I am using several U-Net variants for a medical image segmentation task. I get the following values for the performance measures including Dice, IOU, Area under receiver-operating characteristic (AUC) curves, and Area under Precision-Recall curves (AUPRC), otherwise called the average precision (AP) computed for varying IOU thresholds in the range [0.5:0.95] in intervals of 0.05. From the attached table, I could observe that Model-2 gave better values for the IOU and Dice metrics. I could understand that Dice coefficient gives more weightage for the TPs. However, Model - 1 gives superior values for the AUC, and AP@[0.5:0.95] metrics. What parameters need to be given higher importance in model selection under these circumstances?
See this link
https://www.researchgate.net/post/Dice_vs_IoU_score-which_one_is_most_important_in_semantic_segmentation
Kindly check https://www.jeremyjordan.me/evaluating-image-segmentation-models/amp/
- Badges
- Science topic
- Similar topics
- Engineering
- Computer Engineering
- Computer