Automatic Nodule Detection 2009¶
Computer-aided detection (CAD) of nodules in chest computed tomography (CT) scans has attracted massive interest in the last eight years. There are now multiple commercial systems on the market and a large number of papers have been published that describe systems developed in academia. ANODE09 is an initiative to compare systems that perform automatic detection of pulmonary nodules in chest CT scans on a single common database, with a single evaluation protocol.
How does it work?¶
On this website, teams can register to participate in the study. After registration, they can download an example dataset of 5 annotated scans and a test set of 50 scans without annotations. Results of CAD systems on those test scans, consisting of a list of locations in the scans and a degree of suspicion that this location is a nodule, could be submitted. The submitted results were processed and are published on the results page.
A special session devoted to the ANODE09 study was held at the 2009 CAD Conference of SPIE Medical Imaging and a joint paper with the first results of the study was published in Medical Image Analysis:
B. van Ginneken, S.G. Armato, B. de Hoop, S. van de Vorst, T. Duindam, M. Niemeijer, K. Murphy, A.M.R. Schilham, A. Retico, M.E. Fantacci, N. Camarlinghi, F. Bagagli, I. Gori, T. Hara, H. Fujita, G. Gargano, R. Belloti, F.D. Carlo, R. Megna, S. Tangaro, L. Bolanos, P. Cerello, S.C. Cheran, E.L. Torres and M. Prokop. "Comparing and combining algorithms for computer-aided detection of pulmonary nodules in computed tomography scans: the ANODE09 study", Medical Image Analysis 2010;14:707-722.
The paper can be downloaded on the Medical Image Analysis web site by following this link. If you have no access to this journal, you can request a reprint by following this link.
ANODE09 is currently no longer open for new submissions. You can still sign up and download the data. For those interested in participating in a nodule detection challenge, we refer to the newer LUNA16 challenge that is based on a much larger data set.