...
For some of the images, the laterality stored in the DICOM header and/or image orientation are incorrect. The reference standard "truth" boxes are defined with respect to the corrected image orientation in these instances. Therefore, it is crucial to provide your results for images in the correct image orientation. Python functions for loading image data from a DICOM file into 3D array of pixel values in the correct orientation and for displaying "truth" boxes (if any) are on GitHub. Please see the readme file there for instructions.
DBTex Lesion Detection Challenge Predictions
The DBTex lesion detection challenge tasked participating teams with detecting lesions in the BCS-DBT test set. The challenge had two phases: DBTex1 and DBTex2. Here we provide the BCS-DBT lesion predictions made by all participating teams for both phases, for both the BCS-DBT test and validation sets, as “team_predictions_bothphases.zip”. Please see here under “Output format for the DBTex2 Challenge test set results” for a description of how these results are formatted. Finally, when comparing lesion bounding box predictions to the image data, be sure to load the images correctly according to the above “Required Preprocessing of DBT Images”.
...