient, two neuroradiologists independently segmented the whole solid tumor, the ROI of the area of maximum tumor diameter, and a small 5 mm ROI. The automated analysis pipeline included inter-observer variability, statistical, and machine learning (ML) analyses. We evaluated inter-observer variability with coefficient of variation (COV) and Bland-Altman plots. We estimated DKI metrics accuracy in discriminating among tumor histology with MANOVA analysis. In order to account for class imbalances, we applied SMOTE to balance the dataset. Finally, we performed a Random Forest (RF) machine learning classification analysis based on all DKI metrics from the SMOTE dataset by partitioning 70/30 the training and testing cohort. Results: Tumor histology included medulloblastoma (15), pilocytic astrocytoma (14), and ependymoma (5). VOI-based measurements presented lower variability than ROI-based measurements across all DKI metrics and were used for the analysis. DKI-derived metrics could accurately discriminate between tumor subtypes (Pillai's trace: p < 0.001). SMOTE generated 11 synthetic observations (10 EP and 1 PA), resulting in a balanced dataset with 45 instances (34 original and 11 synthetic). ML analysis yielded an accuracy of 0.928, which correctly predicted all but one lesion in the testing set. Conclusions: VOI-based measurements presented improved repeatability compared to ROI-based measurements across all diffusion metrics. An ML classification algorithm resulted accurate in discriminating PF tumors on a SMOTE-generated dataset. ML techniques based on DKI-derived metrics are useful for the discrimination of pediatric PF tumors.

Exploring the Potential of Machine Learning Algorithms to Improve Diffusion Nuclear Magnetic Resonance Imaging Models Analysis

Purpose

This paper explores different machine learning (ML) algorithms for analyzing diffusion nuclear magnetic resonance imaging (dMRI) models when analytical fitting shows restrictions. It reviews various ML techniques for dMRI analysis and evaluates their performance on different b-values range datasets, comparing them with analytical methods.

Materials and Methods

After standard fitting for reference, four sets of diffusion-weighted nuclear magnetic resonance images were used to train/test various ML algorithms for prediction of diffusion coefficient (D), pseudo-diffusion coefficient (D*), perfusion fraction (f), and kurtosis (K). ML classification algorithms, including extra-tree classifier (ETC), logistic regression, C-support vector, extra-gradient boost, and multilayer perceptron (MLP), were used to determine the existence of diffusion parameters (D, D*, f, and K) within single voxels. Regression algorithms, including linear regression, polynomial regression, ridge, lasso, random forest (RF), elastic-net, and support-vector machines, were used to estimate the value of the diffusion parameters. Performance was evaluated using accuracy (ACC), area under the curve (AUC) tests, and cross-validation root mean square error (RMSECV). Computational timing was also assessed.

Results

ETC and MLP were the best classifiers, with 94.1% and 91.7%, respectively, for the ACC test and 98.7% and 96.3% for the AUC test. For parameter estimation, RF algorithm yielded the most accurate results The RMSECV percentages were: 8.39% for D, 3.57% for D*, 4.52% for f, and 3.53% for K. After the training phase, the ML methods demonstrated a substantial decrease in computational time, being approximately 232 times faster than the conventional methods.

Conclusions

The findings suggest that ML algorithms can enhance the efficiency of dMRI model analysis and offer new perspectives on the microstructural and functional organization of biological tissues. This paper also discusses the limitations and future directions of ML-based dMRI analysis.

Metabolite profiles of medulloblastoma for rapid and non-invasive detection of molecular disease groups

BACKGROUND

The malignant childhood brain tumour, medulloblastoma, is classified clinically into molecular groups which guide therapy. DNA-methylation profiling is the current classification 'gold-standard', typically delivered 3-4 weeks post-surgery. Pre-surgery non-invasive diagnostics thus offer significant potential to improve early diagnosis and clinical management. Here, we determine tumour metabolite profiles of the four medulloblastoma groups, assess their diagnostic utility using tumour tissue and potential for non-invasive diagnosis using in vivo magnetic resonance spectroscopy (MRS).

METHODS

Metabolite profiles were acquired by high-resolution magic-angle spinning NMR spectroscopy (MAS) from 86 medulloblastomas (from 59 male and 27 female patients), previously classified by DNA-methylation array (WNT (n = 9), SHH (n = 22), Group3 (n = 21), Group4 (n = 34)); RNA-seq data was available for sixty. Unsupervised class-discovery was performed and a support vector machine (SVM) constructed to assess diagnostic performance. The SVM classifier was adapted to use only metabolites (n = 10) routinely quantified from in vivo MRS data, and re-tested. Glutamate was assessed as a predictor of overall survival.

FINDINGS

Group-specific metabolite profiles were identified; tumours clustered with good concordance to their reference molecular group (93%). GABA was only detected in WNT, taurine was low in SHH and lipids were high in Group3. The tissue-based metabolite SVM classifier had a cross-validated accuracy of 89% (100% for WNT) and, adapted to use metabolites routinely quantified in vivo, gave a combined classification accuracy of 90% for SHH, Group3 and Group4. Glutamate predicted survival after incorporating known risk-factors (HR = 3.39, 95% CI 1.4-8.1, p = 0.025).

INTERPRETATION

Tissue metabolite profiles characterise medulloblastoma molecular groups. Their combination with machine learning can aid rapid diagnosis from tissue and potentially in vivo. Specific metabolites provide important information; GABA identifying WNT and glutamate conferring poor prognosis.

FUNDING

Children with Cancer UK, Cancer Research UK, Children's Cancer North and a Newcastle University PhD studentship.

Evolutionary gravitational neocognitron neural network optimized with marine predators optimization algorithm for MRI brain tumor classification

Magnetic resonance imaging (MRI) is a powerful tool for tumor diagnosis in human brain. Here, the MRI images are considered to detect the brain tumor and classify the regions as meningioma, glioma, pituitary and normal types. Numerous existing methods regarding brain tumor detection were suggested previously, but none of the methods accurately categorizes the brain tumor and consumes more computation period. To address these problems, an Evolutionary Gravitational Neocognitron Neural Network optimized with Marine Predators Algorithm is proposed in this article for MRI Brain Tumor Classification (EGNNN-VGG16-MPA-MRI-BTC). Initially, the brain MRI pictures are collected under Brats MRI image dataset. By using Savitzky-Golay Denoising approach, these images are pre-processed. The features are extracted utilizing visual geometry group network (VGG16). By utilizing VGG16, the features, like Grey level features, Haralick Texture features are extracted. These extracted features are given to EGNNN classifier, which categorizes the brain tumor as glioma, meningioma, pituitary gland and normal. Batch Normalization (BN) layer of EGNNN is eliminated and included with VGG16 layer. Marine Predators Optimization Algorithm (MPA) optimizes the weight parameters of EGNNN. The simulation is activated in MATLAB. Finally, the EGNNN-VGG16-MPA-MRI-BTC method attains 38.98%, 46.74%, 23.27% higher accuracy, 24.24%, 37.82%, 13.92% higher precision, 26.94%, 47.04%, 38.94% higher sensitivity compared with the existing AlexNet-SVM-MRI-BTC, RESNET-SGD-MRI-BTC and MobileNet-V2-MRI-BTC models respectively.

Posterior fossa tumors in children: current insights

While in adults most intracranial tumors develop around the cerebral hemispheres, 45 to 60% of pediatric lesions are found in the posterior fossa, although this anatomical region represents only 10% of the intracranial volume. The latest edition of the WHO classification for CNS tumors presented some fundamental paradigm shifts that particularly affected the classification of pediatric tumors, also influencing those that affect posterior fossa. Molecular biomarkers play an important role in the diagnosis, prognosis, and treatment of childhood posterior fossa tumors and can be used to predict patient outcomes and response to treatment and monitor its effectiveness. Although genetic studies have identified several posterior fossa tumor types, differing in terms of their location, cell of origin, genetic mechanisms, and clinical behavior, recent management strategies still depend on uniform approaches, mainly based on the extent of resection. However, significant progress has been made in guiding therapy decisions with biological or molecular stratification criteria and utilizing molecularly targeted treatments that address specific tumor biological characteristics. The primary focus of this review is on the latest advances in the diagnosis and treatment of common subtypes of posterior fossa tumors in children, as well as potential therapeutic approaches in the future.   Conclusion: Molecular biomarkers play a central role, not only in the diagnosis and prognosis of posterior fossa tumors in children but also in customizing treatment plans. They anticipate patient outcomes, measure treatment responses, and assess therapeutic effectiveness. Advances in neuroimaging and treatment have significantly enhanced outcomes for children with these tumors. What is Known: • Central nervous system tumors are the most common solid neoplasms in children and adolescents, with approximately 45 to 60% of them located in the posterior fossa. • Multimodal approaches that include neurosurgery, radiation therapy, and chemotherapy are typically used to manage childhood posterior fossa tumors What is New: • Notable progress has been achieved in the diagnosis, categorization and management of posterior fossa tumors in children, leading to improvement in survival and quality of life.

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.

[Pediatric posterior fossa tumors]

CLINICAL ISSUE

Tumors of the posterior fossa account for about 50-55% of brain tumors in childhood.

DIAGNOSTIC WORKUP

The most frequent tumor entities are medulloblastomas, pilocytic astrocytomas, ependymomas, diffuse midline gliomas and atypical teratoid-rhabdoid tumors. Neuroradiological differential diagnosis with magnetic resonance imaging (MRI) is of considerable importance for preoperative planning as well as planning of follow-up therapy.

PERFORMANCE

Most important findings for differential diagnosis of pediatric posterior fossa tumors are tumor location, patient age and the intratumoral apparent diffusion assessed by diffusion-weighted imaging.

ACHIEVEMENTS

Advanced MR techniques like MRI perfusion and MR spectroscopy can be helpful both in the initial differential diagnosis and in tumor surveillance, but exceptional characteristics of certain tumor entities should be kept in mind.

PRACTICAL RECOMMENDATIONS

Standard clinical MRI sequences including diffusion-weighted imaging are the main diagnostic tool in evaluating posterior fossa tumors in children. Advanced imaging methods can be helpful, but should never be interpreted separately from conventional MRI sequences.

Application of Apparent Diffusion Coefficient Histogram Metrics for Differentiation of Pediatric Posterior Fossa Tumors : A Large Retrospective Study and Brief Review of Literature

PURPOSE

This study aimed to evaluate the application of apparent diffusion coefficient (ADC) histogram analysis to differentiate posterior fossa tumors (PFTs) in children.

METHODS

A total of 175 pediatric patients with PFT, including 75 pilocytic astrocytomas (PA), 59 medulloblastomas, 16 ependymomas, and 13 atypical teratoid rhabdoid tumors (ATRT), were analyzed. Tumors were visually assessed using DWI trace and conventional MRI images and manually segmented and post-processed using parametric software (pMRI). Furthermore, tumor ADC values were normalized to the thalamus and cerebellar cortex. The following histogram metrics were obtained: entropy, minimum, 10th, and 90th percentiles, maximum, mean, median, skewness, and kurtosis to distinguish the different types of tumors. Kruskal Wallis and Mann-Whitney U tests were used to evaluate the differences. Finally, receiver operating characteristic (ROC) curves were utilized to determine the optimal cut-off values for differentiating the various PFTs.

RESULTS

Most ADC histogram metrics showed significant differences between PFTs (p < 0.001) except for entropy, skewness, and kurtosis. There were significant pairwise differences in ADC metrics for PA versus medulloblastoma, PA versus ependymoma, PA versus ATRT, medulloblastoma versus ependymoma, and ependymoma versus ATRT (all p < 0.05). Our results showed no significant differences between medulloblastoma and ATRT. Normalized ADC data showed similar results to the absolute ADC value analysis. ROC curve analysis for normalized ADC_median values to thalamus showed 94.9% sensitivity (95% CI: 85-100%) and 93.3% specificity (95% CI: 87-100%) for differentiating medulloblastoma from ependymoma.

CONCLUSION

ADC histogram metrics can be applied to differentiate most types of posterior fossa tumors in children.

The Diagnostic Efficiency of Quantitative Diffusion Weighted Imaging in Differentiating Medulloblastoma from Posterior Fossa Tumors: A Systematic Review and Meta-Analysis

Medulloblastoma (MB) is considered the most common and highly malignant posterior fossa tumor (PFT) in children. The accurate preoperative diagnosis of MB is beneficial in choosing the appropriate surgical methods and treatment strategies. Diffusion-weighted imaging (DWI) has improved the accuracy of differential diagnosis of posterior fossa tumors. Nonetheless, further studies are needed to confirm its value for clinical application. This study aimed to evaluate the performance of DWI in differentiating MB from other PFT. A literature search was conducted using databases PubMed, Embase, and Web of Science for studies reporting the diagnostic performance of DWI for PFT from January 2000 to January 2022. A bivariate random-effects model was employed to evaluate the pooled sensitivities and specificities. A univariable meta-regression analysis was used to assess relevant factors for heterogeneity, and subgroup analyses were performed. A total of 15 studies with 823 patients were eligible for data extraction. Overall pooled sensitivity and specificity of DWI were 0.94 (95% confident interval [CI]: 0.89-0.97) and 0.94 (95% CI: 0.90-0.96) respectively. The area under the curve (AUC) of DWI was 0.98 (95% CI: 0.96-0.99). Heterogeneity was found in the sensitivity (I² = 62.59%) and the specificity (I² = 35.94%). Magnetic field intensity, region of interest definition and DWI diagnostic parameters are the factors that affect the diagnostic performance of DWI. DWI has excellent diagnostic accuracy for differentiating MB from other PFT. Hence, it is necessary to set DWI as a routine examination sequence for posterior fossa tumors.

Machine Learning in the Classification of Pediatric Posterior Fossa Tumors: A Systematic Review

Background: Posterior fossa tumors (PFTs) are a morbid group of central nervous system tumors that most often present in childhood. While early diagnosis is critical to drive appropriate treatment, definitive diagnosis is currently only achievable through invasive tissue collection and histopathological analyses. Machine learning has been investigated as an alternative means of diagnosis. In this systematic review and meta-analysis, we evaluated the primary literature to identify all machine learning algorithms developed to classify and diagnose pediatric PFTs using imaging or molecular data. Methods: Of the 433 primary papers identified in PubMed, EMBASE, and Web of Science, 25 ultimately met the inclusion criteria. The included papers were extracted for algorithm architecture, study parameters, performance, strengths, and limitations. Results: The algorithms exhibited variable performance based on sample size, classifier(s) used, and individual tumor types being investigated. Ependymoma, medulloblastoma, and pilocytic astrocytoma were the most studied tumors with algorithm accuracies ranging from 37.5% to 94.5%. A minority of studies compared the developed algorithm to a trained neuroradiologist, with three imaging-based algorithms yielding superior performance. Common algorithm and study limitations included small sample sizes, uneven representation of individual tumor types, inconsistent performance reporting, and a lack of application in the clinical environment. Conclusions: Artificial intelligence has the potential to improve the speed and accuracy of diagnosis in this field if the right algorithm is applied to the right scenario. Work is needed to standardize outcome reporting and facilitate additional trials to allow for clinical uptake.

Characterization of pediatric brain tumors using pre-diagnostic neuroimaging

Purpose

To evaluate for predictive neuroimaging features of pediatric brain tumor development and quantify tumor growth characteristics in patients who had neuroimaging performed prior to a diagnosis of a brain tumor.

Methods

Retrospective review of 1098 consecutive pediatric patients at a single institution with newly diagnosed brain tumors from January 2009 to October 2021 was performed to identify patients with neuroimaging prior to the diagnosis of a brain tumor. Pre-diagnostic and diagnostic neuroimaging features (e.g., tumor size, apparent diffusion coefficient (ADC) values), clinical presentations, and neuropathology were recorded in those patients who had neuroimaging performed prior to a brain tumor diagnosis. High- and low-grade tumor sizes were fit to linear and exponential growth regression models.

Results

Fourteen of 1098 patients (1%) had neuroimaging prior to diagnosis of a brain tumor (8 females, mean age at definitive diagnosis 8.1 years, imaging interval 0.2-8.7 years). Tumor types included low-grade glioma (n = 4), embryonal tumors (n = 2), pineal tumors (n=2), ependymoma (n = 3), and others (n = 3). Pre-diagnostic imaging of corresponding tumor growth sites were abnormal in four cases (28%) and demonstrated higher ADC values in the region of high-grade tumor growth (p = 0.05). Growth regression analyses demonstrated R²-values of 0.92 and 0.91 using a linear model and 0.64 and 0.89 using an exponential model for high- and low-grade tumors, respectively; estimated minimum velocity of diameter expansion was 2.4 cm/year for high-grade and 0.4 cm/year for low-grade tumors. High-grade tumors demonstrated faster growth rate of diameter and solid tumor volume compared to low-grade tumors (p = 0.02, p = 0.03, respectively).

Conclusions

This is the first study to test feasibility in utilizing pre-diagnostic neuroimaging to demonstrate that linear and exponential growth rate models can be used to estimate pediatric brain tumor growth velocity and should be validated in a larger multi-institutional cohort.

Ensemble learning based functional independence ability estimator for pediatric brain tumor survivors

A history of brain tumor strongly affects children’s cognitive abilities, performance of daily activities, quality of life, and functional outcomes. In light of the difficulties in cognition, communication, physical skills, and behavior that these patients may encounter, occupational therapists should perform a comprehensive needs-led assessment of their global functioning after recovery. Such an assessment would ensure that the patients receive adequate support and services at school, at home, and in the community. By predicting the functional activity performance of children with a history of brain tumor, clinical workers can determine the progress of their ability recovery and the optimal treatment plan. We selected several features for testing and employed common machine learning models to predict Functional Independence Measure (WeeFIM) scores. The ensemble learning models exhibited stronger predictive performance than did the individual machine learning models. The ensemble learning models effectively predicted WeeFIM scores. Machine learning models can help clinical workers predict the functional assessment scores of patients with childhood brain tumors. This study used machine learning models to predict the WeeFIM scores of patients with childhood brain tumors and to demonstrate that ensemble machine learning models are more suitable for this task than are individual machine learning models.

Advanced Neuroimaging Approaches to Pediatric Brain Tumors

Simple Summary

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

Abstract

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

MRI-based Whole-Tumor Radiomics to Classify the Types of Pediatric Posterior Fossa Brain Tumor

BACKGROUND

Differential diagnosis between medulloblastoma (MB), ependymoma (EP) and astrocytoma (PA) is important due to differing medical treatment strategies and predicted survival. The aim of this study was to investigate non-invasive MRI-based radiomic analysis of whole tumors to classify the histologic tumor types of pediatric posterior fossa brain tumor and improve the accuracy of discrimination, using a random forest classifier.

METHODS

MRI images of 99 patients, with 59 MBs, 13 EPs and 27 PAs histologically confirmed by surgery and pathology before treatment, were included in this retrospective study. Registration was performed between the three sequences, and high- throughput features were extracted from manually segmented tumors on MR images of each case. The forest-based feature selection method was adopted to select the top ten significant features. Finally, the results were compared and analyzed according to the classification.

RESULTS

The top ten contributions according to the classifier of wavelet features all came from the ADC sequence. The random forest classifier achieved 100% accuracy on the training data and validated the best accuracy (0.938): sensitivity = 1.000, 0.948 and 0.808, specificity = 0.952, 0.926 and 1.000 for EP, MB and PA, respectively.

CONCLUSION

A random forest classifier based on the ADC sequence of the whole tumor provides more quantitative information than TIWI and T2WI in differentiating pediatric posterior fossa brain tumors. In particular, the histogram percentile value showed great superiority, which added diagnostic value in pediatric neuro-oncology.

Dynamic susceptibility-contrast magnetic resonance imaging with contrast agent leakage correction aids in predicting grade in pediatric brain tumours: a multicenter study

BACKGROUND

Relative cerebral blood volume (rCBV) measured using dynamic susceptibility-contrast MRI can differentiate between low- and high-grade pediatric brain tumors. Multicenter studies are required for translation into clinical practice.

OBJECTIVE

We compared leakage-corrected dynamic susceptibility-contrast MRI perfusion parameters acquired at multiple centers in low- and high-grade pediatric brain tumors.

MATERIALS AND METHODS

Eighty-five pediatric patients underwent pre-treatment dynamic susceptibility-contrast MRI scans at four centers. MRI protocols were variable. We analyzed data using the Boxerman leakage-correction method producing pixel-by-pixel estimates of leakage-uncorrected (rCBVuncorr) and corrected (rCBVcorr) relative cerebral blood volume, and the leakage parameter, K2. Histological diagnoses were obtained. Tumors were classified by high-grade tumor. We compared whole-tumor median perfusion parameters between low- and high-grade tumors and across tumor types.

RESULTS

Forty tumors were classified as low grade, 45 as high grade. Mean whole-tumor median rCBVuncorr was higher in high-grade tumors than low-grade tumors (mean ± standard deviation [SD] = 2.37±2.61 vs. -0.14±5.55; P<0.01). Average median rCBV increased following leakage correction (2.54±1.63 vs. 1.68±1.36; P=0.010), remaining higher in high-grade tumors than low grade-tumors. Low-grade tumors, particularly pilocytic astrocytomas, showed T1-dominant leakage effects; high-grade tumors showed T2*-dominance (mean K2=0.017±0.049 vs. 0.002±0.017). Parameters varied with tumor type but not center. Median rCBVuncorr was higher (mean = 1.49 vs. 0.49; P=0.015) and K2 lower (mean = 0.005 vs. 0.016; P=0.013) in children who received a pre-bolus of contrast agent compared to those who did not. Leakage correction removed the difference.

CONCLUSION

Dynamic susceptibility-contrast MRI acquired at multiple centers helped distinguish between children's brain tumors. Relative cerebral blood volume was significantly higher in high-grade compared to low-grade tumors and differed among common tumor types. Vessel leakage correction is required to provide accurate rCBV, particularly in low-grade enhancing tumors.

Diagnostic accuracy of qualitative MRI in 550 paediatric brain tumours: evaluating current practice in the computational era

BACKGROUND

To investigate the accuracy of qualitative reporting of conventional magnetic resonance imaging (MRI) in the classification of paediatric brain tumours.

METHODS

Preoperative MRI reports of 608 children prior to resection or biopsy of an intracranial lesion were retrospectively reviewed. A total of 550 children had complete radiological and histopathological notes, thereby reaching our inclusion criteria. Concordance between MRI report and final histopathological diagnosis was assessed using an established lexicon derived from the WHO 2016 classification of CNS tumours. Levels of agreement based on cellular origin, tumour type, and tumour grade were evaluated. Diagnostic accuracy, sensitivity, specificity, confidence intervals, and positive and negative predictive values were calculated.

RESULTS

Diagnostic accuracy differed significantly between tumour types and tumour grades. Sensitivities were highest for ependymomas and sellar, pituitary, pineal, and cranial and/or paraspinal nerve tumours (range 80.65-100%). Sensitivity was slightly lower for astrocytic gliomas, oligodendrogliomas, and choroid plexus, neuronal, mixed neuronal-glial, embryonal, and histiocytic tumours (range 63.33-79.59%). Low sensitivities were noted for meningiomas and mesenchymal non-meningothelial, melanocytic, and germ cell tumours (range 0-56.25%). The most correct tumour type predictions were made in the posterior fossa whilst the most incorrect predictions were made in the lobar regions, pineal/tectal plate area, and the supratentorial ventricles.

CONCLUSIONS

This is the largest published series investigating the predictive accuracy of MRI in paediatric brain tumours. We show that diagnostic accuracy varies greatly by tumour type and location. Looking forward, we should develop and leverage computational methods to improve accuracy in the tumour types and anatomical locations where qualitative diagnostic accuracy is lower.

Radiomics and radiogenomics in pediatric neuro-oncology: A review

The current era of advanced computing has allowed for the development and implementation of the field of radiomics. In pediatric neuro-oncology, radiomics has been applied in determination of tumor histology, identification of disseminated disease, prognostication, and molecular classification of tumors (ie, radiogenomics). The field also comes with many challenges, such as limitations in study sample sizes, class imbalance, generalizability of the methods, and data harmonization across imaging centers. The aim of this review paper is twofold: first, to summarize existing literature in radiomics of pediatric neuro-oncology; second, to distill the themes and challenges of the field and discuss future directions in both a clinical and technical context.

Meta-Analysis of Apparent Diffusion Coefficient in Pediatric Medulloblastoma, Ependymoma, and Pilocytic Astrocytoma

BACKGROUND

Medulloblastoma, ependymoma, and pilocytic astrocytoma are common pediatric posterior fossa tumors. These tumors show overlapping characteristics on conventional MRI scans, making diagnosis difficult.

PURPOSE

To investigate whether apparent diffusion coefficient (ADC) values differ between tumor types and to identify optimum cut-off values to accurately classify the tumors using different performance metrics.

STUDY TYPE

Systematic review and meta-analysis.

SUBJECTS

Seven studies reporting ADC in pediatric posterior fossa tumors (115 medulloblastoma, 68 ependymoma, and 86 pilocytic astrocytoma) were included following PubMed and ScienceDirect searches.

SEQUENCE AND FIELD STRENGTH

Diffusion weighted imaging (DWI) was performed on 1.5 and 3 T across multiple institution and vendors.

ASSESSMENT

The combined mean and standard deviation of ADC were calculated for each tumor type using a random-effects model, and the effect size was calculated using Hedge's g.

STATISTICAL TESTS

Sensitivity/specificity, weighted classification accuracy, balanced classification accuracy. A P value < 0.05 was considered statistically significant, and a Hedge's g value of >1.2 was considered to represent a large difference.

RESULTS

The mean (± standard deviation) ADCs of medulloblastoma, ependymoma, and pilocytic astrocytoma were 0.76 ± 0.16, 1.10 ± 0.10, and 1.49 ± 0.16 mm² /sec × 10^-3 . To maximize sensitivity and specificity using the mean ADC, the cut-off was found to be 0.96 mm² /sec × 10^-3 for medulloblastoma and ependymoma and 1.26 mm² /sec × 10^-3 for ependymoma and pilocytic astrocytoma. The meta-analysis showed significantly different ADC distributions for the three posterior fossa tumors. The cut-off values changed markedly (up to 7%) based on the performance metric used and the prevalence of the tumor types.

DATA CONCLUSION

There were significant differences in ADC between tumor types. However, it should be noted that only summary statistics from each study were analyzed and there were differences in how regions of interest were defined between studies.

EVIDENCE LEVEL

1 TECHNICAL EFFICACY: Stage 3.

Combining multi-site magnetic resonance imaging with machine learning predicts survival in pediatric brain tumors

Brain tumors represent the highest cause of mortality in the pediatric oncological population. Diagnosis is commonly performed with magnetic resonance imaging. Survival biomarkers are challenging to identify due to the relatively low numbers of individual tumor types. 69 children with biopsy-confirmed brain tumors were recruited into this study. All participants had perfusion and diffusion weighted imaging performed at diagnosis. Imaging data were processed using conventional methods, and a Bayesian survival analysis performed. Unsupervised and supervised machine learning were performed with the survival features, to determine novel sub-groups related to survival. Sub-group analysis was undertaken to understand differences in imaging features. Survival analysis showed that a combination of diffusion and perfusion imaging were able to determine two novel sub-groups of brain tumors with different survival characteristics (p < 0.01), which were subsequently classified with high accuracy (98%) by a neural network. Analysis of high-grade tumors showed a marked difference in survival (p = 0.029) between the two clusters with high risk and low risk imaging features. This study has developed a novel model of survival for pediatric brain tumors. Tumor perfusion plays a key role in determining survival and should be considered as a high priority for future imaging protocols.

European Society for Paediatric Oncology (SIOPE) MRI guidelines for imaging patients with central nervous system tumours

INTRODUCTION

Standardisation of imaging acquisition is essential in facilitating multicentre studies related to childhood CNS tumours. It is important to ensure that the imaging protocol can be adopted by centres with varying imaging capabilities without compromising image quality.

MATERIALS AND METHOD

An imaging protocol has been developed by the Brain Tumour Imaging Working Group of the European Society for Paediatric Oncology (SIOPE) based on consensus among its members, which consists of neuroradiologists, imaging scientists and paediatric neuro-oncologists. This protocol has been developed to facilitate SIOPE led studies and regularly reviewed by the imaging working group.

RESULTS

The protocol consists of essential MRI sequences with imaging parameters for 1.5 and 3 Tesla MRI scanners and a set of optional sequences that can be used in appropriate clinical settings. The protocol also provides guidelines for early post-operative imaging and surveillance imaging. The complementary use of multimodal advanced MRI including diffusion tensor imaging (DTI), MR spectroscopy and perfusion imaging is encouraged, and optional guidance is provided in this publication.

CONCLUSION

The SIOPE brain tumour imaging protocol will enable consistent imaging across multiple centres involved in paediatric CNS tumour studies.

Artificial intelligence and machine learning for medical imaging: A technology review

Artificial intelligence (AI) has recently become a very popular buzzword, as a consequence of disruptive technical advances and impressive experimental results, notably in the field of image analysis and processing. In medicine, specialties where images are central, like radiology, pathology or oncology, have seized the opportunity and considerable efforts in research and development have been deployed to transfer the potential of AI to clinical applications. With AI becoming a more mainstream tool for typical medical imaging analysis tasks, such as diagnosis, segmentation, or classification, the key for a safe and efficient use of clinical AI applications relies, in part, on informed practitioners. The aim of this review is to present the basic technological pillars of AI, together with the state-of-the-art machine learning methods and their application to medical imaging. In addition, we discuss the new trends and future research directions. This will help the reader to understand how AI methods are now becoming an ubiquitous tool in any medical image analysis workflow and pave the way for the clinical implementation of AI-based solutions.