COMPUTATIONAL CHALLENGES ON SPINE IMAGING
As part of the CSI 2014 Workshop, we are organizing two computational challenges. Researchers are invited to participate and submit their results to the workshop together with a short description of their method.
How to Participate?
Everyone working on related problems and algorithms is invited to join the challenges. The training data for both challenges can be obtained from SpineWeb. It is expected that participants of the challenges also join the MICCAI workshop and present their results. Participants who evaluate previously published methods are encouraged to submit a short paper including a brief method description and their challenge results.
Remote Evaluation
For the localization and identification challenge results on training and testing data will be evaluated remotely via an online evaluation system. The system is now open for submissions.
Challenge 1: Spine and Vertebrae Segmentation
Segmentation is an essential step for many computational spine imaging tasks. Clinical datasets raise many difficulties for automatic methods and ground truth is often scarce. To facilitate and foster the research on this topic, we provide a database of 10 spine CT scans along with manual segmentation of the thoracic and lumbar vertebrae. The data set can be used for development, training and testing of spine segmentation algorithms. More data sets will be provided for the evaluation phase of the challenge.
Challenge 2: Vertebrae Localization and Identification (Testing data now available!)
Accurate localization and identification of vertebrae in spinal CT imaging is important for many clinical tasks such as diagnosis, surgical planning, and post-operative assessment. Clinical datasets raise many difficulties for automatic methods. These arise from the frequent presence of abnormal spine curvature, small field of view, and image artifacts caused by surgical implants. To facilitate the advance of research on this topic, we provide a database of 242 annotated spine CT scans. These scans can be used for development, training and evaluation. Additionally, 60 datasets are now provided for the testing phase of the challenge for which no manual annotations will be made available. Results on testing data need to be submitted to the online evaluation system.
Awards
The winners of the challenges will receive a Microsoft Azure Research Award. Each award provides 170,000 hours of Azure compute and 20TB of data and other services. The awards are sponsored by Microsoft Research Connections.
As part of the CSI 2014 Workshop, we are organizing two computational challenges. Researchers are invited to participate and submit their results to the workshop together with a short description of their method.
How to Participate?
Everyone working on related problems and algorithms is invited to join the challenges. The training data for both challenges can be obtained from SpineWeb. It is expected that participants of the challenges also join the MICCAI workshop and present their results. Participants who evaluate previously published methods are encouraged to submit a short paper including a brief method description and their challenge results.
Remote Evaluation
For the localization and identification challenge results on training and testing data will be evaluated remotely via an online evaluation system. The system is now open for submissions.
Challenge 1: Spine and Vertebrae Segmentation
Segmentation is an essential step for many computational spine imaging tasks. Clinical datasets raise many difficulties for automatic methods and ground truth is often scarce. To facilitate and foster the research on this topic, we provide a database of 10 spine CT scans along with manual segmentation of the thoracic and lumbar vertebrae. The data set can be used for development, training and testing of spine segmentation algorithms. More data sets will be provided for the evaluation phase of the challenge.
Challenge 2: Vertebrae Localization and Identification (Testing data now available!)
Accurate localization and identification of vertebrae in spinal CT imaging is important for many clinical tasks such as diagnosis, surgical planning, and post-operative assessment. Clinical datasets raise many difficulties for automatic methods. These arise from the frequent presence of abnormal spine curvature, small field of view, and image artifacts caused by surgical implants. To facilitate the advance of research on this topic, we provide a database of 242 annotated spine CT scans. These scans can be used for development, training and evaluation. Additionally, 60 datasets are now provided for the testing phase of the challenge for which no manual annotations will be made available. Results on testing data need to be submitted to the online evaluation system.
- Download: Training data.
- Online evaluation system for training results
- Download: Testing data.
- Online evaluation system for testing results
Awards
The winners of the challenges will receive a Microsoft Azure Research Award. Each award provides 170,000 hours of Azure compute and 20TB of data and other services. The awards are sponsored by Microsoft Research Connections.