Introduction
Project B‐TIC‐324‐UGR20 funded by
The research project Development and evaluation of methods for deblurring and color normalization of histological images (Desarrollo y evaluación de métodos para la eliminación de emborronamiento y normalización de color de imágenes histológicas) is funded by FEDER/Junta de Andalucía - Consejería de Transformación Económica, Industria, Conocimiento y Universidades from 2021 to 2022.
The Development and evaluation of methods for deblurring and color normalization of histological images research team consists of five doctors from the Visual Information Processing Group (VIP) at Universidad de Granada and two experts pathologist from the Anatomic Pathology Unit from Hospital Universitario Clínico San Cecilio (HUSC) and Hospital Universitario Virgen de las Nieves (HUVN), two undergraduates with extensive experience in the project topic and a doctor from the Northwestern University (Evanston, Illinois, USA).
This page will provide information on the project results and publications.
Summary
The use of Whole-Slide-Images (WSI) is undoubtedly one of the major technical advances in computational pathology. However, the acquisition process of WSIs introduces degradations that need to be eliminated. Currently, by ocular inspection, about 5% of the images need to be re-scanned mainly due to blurring or staining quality problems. This leads to an increase in health and human costs and results in a delay in diagnosis. In addition, the lack of objective criteria to detect the presence of artifacts in the images leads to incorrect diagnoses due to misinterpretation of the image.
Color differences in WSI
In this project, a multidisciplinary team of scientists, engineers and medical doctors will address the development and evaluation, qualitative and quantitative, of methods for the detection and elimination of blurring and color normalization of histological images. We will develop inverse models for higher quality imaging, which will facilitate the work of pathologists and automatic diagnostic support algorithms. The analysis and diagnosis process will be accelerated, decreasing its cost and increasing its accuracy. The results obtained will be published in high impact index journals and prestigious conferences. The software developed will be transferred to the company.
Objectives
The objective of this project is to enhance the histological images, detecting and eliminating any blurring that may exist and normalizing the color of the images, in order to increase their quality and achieve a better diagnosis, either by means of automatic algorithms or by the team of pathologists.
This general objective is divided into the following specific objectives:
SO1. Build and compile databases of WSIs that allow training and validation of the algorithms.
SO2. Develop methods for detection and removal of smearing to improve the quality of the images.
SO3. Develop methods for normalizing the color of WSIs to make the system robust to inter- and intra-hospital variations.
SO4. Validate the developed algorithms by qualitative and quantitative methods, using reference WSIs and on diagnostic analysis systems.