|
 
The Cancer Imaging Archive Updates : May 2024
|
|
NCI Clinical Trials Network (NCTN) Datasets:
Expert-Generated Tumor Annotations
Annotations for A Randomized Phase III Study Comparing Carboplatin/Paclitaxel or Carboplatin/Paclitaxel/Bevacizumab With or Without Concurrent Cetuximab in Patients With Advanced Non-small Cell Lung Cancer (S0819-Tumor-Annotations) have been published. For all 1,296 patients, scans were reviewed to annotate the clinically relevant time points and sequences/series. PERCIST criteria was followed for PET imaging and RECIST 1.1 was otherwise followed for any MR and CT imaging. Lesion volumes were calculated as well. These expert annotations further increase the value of the imaging data from this trial from the NCI Clinical Trial Network dataset, which can be reviewed along with the rich supporting clinical data available.
|
|
|
The recent joint statement by STM, DataCite, and Crossref has set new standards for citing datasets. At The Cancer Imaging Archive, we’re proud to be part of this change. We’re taking care of items #1 and #3 in their list by assigning persistent identifiers (DOIs) and enabling FAIR sharing of research outputs. We encourage our users to follow item #2 by citing our datasets using the full dataset citation and Digital Object Identifiers (DOIs) that we assign in the reference section of your manuscripts. Let’s make research data more findable and reusable together! 🌐🔬 |
|
New and Updated Collections
“Standard and Delayed Contrast-Enhanced MRI of Malignant and Benign Breast Lesions with Histological and Clinical Supporting Data (Advanced-MRI-Breast-Lesions)” is now available. It contains 1.5T MRIs for 632 subjects. There is supporting data for 200 patients including clinical notes, radiology reports and pathology reports, with patient age, lesion location, malignant/benign outcome, pathology and grade (if applicable), receptor status and KI67 (if applicable). All enhancing/suspicious positions that were mentioned in the radiological reports were delineated manually by a certified radiologist with 23 years of experience. This dataset is unique in that it contains the standard protocol series (T1-weighted DCE & T2-weighted MRI with/without fat suppression) plus the delayed contrast T1-images acquired 20-28 minutes after contrast and the calculated treatment response assessment maps (TRAMs).
|
“A Curated Benchmark Dataset for Ultrasound Based Breast Lesion Analysis (Breast-Lesions-USG)” has been published. It consists of 256 patients with patient, image, and tumor-level labels with all cases confirmed by follow-up care or biopsy result. Each scan was manually annotated and labeled by a radiologist experienced in breast ultrasound examination: each tumor was identified in the image via a freehand annotation and labeled according to BIRADS features. The tumor histopathological classification is stated for patients who underwent a biopsy.
|
“Computed Tomography Images from Large Head and Neck Cohort (RADCURE)” has been updated to add 3,337 contours (primary gross tumor volume (GTVp), nodal gross tumor volume (GTVn), and 19 OARs). These new data points increase the value of this dataset for a variety of quantitative image analysis research initiatives including the application of machine learning or artificial intelligence methods to expedite routine clinical practices, discovery of new non-invasive biomarkers, or development of prognostic models.
|
|
|
TCIA has a new free-text search function that makes it easier to find datasets with the characteristics you are looking for. You can find the search icon in the top right corner of each page, and also on the Browse Collection and Browse Analysis Result pages. |
|
|
|
|
