Kvasir-Instruments and Polyp Segmentation Using UNet

Authors

  • Sumit Pandey Department of Medical Imaging and Intervention, Chang Gung Memorial Hospital
  • Arvind Keprate OsloMet University

DOI:

https://doi.org/10.5617/nmi.9130

Keywords:

UNet, segmentation, deep learning, polyp, instrumentation

Abstract

This paper aims to describe the methodology used to develop, fine-tune and analyze a UNet model for creating masks for two datasets: Polyp Segmentation and Instrument Segmentation, which are part of MedAI challenge. For training and validation, we have used the same methodology on both tasks and finally on the hidden testing dataset the model resulted in an accuracy of 0.9721, dice score of 0.7980 for the instrumentation task, and the accuracy of 0.5646 and a dice score of 0.4100 was achieved for the Polyp Segmentation.

Downloads

Published

2021-11-01 — Updated on 2021-12-09

Versions