The Cancer Imaging Archive Continues to Evolve: A New Hybrid Infrastructure for Open Science
The Cancer Imaging Archive (TCIA) continues to evolve. Here’s what that means for researchers. To align with new NIH policies and strengthen integration with the NCI Cancer Research Data Commons (CRDC), TCIA is modernizing its infrastructure while keeping the user-friendly experience the research community relies on. The key updates highlighted below enhance security, improve scalability, and maintain the simplicity that makes TCIA a trusted resource for open science. Learn more about TCIA’s new hybrid infrastructure here.
New Collections
The University of California San Diego Longitudinal Vestibular Schwannoma Dataset
The University of California San Diego Longitudinal Vestibular Schwannoma Dataset (UCSD-VS-Longitudinal) is now available. This extensive resource features 570 longitudinal multimodal MRI exams from 191 patients with vestibular schwannoma. Offering two to four consecutive timepoints per patient, the collection uniquely pairs real-world, multi-vendor imaging data with detailed clinical histories. It covers conservative observation, surgical resection, and radiotherapy; complete with expert-reviewed, volumetric tumor segmentations on T1 post-contrast images. This dataset provides a critical foundation for researchers looking to track tumor evolution, assess treatment responses over time, and develop advanced automated segmentation and prognostic tools.
|
Segmentation of spinal multiple myeloma lesions in dual-energy CT
Researchers and AI developers in the medical imaging field have a valuable new resource with the release of the Segmentation of spinal multiple myeloma lesions in dual-energy CT (Spinal-Multiple-Myeloma-SEG ). This unique public collection addresses a critical shortage of detailed, lesion-level annotations necessary for validating detection algorithms in multiple myeloma. The dataset includes 72 whole-body dual-energy CT (DECT) scans from 67 patients, offering diverse image types such as virtual monoenergetic and calcium-suppressed images alongside precise manual segmentations of vertebrae and lytic bone lesions. By making this comprehensive imaging and clinical data openly accessible, this resource aims to accelerate the development of advanced tools for more accurate disease staging, prognosis, and treatment planning.
|
The Clinical Proteomic Tumor Analysis Consortium Stomach Adenocarcinoma
The Clinical Proteomic Tumor Analysis Consortium Stomach Adenocarcinoma (CPTAC-STAD) collection has recently been updated on TCIA. The collection features high-resolution histopathology and whole-slide tissue images, alongside clinical radiology scans such as CT and ultrasound. By making this dataset publicly available, the National Cancer Institute (NCI) aims to empower researchers to investigate stomach cancer phenotypes and integrate these imaging characteristics with corresponding proteomic, genomic, and clinical data, ultimately accelerating our understanding of the disease's molecular basis.
|
This collection provides data which could be used to integrate imaging features to expand on a recent study published in Cell Reports Medicine, titled “A 15-layer multi-omics analysis of gastric cancer ecotypes provides therapeutic insights”. Researchers integrated the physical characteristics found in these images with a the 15-layer atlas of genomic, proteomic, and microbiome data. The study identifies distinct gastric cancer ecotypes, offering new avenues for precision medicine and targeted therapies.
|
