Comprehensive exploration of UNet adaptations for improved COVID-19 CT segmentation

Bowen Ji

doi:10.1117/12.3004202

11 October 2023 Comprehensive exploration of UNet adaptations for improved COVID-19 CT segmentation

Bowen Ji

Proceedings Volume 12800, Sixth International Conference on Computer Information Science and Application Technology (CISAT 2023); 1280065 (2023) https://doi.org/10.1117/12.3004202
Event: 6th International Conference on Computer Information Science and Application Technology (CISAT 2023), 2023, Hangzhou, China

Abstract

The COVID-19 pandemic has led to a widespread need for accurate and efficient medical image analysis, particularly in the context of lung CT scans. In this paper, we present a comprehensive exploration of UNet adaptations for COVID-19 CT segmentation, focusing on the impact of different loss functions and modifications to the UNet's Convolutional Neural Network (CNN) architecture. Our experimental results demonstrate the efficacy of these approaches in improving segmentation performance. We provide a thorough comparison of various loss functions and UNet modifications, highlighting the most effective combinations for robust and accurate COVID-19 CT segmentation.

(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Bowen Ji "Comprehensive exploration of UNet adaptations for improved COVID-19 CT segmentation", Proc. SPIE 12800, Sixth International Conference on Computer Information Science and Application Technology (CISAT 2023), 1280065 (11 October 2023); https://doi.org/10.1117/12.3004202

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available