The‌ ‌Cancer‌ ‌Imaging‌ ‌Archive‌ ‌Updates:‌ March 2023

[The Cancer Imaging Program]

The‌ ‌Cancer‌ ‌Imaging‌ ‌Archive‌ ‌Updates:‌ March 2023 New Collections The "Segmentation and Classification of Grade I and II Meningiomas from Magnetic Resonance Imaging: An Open Annotated Dataset (Meningioma-SEG-CLASS)" is now available as a restricted access collection. The study includes 96 consecutive treatment naïve patients with intracranial meningiomas treated with surgical resection. All patients had pre-operative T1, T1-CE, and T2-FLAIR MR images with subsequent subtotal or gross total resection of pathologically confirmed grade I or grade II meningiomas. Meningioma labels on T1-CE and T2-FLAIR images are provided as well as contours on each MRI of the hyperintense T1-contrast enhancing tumor and hyperintense T2-FLAIR and tumor. Video tutorials are available to help users learn to navigate TCIA and to take advantage of all its features. AREN0532-Tumor-Annotations has been published. This TCIA Analysis Data Set augments the previously published TCIA collection “Vincristine, Dactinomycin, and Doxorubicin With or Without Radiation Therapy or Observation Only in Treating Younger Patients Who Are Undergoing Surgery for Newly Diagnosed Stage I, Stage II, or Stage III Wilms' Tumor (AREN0532)“.   For each patient, every DICOM Study and DICOM Series was reviewed to identify and annotate the clinically relevant time points and sequences. This new collection contains 3D segmentations and seed points of tumors on the CT and MRI images with labels and metadata. The selected sequences inform whether there is residual or recurrent tumor and assess response to therapy. There is also a Jupyter Notebook detailing how to access the data from the TCIA API using Python. In order to facilitate algorithmic analysis of the imaging data, NCI is supporting the addition of 3D segmentations and seed points to identify tumor locations in clinical trials and other selected TCIA collections.  The resulting “analysis datasets” are being made freely available on TCIA, along with supporting documentation and sample code (e.g. Jupyter Notebooks) that will facilitate streamlined analysis on widely used data science platforms. It is hoped that radiology and non-radiology researchers will leverage these annotations to develop predictive radiomic tools, and to develop quantitative imaging software and machine learning models for tumor segmentation, detection and total burden estimations to assess response assessment. A list of the currently available segmentation datasets can be found here. Imaging data from "Sorafenib Tosylate in Treating Patients With Desmoid Tumors or Aggressive Fibromatosis (A091105)" has been published on TCIA. This collection contains CT and MR imaging and supporting clinical data for 83 subjects. This clinical trial was sponsored by NCI and performed by the Alliance for Clinical Trials in Oncology under study number A091105. This randomized phase III trial compares the effects, good and/or bad, of sorafenib tosylate in treating patients with desmoid tumors or aggressive fibromatosis.  Sorafenib tosylate may stop the growth of tumor cells by blocking some of the proteins needed for cell growth. Collaborations Researchers from Frederick National Laboratory for Cancer Research and Kitware recently collaborated to create a series of Jupyter notebooks that demonstrate common workflows to programmatically access and visualize TCIA data. These notebooks use Amazon Sagemaker Studio Lab to allow researchers to run the notebooks without the need to set up their own local Jupyter development environment, enabling quick exploration of new ideas and simple integration of your work into interactive presentations, workshops and tutorials at conferences. The  Histopathology Portal allows users to filter and download the TCIA pathology datasets.   You can also visualize slides in your browser before downloading! Stay Connected with The Cancer Imaging Archive  Manage Subscriptions | Help The Cancer imaging Archive is funded in part by Frederick Natational Lab for Cancer Research.

 

This email was sent to tcia-anno...@googlegroups.com using GovDelivery Communications Cloud on behalf of: National Cancer Institute · BG 9609 MSC 9760 · 9609 Medical Center Drive · Bethesda, MD 20892