This is the dataset we chose we conducting the experiment for the project.
Each radiologist marked lesions they identified as non-nodule, nodule < 3 mm, and nodules >= 3 mms
annotations.csv: CSV file that contains the annotations used as reference standard for the 'nodule detection' track.
These Annotations have X,Y,Z, and the diameter of the nodule along with the unique id of the CT scan
Below 3D is the lung illustration from the dataset and a sample of the CT scan with the largest nodule ( 28mm):
This is the distribution of the diameters of the nodules present in the dataset
Segmenting the nodules for a better model training and outcome, with nodules as class '1' and without nodules as the class '0'