Empowering Data Sharing in Neuroscience: A Deep Learning Deidentification Method for Pediatric Brain MRIs

American Journal of Neuroradiology · 2024 · 4 citations

• Medicine
• Neuroscience
• Cognitive science

BACKGROUND AND PURPOSE: Privacy concerns, such as identifiable facial features within brain scans, have hindered the availability of pediatric neuroimaging data sets for research. Consequently, pediatric neuroscience research lags adult counterparts, particularly in rare disease and under-represented populations. The removal of face regions (image defacing) can mitigate this; however, existing defacing tools often fail with pediatric cases and diverse image types, leaving a critical gap in data accessibility. Given recent National Institutes of Health data sharing mandates, novel solutions are a critical need. MATERIALS AND METHODS: To develop an artificial intelligence (AI)-powered tool for automatic defacing of pediatric brain MRIs, deep learning methodologies (nnU-Net) were used by using a large, diverse multi-institutional data set of clinical radiology images. This included multiparametric MRIs (T1-weighted [T1W], T1W-contrast-enhanced, T2-weighted [T2W], T2W-FLAIR) with 976 total images from 208 patients with brain tumor (Children's Brain Tumor Network, CBTN) and 36 clinical control patients (Scans with Limited Imaging Pathology, SLIP) ranging in age from 7 days to 21 years old. RESULTS: < .0001). CONCLUSIONS: The defacing model demonstrates efficacy in removing facial regions across multiple MRI types and exhibits minimal impact on downstream research usage. A software package with the trained model is freely provided for wider use and further development (pediatric-auto-defacer; https://github.com/d3b-center/pediatric-auto-defacer-public). By offering a solution tailored to pediatric cases and multiple MRI sequences, this defacing tool will expedite research efforts and promote broader adoption of data sharing practices within the neuroscience community.

Automated tumor segmentation and brain tissue extraction from multiparametric MRI of pediatric brain tumors: A multi-institutional study

Neuro-Oncology Advances · 2023 · 50 citations

• Computer Science
• Artificial Intelligence
• Medicine

Background: Brain tumors are the most common solid tumors and the leading cause of cancer-related death among all childhood cancers. Tumor segmentation is essential in surgical and treatment planning, and response assessment and monitoring. However, manual segmentation is time-consuming and has high interoperator variability. We present a multi-institutional deep learning-based method for automated brain extraction and segmentation of pediatric brain tumors based on multi-parametric MRI scans. Methods: = 21) subsets. DeepMedic, a three-dimensional convolutional neural network, was trained and the model parameters were tuned. Finally, the network was evaluated on the withheld internal and external test cohorts. Results: Dice similarity score (median ± SD) was 0.91 ± 0.10/0.88 ± 0.16 for the whole tumor, 0.73 ± 0.27/0.84 ± 0.29 for ET, 0.79 ± 19/0.74 ± 0.27 for union of all non-enhancing components (i.e., NET, CC, ED), and 0.98 ± 0.02 for brain tissue in both internal/external test sets. Conclusions: Our proposed automated brain extraction and tumor subregion segmentation models demonstrated accurate performance on segmentation of the brain tissue and whole tumor regions in pediatric brain tumors and can facilitate detection of abnormal regions for further clinical measurements.

Imaging of pediatric brain tumors: A COG Diagnostic Imaging Committee/SPR Oncology Committee/ASPNR White Paper

Pediatric Blood & Cancer · 2022 · 26 citations

• Medicine
• Medical physics
• Radiology

Tumors of the central nervous system are the most common solid malignancies in children and the most common cause of pediatric cancer-related mortality. Imaging plays a central role in diagnosis, staging, treatment planning, and response assessment of pediatric brain tumors. However, the substantial variability in brain tumor imaging protocols across institutions leads to variability in patient risk stratification and treatment decisions, and complicates comparisons of clinical trial results. This White Paper provides consensus-based imaging recommendations for evaluating pediatric patients with primary brain tumors. The proposed brain magnetic resonance imaging protocol recommendations balance advancements in imaging techniques with the practicality of deployment across most imaging centers.

Invasive brain tissue oxygen and intracranial pressure (ICP) monitoring versus ICP-only monitoring in pediatric severe traumatic brain injury

Journal of Neurosurgery Pediatrics · 2022 · 22 citations

• Medicine
• Anesthesia
• Pediatrics

OBJECTIVE: Severe traumatic brain injury (TBI) is a leading cause of disability and death in the pediatric population. While intracranial pressure (ICP) monitoring is the gold standard in acute neurocritical care following pediatric severe TBI, brain tissue oxygen tension (PbtO2) monitoring may also help limit secondary brain injury and improve outcomes. The authors hypothesized that pediatric patients with severe TBI and ICP + PbtO2 monitoring and treatment would have better outcomes than those who underwent ICP-only monitoring and treatment. METHODS: Patients ≤ 18 years of age with severe TBI who received ICP ± PbtO2 monitoring at a quaternary children's hospital between 1998 and 2021 were retrospectively reviewed. The relationships between conventional measurements of TBI were evaluated, i.e., ICP, cerebral perfusion pressure (CPP), and PbtO2. Differences were analyzed between patients with ICP + PbtO2 versus ICP-only monitoring on hospital and pediatric intensive care unit (PICU) length of stay (LOS), length of intubation, Pediatric Intensity Level of Therapy scale score, and functional outcome using the Glasgow Outcome Score-Extended (GOS-E) scale at 6 months postinjury. RESULTS: Forty-nine patients, including 19 with ICP + PbtO2 and 30 with ICP only, were analyzed. There was a weak negative association between ICP and PbtO2 (β = -0.04). Conversely, there was a strong positive correlation between CPP ≥ 40 mm Hg and PbtO2 ≥ 15 and ≥ 20 mm Hg (β = 0.30 and β = 0.29, p < 0.001, respectively). An increased number of events of cerebral PbtO2 < 15 mm Hg or < 20 mm Hg were associated with longer hospital (p = 0.01 and p = 0.022, respectively) and PICU (p = 0.015 and p = 0.007, respectively) LOS, increased duration of mechanical ventilation (p = 0.015 when PbtO2 < 15 mm Hg), and an unfavorable 6-month GOS-E score (p = 0.045 and p = 0.022, respectively). An increased number of intracranial hypertension episodes (ICP ≥ 20 mm Hg) were associated with longer hospital (p = 0.007) and PICU (p < 0.001) LOS and longer duration of mechanical ventilation (p < 0.001). Lower minimum hourly and average daily ICP values predicted favorable GOS-E scores (p < 0.001 for both). Patients with ICP + PbtO2 monitoring experienced longer PICU LOS (p = 0.018) compared to patients with ICP-only monitoring, with no significant GOS-E score difference between groups (p = 0.733). CONCLUSIONS: An increased number of cerebral hypoxic episodes and an increased number of intracranial hypertension episodes resulted in longer hospital LOS and longer duration of mechanical ventilator support. An increased number of cerebral hypoxic episodes also correlated with less favorable functional outcomes. In contrast, lower minimum hourly and average daily ICP values, but not the number of intracranial hypertension episodes, were associated with more favorable functional outcomes. There was a weak correlation between ICP and PbtO2, supporting the importance of multimodal invasive neuromonitoring in pediatric severe TBI.

The children's brain tumor network (CBTN) - Accelerating research in pediatric central nervous system tumors through collaboration and open science

Neoplasia · 2022 · 86 citations

• Medicine
• Bioinformatics
• Oncology

Pediatric brain tumors are the leading cause of cancer-related death in children in the United States and contribute a disproportionate number of potential years of life lost compared to adult cancers. Moreover, survivors frequently suffer long-term side effects, including secondary cancers. The Children's Brain Tumor Network (CBTN) is a multi-institutional international clinical research consortium created to advance therapeutic development through the collection and rapid distribution of biospecimens and data via open-science research platforms for real-time access and use by the global research community. The CBTN's 32 member institutions utilize a shared regulatory governance architecture at the Children's Hospital of Philadelphia to accelerate and maximize the use of biospecimens and data. As of August 2022, CBTN has enrolled over 4700 subjects, over 1500 parents, and collected over 65,000 biospecimen aliquots for research. Additionally, over 80 preclinical models have been developed from collected tumors. Multi-omic data for over 1000 tumors and germline material are currently available with data generation for > 5000 samples underway. To our knowledge, CBTN provides the largest open-access pediatric brain tumor multi-omic dataset annotated with longitudinal clinical and outcome data, imaging, associated biospecimens, child-parent genomic pedigrees, and in vivo and in vitro preclinical models. Empowered by NIH-supported platforms such as the Kids First Data Resource and the Childhood Cancer Data Initiative, the CBTN continues to expand the resources needed for scientists to accelerate translational impact for improved outcomes and quality of life for children with brain and spinal cord tumors.

The Human Tumor Atlas Network: Charting Tumor Transitions across Space and Time at Single-Cell Resolution

Cell · 2020 · 581 citations

• Biology
• Computational biology
• Bioinformatics

Crucial transitions in cancer-including tumor initiation, local expansion, metastasis, and therapeutic resistance-involve complex interactions between cells within the dynamic tumor ecosystem. Transformative single-cell genomics technologies and spatial multiplex in situ methods now provide an opportunity to interrogate this complexity at unprecedented resolution. The Human Tumor Atlas Network (HTAN), part of the National Cancer Institute (NCI) Cancer Moonshot Initiative, will establish a clinical, experimental, computational, and organizational framework to generate informative and accessible three-dimensional atlases of cancer transitions for a diverse set of tumor types. This effort complements both ongoing efforts to map healthy organs and previous large-scale cancer genomics approaches focused on bulk sequencing at a single point in time. Generating single-cell, multiparametric, longitudinal atlases and integrating them with clinical outcomes should help identify novel predictive biomarkers and features as well as therapeutically relevant cell types, cell states, and cellular interactions across transitions. The resulting tumor atlases should have a profound impact on our understanding of cancer biology and have the potential to improve cancer detection, prevention, and therapeutic discovery for better precision-medicine treatments of cancer patients and those at risk for cancer.

Suppurative Intracranial Complications of Pediatric Sinusitis: A Single-Center Experience

Journal of the Pediatric Infectious Diseases Society · 2020 · 44 citations

• Medicine
• Pediatrics
• Intensive care medicine

BACKGROUND: Suppurative intracranial complications of sinusitis are rare events in children and can lead to harmful neurologic sequelae and significant morbidity. We sought to review the presentation and management of patients admitted at our hospital with these conditions. METHODS: This was a retrospective study of pediatric patients admitted to a quaternary children's hospital from 2007 to 2019 for operative management of sinusitis with intracranial extension. Clinical characteristics, including surgical and microbiological data, were collected and analyzed. RESULTS: Fifty-four patients were included; the median age was 11.0 years, and there was a male predominance. Eighty-nine percent of patients had prior healthcare visits for the current episode of sinusitis; 46% of patients had an abnormal neurologic exam on admission. Epidural abscess and subdural empyema were the most common complications, and subdural empyema was associated with repeat surgical intervention. The dominant pathogens were Streptococcus anginosus group organisms (74%). The majority of patients completed treatment parenterally, with a median duration of therapy of 35 days. Neurological sequelae, including epilepsy or ongoing focal deficits, occurred in 22% of patients. History of seizure or an abnormal neurological exam at admission were associated with neurological sequelae. CONCLUSIONS: Clinicians should consider intracranial complications of sinusitis in patients with symptoms of sinusitis for >1 week. Patients should undergo urgent neuroimaging, as neurosurgical intervention is essential for these patients. Subdural empyema was associated with repeat neurosurgical intervention. Neurological sequelae occurred in 22% of patients, and new onset seizure or an abnormal neurological exam at admission were associated with neurological sequelae.

Integrated Proteogenomic Characterization across Major Histological Types of Pediatric Brain Cancer

Cell · 2020 · 337 citations

• Biology
• Computational biology
• Genetics