|
 
The Cancer Imaging Archive Updates : February 2026
|
|
Advancing Pediatric Neuro-Oncology: New Dataset Focuses on AI-Powered MRI Tumor Segmentation
 |
Pediatric central nervous system (CNS) tumors are the leading cause of cancer-related mortality in children. Pediatric high-grade gliomas, particularly diffuse midline gliomas (DMGs), have a dismal prognosis, with five-year survival rates below 20%. The Brain Tumor Segmentation in Pediatric Magnetic Resonance Imaging (BraTS-PEDs) dataset provides a comprehensive, multi-institutional, international resource focused on this disease.
It includes 457 pediatric patients with high-grade gliomas, primarily DMGs, aggregated from major pediatric neuro-oncology consortia and institutions, including the Children’s Brain Tumor Network (CBTN), DMG/DIPG Registry, and multiple academic centers. The dataset contains multiparametric structural MRI, expert tumor segmentations, imaging acquisition parameters, demographic variables, and clinical outcomes such as overall and progression-free survival. This dataset is unique in offering one of the first large-scale, multi-institutional, and standardized pediatric resources.
By integrating standardized imaging, annotations, and clinical data, BraTS-PEDs enables reproducible research and accelerates translation of imaging science into clinical impact. This dataset aims to promote innovation in algorithm development tailored to children, support the training of neuroradiologists in consistent tumor delineation, and enable outcome prediction. BraTS-PEDs can serve as a benchmark reference that will accelerate research and improve clinical decision-making for pediatric brain tumors.
|
|
Interested in publishing data on
The Cancer Imaging Archive?
Researchers interested in publishing new Image Collections or Analysis Result datasets must submit a proposal for review by the NCI TCIA Advisory Group. TCIA reviews proposals on a quarterly basis.
👉 Click here to learn about TCIA submission guidelines and to submit a proposal.
How to Submit a Proposal to TCIA
|
|
Featured Collections
Yale longitudinal dataset of brain metastases on MRI with associated clinical data
Brain metastases are associated with significant symptoms, which require patients to have frequent radiologic assessments that aid in evaluation, treatment response and disease progression. Magnetic resonance imaging (MRI) is a cornerstone in the management of central nervous system metastases, providing critical insights over time. The Yale longitudinal dataset of brain metastases on MRI with associated clinical data (Yale-Brain-Mets-Longitudinal) is a collection of 11,892 longitudinal brain MRI studies from 1,430 patients with clinically confirmed brain metastases.
The dataset includes T1-weighted pre-contrast, T1-weighted post-contrast, T2-weighted, and fluid-attenuated inversion recovery MRI sequence images in NIfTI format. Patient demographic information, scanner details, and image acquisition parameters are provided.
This dataset is a resource that can enable diverse research applications, from conventional radiologic investigations to state-of-the-art machine learning approaches, ultimately contributing to better patient outcomes and a more comprehensive understanding of brain metastases. The inclusion of both imaging and clinical data makes this dataset a valuable asset for researchers in oncology, neuroradiology, and data science. This dataset can also help facilitate the development of AI models to assist in the long-term management of patients with brain metastasis.
|
A Multi-Centric Anthropomorphic 3D CT Phantom-Based Benchmark Dataset for Harmonization
The Multi-Centric Anthropomorphic 3D CT Phantom-Based Benchmark Dataset for Harmonization (CT4Harmonization-Multicentric) collection introduces an open-source, anthropomorphic phantom-based dataset of CT scans for developing harmonization methods for deep learning-based models. The phantom mimics human anatomy, allowing repeated scans without radiation delivery to real patients and isolating scanner effects by removing inter- and intra-patient variations.
The dataset includes 268 image series from 13 scanners, 4 manufacturers, and 8 institutions, repeated 18-30 times at a 10 mGy dose using a harmonized protocol. An additional 1,378 image series were acquired with the same 13 scanners and harmonized protocol but including additional acquisition doses. The presented phantom scans consist of three compartments from thorax, liver and test patterns. The 3D-printed liver includes three types of abnormal regions of interest, including two cysts, a metastasis, and a hemangioma, with ground truth segmentation masks that could be used for classification and segmentation.
|
Pre and post treatment MRI and radiotherapy plans of patients with glioblastoma: the CFB-GBM cohort
Glioblastoma is the most frequent primary malignant brain tumor in adults. Treatment consists of surgery followed by Stupp protocol association radiotherapy and chemotherapy. However, median survival is still limited to about 15 months from time of diagnosis. Despite these therapeutic approaches, treatment efficacy remains heterogeneous, with some patients showing favorable response while others do not. MRI plays a key role in diagnosis and evaluating treatment response.
The Pre and post treatment MRI and radiotherapy plans of patients with glioblastoma: the CFB-GBM cohort (CFB-GBM) dataset looks to develop treatments for these cases. It includes 264 patients with glioblastoma involved in the clinical routine. For all patients, pre-treatment MRI (at t0) were acquired. There are also included CT scans and treatment imaging, such as radiotherapy doses and tumor delineations, when available. The pre-processing pipeline involved skull-stripping and co-registering the MRI, as well as defacing the CT scans. Patients clinical data, treatment data, and CT and MRI machines are provided when available.
Pre and post treatment MRI and radiotherapy plans of patients with glioblastoma: the CFB-GBM cohort
|
|
|
|
|
