An international study presenting a federated learning AI platform for pediatric brain tumors.

While multiple factors impact disease, artificial intelligence (AI) studies in medicine often use small, non-diverse patient cohorts due to data sharing and privacy issues. Federated learning (FL) has emerged as a solution, enabling training across hospitals without direct data sharing. Here, we present FL-PedBrain, an FL platform for pediatric posterior fossa brain tumors, and evaluate its performance on a diverse, realistic, multi-center cohort. Pediatric brain tumors were targeted due to the scarcity of such datasets, even in tertiary care hospitals. Our platform orchestrates federated training for joint tumor classification and segmentation across 19 international sites. FL-PedBrain exhibits less than a 1.5% decrease in classification and a 3% reduction in segmentation performance compared to centralized data training. FL boosts segmentation performance by 20 to 30% on three external, out-of-network sites. Finally, we explore the sources of data heterogeneity and examine FL robustness in real-world scenarios with data imbalances.

Beyond hand-crafted features for pretherapeutic molecular status identification of pediatric low-grade gliomas.

The use of targeted agents in the treatment of pediatric low-grade gliomas (pLGGs) relies on the determination of molecular status. It has been shown that genetic alterations in pLGG can be identified non-invasively using MRI-based radiomic features or convolutional neural networks (CNNs). We aimed to build and assess a combined radiomics and CNN non-invasive pLGG molecular status identification model. This retrospective study used the tumor regions, manually segmented from T2-FLAIR MR images, of 336 patients treated for pLGG between 1999 and 2018. We designed a CNN and Random Forest radiomics model, along with a model relying on a combination of CNN and radiomic features, to predict the genetic status of pLGG. Additionally, we investigated whether CNNs could predict radiomic feature values from MR images. The combined model (mean AUC: 0.824) outperformed the radiomics model (0.802) and CNN (0.764). The differences in model performance were statistically significant (p-values < 0.05). The CNN was able to learn predictive radiomic features such as surface-to-volume ratio (average correlation: 0.864), and difference matrix dependence non-uniformity normalized (0.924) well but was unable to learn others such as run-length matrix variance (- 0.017) and non-uniformity normalized (- 0.042). Our results show that a model relying on both CNN and radiomic-based features performs better than either approach separately in differentiating the genetic status of pLGGs, and that CNNs are unable to express all handcrafted features.

Developmental Curves of the Paediatric Brain Using FLAIR MRI Texture Biomarkers.

Purpose: Analysis of FLAIR MRI sequences is gaining momentum in brain maturation studies, and this study aimed to establish normative developmental curves for FLAIR texture biomarkers in the paediatric brain. Methods: A retrospective, single-centre dataset of 465/512 healthy paediatric FLAIR volumes was used, with one pathological volume for proof-of-concept. Participants were included if the MRI was unremarkable as determined by a neuroradiologist. An automated intensity normalization algorithm was used to standardize FLAIR signal intensity across MRI scanners and individuals. FLAIR texture biomarkers were extracted from grey matter (GM), white matter (WM), deep GM, and cortical GM regions. Sex-specific percentile curves were reported and modelled for each tissue type. Correlations between texture and established biomarkers including intensity volume were examined. Biomarkers from the pathological volume were extracted to demonstrate clinical utility of normative curves. Results: This study analyzed 465 FLAIR sequences in children and adolescents (mean age 10.65 ± 4.22 years, range 2-19 years, 220 males, 245 females). In the WM, texture increased to a maximum at around 8 to 10 years, with different trends between females and males in adolescence. In the GM, texture increased over the age range while demonstrating a local maximum at 8 to 10 years. Texture had an inverse relationship with intensity in the WM across all ages. WM and edema in a pathological brain exhibited abnormal texture values outside of the normative growth curves. Conclusion: Normative curves for texture biomarkers in FLAIR sequences may be used to assess brain maturation and microstructural changes over the paediatric age range.

Pediatric-type diffuse low-grade gliomas.

The World Health Organization's 5th edition classification of Central Nervous System (CNS) tumors differentiates diffuse gliomas into adult and pediatric variants. Pediatric-type diffuse low-grade gliomas (pDLGGs) are distinct from adult gliomas in their molecular characteristics, biological behavior, clinical progression, and prognosis. Various molecular alterations identified in pDLGGs are crucial for treatment. There are four distinct entities of pDLGGs. All four of these tumor subtypes exhibit diffuse growth and share overlapping histopathological and imaging characteristics. Molecular analysis is essential for differentiating these lesions.

Identification of Multiclass Pediatric Low-Grade Neuroepithelial Tumor Molecular Subtype with ADC MR Imaging and Machine Learning.

BACKGROUND AND PURPOSE: Molecular biomarker identification increasingly influences the treatment planning of pediatric low-grade neuroepithelial tumors (PLGNTs). We aimed to develop and validate a radiomics-based ADC signature predictive of the molecular status of PLGNTs. MATERIALS AND METHODS: In this retrospective bi-institutional study, we searched the PACS for baseline brain MRIs from children with PLGNTs. Semiautomated tumor segmentation on ADC maps was performed using the semiautomated level tracing effect tool with 3D Slicer. Clinical variables, including age, sex, and tumor location, were collected from chart review. The molecular status of tumors was derived from biopsy. Multiclass random forests were used to predict the molecular status and fine-tuned using a grid search on the validation sets. Models were evaluated using independent and unseen test sets based on the combined data, and the area under the receiver operating characteristic curve (AUC) was calculated for the prediction of 3 classes: KIAA1549-BRAF fusion, BRAF V600E mutation, and non-BRAF cohorts. Experiments were repeated 100 times using different random data splits and model initializations to ensure reproducible results. RESULTS: Two hundred ninety-nine children from the first institution and 23 children from the second institution were included (53.6% male; mean, age 8.01 years; 51.8% supratentorial; 52.2% with KIAA1549-BRAF fusion). For the 3-class prediction using radiomics features only, the average test AUC was 0.74 (95% CI, 0.73-0.75), and using clinical features only, the average test AUC was 0.67 (95% CI, 0.66-0.68). The combination of both radiomics and clinical features improved the AUC to 0.77 (95% CI, 0.75-0.77). The diagnostic performance of the per-class test AUC was higher in identifying KIAA1549-BRAF fusion tumors among the other subgroups (AUC = 0.81 for the combined radiomics and clinical features versus 0.75 and 0.74 for BRAF V600E mutation and non-BRAF, respectively). CONCLUSIONS: ADC values of tumor segmentations have differentiative signals that can be used for training machine learning classifiers for molecular biomarker identification of PLGNTs. ADC-based pretherapeutic differentiation of the BRAF status of PLGNTs has the potential to avoid invasive tumor biopsy and enable earlier initiation of targeted therapy.

Brain Volumes in Opsoclonus-Myoclonus Ataxia Syndrome: A Longitudinal Study.

INTRODUCTION: Little is known about the longitudinal trajectory of brain growth in children with opsoclonus-myoclonus ataxia syndrome. We performed a longitudinal evaluation of brain volumes in pediatric opsoclonus-myoclonus ataxia syndrome patients compared with age- and sex-matched healthy children. PATIENTS AND METHODS: This longitudinal case-control study included brain magnetic resonance imaging (MRI) scans from consecutive pediatric opsoclonus-myoclonus ataxia syndrome patients (2009-2020) and age- and sex-matched healthy control children. FreeSurfer analysis provided automatic volumetry of the brain. Paired t tests were performed on the curvature of growth trajectories, with Bonferroni correction. RESULTS: A total of 14 opsoclonus-myoclonus ataxia syndrome patients (12 female) and 474 healthy control children (406 female) were included. Curvature of the growth trajectories of the cerebral white and gray matter, cerebellar white and gray matter, and brainstem differed significantly between opsoclonus-myoclonus ataxia syndrome patients and healthy control children (cerebral white matter, P = .01; cerebral gray matter, P = .01; cerebellar white matter, P < .001; cerebellar gray matter, P = .049; brainstem, P < .01). DISCUSSION/CONCLUSION: We found abnormal brain maturation in the supratentorial brain, brainstem, and cerebellum in children with opsoclonus-myoclonus ataxia syndrome.

Brain Tumor Imaging in Adolescents and Young Adults: 2021 WHO Updates for Molecular-based Tumor Types.

Published in 2021, the fifth edition of the World Health Organization (WHO) classification of tumors of the central nervous system (CNS) introduced new molecular criteria for tumor types that commonly occur in either pediatric or adult age groups. Adolescents and young adults (AYAs) are at the intersection of adult and pediatric care, and both pediatric-type and adult-type CNS tumors occur at that age. Mortality rates for AYAs with CNS tumors have increased by 0.6% per year for males and 1% per year for females from 2007 to 2016. To best serve patients, it is crucial that both pediatric and adult radiologists who interpret neuroimages are familiar with the various pediatric- and adult-type brain tumors and their typical imaging morphologic characteristics. Gliomas account for approximately 80% of all malignant CNS tumors in the AYA age group, with the most common types observed being diffuse astrocytic and glioneuronal tumors. Ependymomas and medulloblastomas also occur in the AYA population but are seen less frequently. Importantly, biologic behavior and progression of distinct molecular subgroups of brain tumors differ across ages. This review discusses newly added or revised gliomas in the fifth edition of the CNS WHO classification, as well as other CNS tumor types common in the AYA population.

The unintended consequences of artificial intelligence in paediatric radiology.

Over the past decade, there has been a dramatic rise in the interest relating to the application of artificial intelligence (AI) in radiology. Originally only 'narrow' AI tasks were possible; however, with increasing availability of data, teamed with ease of access to powerful computer processing capabilities, we are becoming more able to generate complex and nuanced prediction models and elaborate solutions for healthcare. Nevertheless, these AI models are not without their failings, and sometimes the intended use for these solutions may not lead to predictable impacts for patients, society or those working within the healthcare profession. In this article, we provide an overview of the latest opinions regarding AI ethics, bias, limitations, challenges and considerations that we should all contemplate in this exciting and expanding field, with a special attention to how this applies to the unique aspects of a paediatric population. By embracing AI technology and fostering a multidisciplinary approach, it is hoped that we can harness the power AI brings whilst minimising harm and ensuring a beneficial impact on radiology practice.

Increased confidence of radiomics facilitating pretherapeutic differentiation of BRAF-altered pediatric low-grade glioma.

OBJECTIVES: Currently, the BRAF status of pediatric low-grade glioma (pLGG) patients is determined through a biopsy. We established a nomogram to predict BRAF status non-invasively using clinical and radiomic factors. Additionally, we assessed an advanced thresholding method to provide only high-confidence predictions for the molecular subtype. Finally, we tested whether radiomic features provide additional predictive information for this classification task, beyond that which is embedded in the location of the tumor. METHODS: Random forest (RF) models were trained on radiomic and clinical features both separately and together, to evaluate the utility of each feature set. Instead of using the traditional single threshold technique to convert the model outputs to class predictions, we implemented a double threshold mechanism that accounted for uncertainty. Additionally, a linear model was trained and depicted graphically as a nomogram. RESULTS: The combined RF (AUC: 0.925) outperformed the RFs trained on radiomic (AUC: 0.863) or clinical (AUC: 0.889) features alone. The linear model had a comparable AUC (0.916), despite its lower complexity. Traditional thresholding produced an accuracy of 84.5%, while the double threshold approach yielded 92.2% accuracy on the 80.7% of patients with the highest confidence predictions. CONCLUSION: Models that included radiomic features outperformed, underscoring their importance for the prediction of BRAF status. A linear model performed similarly to RF but with the added benefit that it can be visualized as a nomogram, improving the explainability of the model. The double threshold technique was able to identify uncertain predictions, enhancing the clinical utility of the model. CLINICAL RELEVANCE STATEMENT: Radiomic features and tumor location are both predictive of BRAF status in pLGG patients. We show that they contain complementary information and depict the optimal model as a nomogram, which can be used as a non-invasive alternative to biopsy. KEY POINTS: • Radiomic features provide additional predictive information for the determination of the molecular subtype of pediatric low-grade gliomas patients, beyond what is embedded in the location of the tumor, which has an established relationship with genetic status. • An advanced thresholding method can help to distinguish cases where machine learning models have a high chance of being (in)correct, improving the utility of these models. • A simple linear model performs similarly to a more powerful random forest model at classifying the molecular subtype of pediatric low-grade gliomas but has the added benefit that it can be converted into a nomogram, which may facilitate clinical implementation by improving the explainability of the model.

MRI-Based End-To-End Pediatric Low-Grade Glioma Segmentation and Classification.

Purpose: MRI-based radiomics models can predict genetic markers in pediatric low-grade glioma (pLGG). These models usually require tumour segmentation, which is tedious and time consuming if done manually. We propose a deep learning (DL) model to automate tumour segmentation and build an end-to-end radiomics-based pipeline for pLGG classification. Methods: The proposed architecture is a 2-step U-Net based DL network. The first U-Net is trained on downsampled images to locate the tumour. The second U-Net is trained using image patches centred around the located tumour to produce more refined segmentations. The segmented tumour is then fed into a radiomics-based model to predict the genetic marker of the tumour. Results: Our segmentation model achieved a correlation value of over 80% for all volume-related radiomic features and an average Dice score of .795 in test cases. Feeding the auto-segmentation results into a radiomics model resulted in a mean area under the ROC curve (AUC) of .843, with 95% confidence interval (CI) [.78-.906] and .730, with 95% CI [.671-.789] on the test set for 2-class (BRAF V600E mutation BRAF fusion) and 3-class (BRAF V600E mutation BRAF fusion and Other) classification, respectively. This result was comparable to the AUC of .874, 95% CI [.829-.919] and .758, 95% CI [.724-.792] for the radiomics model trained and tested on the manual segmentations in 2-class and 3-class classification scenarios, respectively. Conclusion: The proposed end-to-end pipeline for pLGG segmentation and classification produced results comparable to manual segmentation when it was used for a radiomics-based genetic marker prediction model.

Accuracy of Information and References Using ChatGPT-3 for Retrieval of Clinical Radiological Information.

Purpose: To assess the accuracy of answers provided by ChatGPT-3 when prompted with questions from the daily routine of radiologists and to evaluate the text response when ChatGPT-3 was prompted to provide references for a given answer. Methods: ChatGPT-3 (San Francisco, OpenAI) is an artificial intelligence chatbot based on a large language model (LLM) that has been designed to generate human-like text. A total of 88 questions were submitted to ChatGPT-3 using textual prompt. These 88 questions were equally dispersed across 8 subspecialty areas of radiology. The responses provided by ChatGPT-3 were assessed for correctness by cross-checking them with peer-reviewed, PubMed-listed references. In addition, the references provided by ChatGPT-3 were evaluated for authenticity. Results: A total of 59 of 88 responses (67%) to radiological questions were correct, while 29 responses (33%) had errors. Out of 343 references provided, only 124 references (36.2%) were available through internet search, while 219 references (63.8%) appeared to be generated by ChatGPT-3. When examining the 124 identified references, only 47 references (37.9%) were considered to provide enough background to correctly answer 24 questions (37.5%). Conclusion: In this pilot study, ChatGPT-3 provided correct responses to questions from the daily clinical routine of radiologists in only about two thirds, while the remainder of responses contained errors. The majority of provided references were not found and only a minority of the provided references contained the correct information to answer the question. Caution is advised when using ChatGPT-3 to retrieve radiological information.

Increased rate of significant findings on brain MRI during the early stage of the COVID-19 pandemic.

OBJECTIVES: To assess the effect of the COVID-19 pandemic on the proportion of abnormal paediatric neuroimaging findings as a surrogate marker for potential underutilisation. METHODS: Consecutive paediatric brain MRIs performed between March 27th and June 19th 2019 (Tbaseline) and March 23rd and June 1st 2020 (Tpandemic) were reviewed and classified according to presence or absence and type of imaging abnormality, and graded regarding severity on a 5-point Likert scale, where grade 4 was defined as abnormal finding requiring non-urgent intervention and grade 5 was defined as acute illness prompting urgent medical intervention. Non-parametric statistical testing was used to assess for significant differences between Tpandemic vs. Tbaseline. RESULTS: Fewer paediatric MRI brains were performed during Tpandemic compared to Tbaseline (12.2 vs 14.7 examinations/day). No significant difference was found between the two time periods regarding sex and age (Tbaseline: 557 females (44.63%), 7.95 ± 5.49 years, Tpandemic: 385 females (44.61%), 7.64 ± 6.11 years; p = 1 and p = .079, respectively). MRI brain examinations during Tpandemic had a higher likelihood of being abnormal, 41.25% vs. 25.32% (p<.0001). Vascular abnormalities were more frequent during Tpandemic (11.01% vs 8.01%, p = .02), congenital malformations were less common (8.34% vs 12.34%, p = .004). Severity of MRI brain examinations was significantly different when comparing group 4 and group 5 individually and combined between Tbaseline and Tpandemic (p = .0018, p < .0001, and p <.0001, respectively). CONCLUSIONS: The rate of abnormality and severity found on paediatric brain MRI was significantly higher during the early phase of the pandemic, likely due to underutilisation.

Volumetric differences of thalamic nuclei in children with trisomy 21.

OBJECTIVES: Histological studies have shown alterations of thalamic nuclei in patients with Down syndrome (DS). The correlation of these changes on MRI (magnetic resonance imaging) is unclear. Therefore, this study investigates volumetric differences of thalamic nuclei in children with DS compared to controls. METHODS: Patients were retrospectively identified between 01/2000 and 10/2021. Patient inclusion criteria were: (1) 0-18 years of age, (2) diagnosis of DS, and (3) availability of a brain MRI without parenchymal injury and a non-motion-degraded volumetric T1-weighted sequence. Whole thalamus and thalamic nuclei (n = 25) volumes were analyzed bilaterally relative to the total brain volume (TBV). Two-sided t-tests were used to evaluate differences between groups. Differences were considered significant if the adjusted p-value was <0.05 after correction for multiple hypothesis testing using the Holm-Bonferroni method. RESULTS: 21 children with DS (11 females, 52.4%, mean age: 8.6 ± 4.3 years) and 63 age- and sex-matched controls (32 females, 50.8%, 8.6 ± 4.3 years) were studied using automated volumetric segmentation. Significantly smaller ratios were found for nine thalamic nuclei and the whole thalamus on the right and five thalamic nuclei on the left. TBV was significantly smaller in patients with DS (p < 0.001). No significant differences were found between the groups for age and sex. CONCLUSIONS: In this exploratory volumetric analysis of the thalamus and thalamic nuclei, we observed statistically significant volumetric changes in children with DS. Our findings confirm prior neuroimaging and histological studies and extend the range of involved thalamic nuclei in pediatric DS.

MRI Volumetric Analysis of the Hypothalamus and Limbic System across the Pediatric Age Span.

PURPOSE: Literature is scarce regarding volumetric measures of limbic system components across the pediatric age range. The purpose of this study is to remedy this scarcity by reporting continuous volumetric measurements of limbic system components, and to provide consistent stratification data including age-related trajectories and sex-related differences in the pediatric age range in order to improve the recognition of structural variations that might reflect pathology. METHODS: In this retrospective study, MRI sequences of children with normal clinical MRI examinations of the brain acquired between January 2010 and December 2019 were included. Isotropic 3D T1-weighted were processed using FreeSurfer version 7.3. Total brain volume and volumes of the limbic system including the hippocampus, parahippocampal gyrus, amygdala, hypothalamus, cingulate gyrus, entorhinal cortex, anteroventral thalamic nucleus, and whole thalamus were assessed. Parcellated output was displayed with the respective label map overlay and images were visually inspected for accuracy of regional segmentation results. Continuous data are provided as mean and standard deviation with quadratic trendlines and as mean and 95% confidence intervals. Categorical data are presented as integers and percentages (%). RESULTS: A total of 724 children (401 female, 55.4%), with a mean age at time of MRI of 10.9 ± 4.2 years (range: 1.9-18.2 years), were included in the study. For females, the total brain volume increased from 955 ± 70 mL at the age of 2-3 years to 1140 ± 110 mL at the age of 17-18 years. Similarly, the total brain volume increased for males from 1004 ± 83 mL to 1263 ± 96 mL. The maximum volume was noted at 11-12 years for females (1188 ± 90 mL) and at 14-15 years for males (1310 ± 159 mL). Limbic system structures reached their peak volume more commonly between the 13-14 years to 17-18 years age groups. The male cingulate gyrus, entorhinal cortex, and anteroventral thalamic nucleus reached peak volume before or at 9-10 years. CONCLUSION: This study provides unique age- and sex-specific volumes of the components of the limbic system throughout the pediatric age range to serve as normal values in comparative studies. Quantification of volumetric abnormalities of the limbic system on brain MRI may offer insights into phenotypical variations of diseases and may help elucidate new pathological phenotypes.

Fet-Net Algorithm for Automatic Detection of Fetal Orientation in Fetal MRI.

Identifying fetal orientation is essential for determining the mode of delivery and for sequence planning in fetal magnetic resonance imaging (MRI). This manuscript describes a deep learning algorithm named Fet-Net, composed of convolutional neural networks (CNNs), which allows for the automatic detection of fetal orientation from a two-dimensional (2D) MRI slice. The architecture consists of four convolutional layers, which feed into a simple artificial neural network. Compared with eleven other prominent CNNs (different versions of ResNet, VGG, Xception, and Inception), Fet-Net has fewer architectural layers and parameters. From 144 3D MRI datasets indicative of vertex, breech, oblique and transverse fetal orientations, 6120 2D MRI slices were extracted to train, validate and test Fet-Net. Despite its simpler architecture, Fet-Net demonstrated an average accuracy and F1 score of 97.68% and a loss of 0.06828 on the 6120 2D MRI slices during a 5-fold cross-validation experiment. This architecture outperformed all eleven prominent architectures (p < 0.05). An ablation study proved each component's statistical significance and contribution to Fet-Net's performance. Fet-Net demonstrated robustness in classification accuracy even when noise was introduced to the images, outperforming eight of the 11 prominent architectures. Fet-Net's ability to automatically detect fetal orientation can profoundly decrease the time required for fetal MRI acquisition.

European Society of Paediatric Radiology Artificial Intelligence taskforce: a new taskforce for the digital age.

A new task force dedicated to artificial intelligence (AI) with respect to paediatric radiology was created in 2021 at the International Paediatric Radiology (IPR) meeting in Rome, Italy (a joint society meeting by the European Society of Pediatric Radiology [ESPR] and the Society for Pediatric Radiology [SPR]). The concept of a separate task force dedicated to AI was borne from an ESPR-led international survey of health care professionals' opinions, expectations and concerns regarding AI integration within children's imaging departments. In this survey, the majority (> 80%) of ESPR respondents supported the creation of a task force and helped define our key objectives. These include providing educational content about AI relevant for paediatric radiologists, brainstorming ideas for future projects and collaborating on AI-related studies with respect to collating data sets, de-identifying images and engaging in multi-case, multi-reader studies. This manuscript outlines the starting point of the ESPR AI task force and where we wish to go.

An Educational Graphical User Interface to Construct Convolutional Neural Networks for Teaching Artificial Intelligence in Radiology.

Deep learning techniques using convolutional neural networks (CNNs) have been successfully developed for various medical image analysis tasks. However, the skills to understand and develop deep learning models are not usually taught during radiology training, which constitutes a barrier for radiologists looking to integrate machine learning (ML) into their research or clinical practice. In this work, we developed and evaluated an educational graphical user interface (GUI) to construct CNNs for teaching deep learning concepts to radiology trainees. The GUI was developed in Python using the PyQt and PyTorch frameworks. The functionality of the GUI was demonstrated through a binary classification task on a dataset of MR images of the brain. The usability of the GUI was assessed through 45-min user testing sessions with 5 neuroradiologists and neuroradiology fellows, assessing mean task completion times, the System Usability Scale (SUS), and a qualitative questionnaire as metrics. Task completion times were compared against a ML expert who performed the same tasks. After a 20-min introduction to CNNs and a walkthrough of the GUI, users were able to perform all assigned tasks successfully. There was no significant difference in task completion time compared to a ML expert. The educational GUI achieved a score of 82.5 on the SUS, suggesting that the system is highly usable. Users indicated that the GUI seems useful as an educational tool to teach ML topics to radiology trainees. An educational GUI allows interactive teaching in ML that can be incorporated into radiology training.

Radiomic Features Based on MRI Predict Progression-Free Survival in Pediatric Diffuse Midline Glioma/Diffuse Intrinsic Pontine Glioma.

Purpose: Biopsy-based assessment of H3 K27 M status helps in predicting survival, but biopsy is usually limited to unusual presentations and clinical trials. We aimed to evaluate whether radiomics can serve as prognostic marker to stratify diffuse intrinsic pontine glioma (DIPG) subsets. Methods: In this retrospective study, diagnostic brain MRIs of children with DIPG were analyzed. Radiomic features were extracted from tumor segmentations and data were split into training/testing sets (80:20). A conditional survival forest model was applied to predict progression-free survival (PFS) using training data. The trained model was validated on the test data, and concordances were calculated for PFS. Experiments were repeated 100 times using randomized versions of the respective percentage of the training/test data. Results: A total of 89 patients were identified (48 females, 53.9%). Median age at time of diagnosis was 6.64 years (range: 1-16.9 years) and median PFS was 8 months (range: 1-84 months). Molecular data were available for 26 patients (29.2%) (1 wild type, 3 K27M-H3.1, 22 K27M-H3.3). Radiomic features of FLAIR and nonenhanced T1-weighted sequences were predictive of PFS. The best FLAIR radiomics model yielded a concordance of .87 [95% CI: .86-.88] at 4 months PFS. The best T1-weighted radiomics model yielded a concordance of .82 [95% CI: .8-.84] at 4 months PFS. The best combined FLAIR + T1-weighted radiomics model yielded a concordance of .74 [95% CI: .71-.77] at 3 months PFS. The predominant predictive radiomic feature matrix was gray-level size-zone. Conclusion: MRI-based radiomics may predict progression-free survival in pediatric diffuse midline glioma/diffuse intrinsic pontine glioma.

Improving the Segmentation of Pediatric Low-Grade Gliomas Through Multitask Learning.

Brain tumor segmentation is a critical task for tumor volumetric analyses and AI algorithms. However, it is a time-consuming process and requires neuroradiology expertise. While there has been extensive research focused on optimizing brain tumor segmentation in the adult population, studies on AI guided pediatric tumor segmentation are scarce. Furthermore, MRI signal characteristics of pediatric and adult brain tumors differ, necessitating the development of segmentation algorithms specifically designed for pediatric brain tumors. We developed a segmentation model trained on magnetic resonance imaging (MRI) of pediatric patients with low-grade gliomas (pLGGs) from The Hospital for Sick Children (Toronto, Ontario, Canada). The proposed model utilizes deep Multitask Learning (dMTL) by adding tumor's genetic alteration classifier as an auxiliary task to the main network, ultimately improving the accuracy of the segmentation results.