See you in two years!
11-16 April 2026
Conference 13006 > Paper 13006-66
Paper 13006-66

Analysing histology hyperspectral images: does tissue thickness matter?

11 April 2024 • 16:30 - 16:50 CEST | Etoile B, Niveau/Level 1

Abstract

Cancer is one of the worldwide leading causes of death. They are diagnosed and graded histologically using biopsies, sliced at 2-5 μm, and stained with Haematoxylin and Eosin (H&E). Hyperspectral (HS) Imaging (HSI) has been arising to improve the spectral information given by RGB (Red-Green-Blue) cameras. This work aims to characterize the influence of tissue thickness on the spectral information. Captures were taken over breast tissue samples sectioned at 2 and 3 μm. They were pre-processed to reduce noise and normalize the data. Then, based on H&E absorption spectra peaks, HS images could be segmented into stroma (eosin-stained basic structures), nucleus (haematoxylin-stained acidic structures), and background (non-stained structures). Concluding, results show that 3 μm slides proportionate a bigger count of cells and a higher spectral contrast than 2 μm slides. However, further research is needed to study other thickness values, including more samples.

Presenter

Javier Santana-Nunez
Instituto Univ. de Microelectrónica Aplicada, Univ. de Las Palmas de Gran Canaria (Spain)
Javier Santana Nunez, graduated in 2023 with a bachelor’s degree in Industrial Electronics and Automation Engineering from the University of Las Palmas de Gran Canaria (ULPGC), Spain. He developed his bachelor thesis at the Research Institute for Applied Microelectronics (IUMA) of the ULPGC, focusing on designing and evaluating hyperspectral image registration techniques. Currently, he is working at IUMA as a researcher and starting his PhD Thesis. In parallel, he is coursing a Master of Science degree in Intelligent Systems and Numerical Applications Applied to Engineering, focusing on developing intelligent systems, such as neural networks, computer vision, and scientific research skills. His research interests are related to microscopic hyperspectral imaging applied to biological samples analysis, especially for histopathology applications.
Presenter/Author
Javier Santana-Nunez
Instituto Univ. de Microelectrónica Aplicada, Univ. de Las Palmas de Gran Canaria (Spain)
Author
Instituto Univ. de Microelectrónica Aplicada, Univ. de Las Palmas de Gran Canaria (Spain)
Author
Instituto Univ. de Microelectrónica Aplicada, Univ. de Las Palmas de Gran Canaria (Spain), Fundación Canaria Instituto de Investigación Sanitaria de Canarias (Spain)
Author
Nofima (Norway), Instituto Univ. de Microelectrónica Aplicada, Univ. de Las Palmas de Gran Canaria (Spain)
Author
Hospital de Tortosa Verge de la Cinta, Institut Català de la Salut, Institut d'Investigació Sanitària Pere Virgili (Spain), Univ. Rovira i Virgili (Spain)
Author
Univ. Rovira i Virgili (Spain), Hospital de Tortosa Verge de la Cinta, Institut Català de la Salut, Institut d'Investigació Sanitària Pere Virgili (Spain)
Author
Hospital de Tortosa Verge de la Cinta, Institut Català de la Salut, Institut d'Investigació Sanitària Pere Virgili (Spain)
Author
Hospital de Tortosa Verge de la Cinta, Institut Català de la Salut, Institut d'Investigació Sanitària Pere Virgili (Spain), Univ. Rovira i Virgili (Spain)
Author
Instituto Univ. de Microelectrónica Aplicada, Univ. de Las Palmas de Gran Canaria (Spain)