Abstract:
In this study, we present a novel approach to address the task of identifying nuclei in a diverse set of microscopy scan images containing slides from several distinct ca...View moreMetadata
Abstract:
In this study, we present a novel approach to address the task of identifying nuclei in a diverse set of microscopy scan images containing slides from several distinct cancerous tissue types. Our approach is based on YOLOv5 algorithm. We thoroughly discuss the crucial training specifics that contributed to our outcomes. Additionally, we propose an intuitive strategy for combining multiple segmentation results to further improve the accuracy of nucleus identification. By employing this lightweight model, we successfully achieved results better or close to the state-of-the-art performance.
Date of Conference: 05-08 December 2023
Date Added to IEEE Xplore: 18 January 2024
ISBN Information: