This challenge was completed in February 2015. An article describing the lessons learned can be found at http://medicalimaging.spiedigitallibrary.org/article.aspx?articleid=2375472. The corresponding scientific manuscript is available at http://dx.doi.org/10.1117/1.JMI.3.4.044506. The full data set, including the answers to the test data, can be found on the SPIE-AAPM Lung CT Challenge page. |
As part of the 2015 SPIE Medical Imaging Conference, SPIE – with the support of American Association of Physicists in Medicine (AAPM) and the National Cancer Institute (NCI) – will conduct a “Grand Challenge” on quantitative image analysis methods for the diagnostic classification of malignant and benign lung nodules. The LUNGx Challenge will provide a unique opportunity for participants to compare their algorithms to those of others from academia, industry, and government in a structured, direct way using the same data sets. |
Disclaimer: Anyone wishing to use the downloaded images for presentation or publication purposes outside of the LUNGx Challenge should acknowledge the SPIE, the NCI, the AAPM, and The University of Chicago: “Data used in this research were obtained from The Cancer Imaging Archive (TCIA) sponsored by the SPIE, NCI/NIH, AAPM and The University of Chicago.” |
The 10 calibration cases released in November were made available specifically so that participants could become familiar with the image acquisition parameters and DICOM files of the one institution that is supplying the clinical cases for the LUNGx Challenge (The University of Chicago). The calibration cases were intended to orient participants to relevant technical issues surrounding the compatibility of the Challenge cases with participants’ algorithms and were not intended for algorithm development or training.
The 60 CT scans that comprise the LUNGx Challenge test set are now available. Most of these cases are contrast-enhanced scans. The location of each nodule for analysis is specified in the accompanying Excel spreadsheet that includes case name, the number of nodules for analysis in the scan (either 1 or 2), and the coordinates of each nodule’s approximate centroid. Important reference and explanatory information appears at the bottom of this spreadsheet. The nodules contained in the test set cases have been determined by a radiologist to be either primary lung cancer or benign based on pathologic assessment and/or follow-up imaging examinations.
A Google Group mailing list has been established as the primary mechanism through which news, updates, and answers to questions will be communicated by the organizers of the Challenge. It is essential that all participants sign up to receive these emails at: https://groups.google.com/forum/#!forum/lungx-challenge. This URL may also be used to view the archive of prior announcements. Questions about the Challenge should be directed to Sam Armato (s-armato@uchicago.edu) or Karen Drukker (kdrukker@uchicago.edu); please note that answers to any questions will be distributed only through the Google Group mailing list.
No more than two sets of results obtained from distinct methods may be submitted by a single group along with a description of the methods used. By submitting results for the LUNGx Challenge, participants agree that their results (along with a very brief description of their method) may be presented as part of publications prepared about the LUNGx Challenge by the organizers.
Disclaimer: Anyone wishing to use the downloaded images for presentation or publication purposes outside of the LUNGx Challenge should acknowledge the SPIE, the NCI, the AAPM, and The University of Chicago: “Data used in this research were obtained from The Cancer Imaging Archive (TCIA) sponsored by the SPIE, NCI/NIH, AAPM and The University of Chicago.” |
In order to facilitate performance assessment of your method for the test cases in the LUNGx Challenge (and to compare your performance to that of the methods of other participants), please follow these guidelines:
Feng Li, University of Chicago (feng@uchicago.edu)