Computational methods for analyses of public H&E histopathological images: a systematic review

• Grant number: 23/18404-2
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: February 01, 2024
• End date: December 31, 2024
• Field of knowledge: Engineering - Biomedical Engineering
• Principal Investigator: Thaína Aparecida Azevedo Tosta
• Grantee: Anna Clara Medina Roissmann
• Host Institution: Instituto de Ciência e Tecnologia (ICT). Universidade Federal de São Paulo (UNIFESP). Campus São José dos Campos. São José dos Campos , SP, Brazil
• Associated research grant: 22/03020-1 - Normalization of H&E stain by autoencoders with analysis of ensemble learning for histological images, AP.PNGP.PI

Abstract

The increasing incidence of cancer cases promotes scientific investigation of these diseases in different areas of knowledge. The digitization of tissue samples stained with Hematoxylin-Eosin (H&E) allows the obtaining of the so-called histological images, used for the definition of diagnoses and, consequently, for the choice of effective treatments. With these images, computational systems are being developed to aid pathologists in defining cancer diagnoses. These systems can use pre-processing, segmentation, feature extraction, and classification steps. However, the availability of public images has been an obstacle already recognized in the related literature for new proposals of these computational systems. In that way, this project proposes to perform a systematic review for mapping and identifying public datasets already available and used by published studies. Besides, we aim to identify the techniques used for processing these images, which allow us to investigate advances already reached, identify issues and challenges still open, and define possible works to be made in the future.