Dr. Andreas Beck
HepaTux
Performanca data
MICCAI Grand ChallengeThe scoring systemFrom our results in the MICCAI Grand Challenge, which were consistent between the pre-competition test data and the on-site test, we derive the following conclusion about the available precision:The MICCAI Grand Challenge scoring system was set up to give a score of 75 for a second human observer and 100 for perfect agreement with the first reading. The first reading was done by an expert, the second reading by a trained student. So it is to be expected that a perfect system would score at around 75, maybe a bit higher to account for the difference in experience. Our performanceWe are thus very pleased to see our system perform at 78 points at average with no single total score below 70. These results seem acceptable, especially as the segmentation was done by a computer scientist with only rudimentary anatomical knowledge. |  Flash nicht gefunden. Live-Video nicht möglich. |