Interested in a custom solution?
On Monday June 3rd, Ruth Wijma (24 years old, Master Student Artificial Intelligence at the University of Amsterdam) was proclaimed the winner of the ZiuZ Pill Challenge. The results of the algorithm developed by Ruth corresponded most with the predetermined ground truth. After his final presentation, Ruth was presented with the winner’s check of € 2,500. On behalf of everyone at ZiuZ, we want to congratulate Ruth with the achieved results!
Because both medicine bags and medication are changing continuously, ZiuZ must look critically at its own segmentation methods. Last year, ZiuZ issued the ZiuZ Pill Challenge with the aim of finding out what is currently the best and fastest segmentation method for medicine bags. Based on a training database with 1,743 images (581 medicine bags), participants were asked to develop an algorithm for the segmentation of medicines into medicine bags. Two weeks before the expiry of the Pill Challenge, a testing database with 642 images (321 medicine bags) was published. Based on the results of the developed algorithm on the testing database, we determined who could claim to be the winner of the Pill Challenge.
A total of 11 people signed up for the Pill Challenge, of which five actually made a submission. To assess these submissions, ZiuZ measured how well the results match the predetermined ground truth. Based on the Jaccard Index and the Sørensen-Dice coefficient, we ultimately determined the winner of the Pill Challenge. With a score of 0.90855 on the Jaccard index and 0.12182 on the Sørensen-Dice coefficient, the algorithm developed by Ruth Wijma attained the best results.
ZiuZ would like to thank all participants of the Pill Challenge for participating. In the future, we will be launching new Challenges, so keep an eye on our Challenge page!