The Seville Institute of Biomedicine (Hospital Universitario Virgen del Rocío/CSIC/University of Seville) has recently collaborated with the University of the Basque Country, the Biofisika Institute, the International Physics Centre of Donostia and other prestigious national and international centres to develop CartoCell, a high-content 3D image analysis tool that reveals hidden patterns of cell morphology. This open source software has great potential for application both in basic science, with which it has already been tested, and in clinical and medical development.
Hidden Patterns
Animals and plants, as well as other structures in nature, show patterns that we can easily detect. However, this same phenomenon occurs at the microscopic level, in tissues. «These patterns also appear during embryonic development,» explains Luis M. Escudero, one of the authors and a researcher at the Institute of Biomedicine of Seville (IBiS) and professor at the University of Seville. «For example, there are genes that are expressed in alternating patterns of stripes and that are fundamental for establishing insect segments». «In our work we have found patterns that were still hidden…. And that could only be revealed by using artificial intelligence tools for the analysis of microscopic tissue images.
CartoCell, capable of mapping epithelial tissues
CartoCell is a digital tool capable of automatically working with a huge amount of information, specialising in identifying and studying the shape and spatial distribution of each cell in epithelial tissue, allowing the discovery of patterns associated with different types of tissue from images. So far, the software has been used for basic science studies, mapping the morphology of cells with images of different tissue types. But in addition to hidden patterns and biological tissue research, CartoCell also shows great potential for clinical application. Its advanced image analysis could detect subtle changes at the cellular level, which may be important in the long run, of the effect of each drug against a given disease.
Neural networks and open source
The heart of CartoCell is an artificial neural network. The architecture of the network was designed based on other models previously developed by the research group in artificial vision for bioimaging at the Biofisika Institute and the Donostia International Physics Center (DIPC), in the Basque Country, which is participating in the project.
One of the characteristics of CartoCell is its versatility, according to Dr. Pedro Gómez-Gálvez, also an author of the study and currently at the Department of Physiology, Development and Neuroscience at the University of Cambridge (UK). In addition, CartoCell is an open source tool, which gives other experts the opportunity to freely use, improve or adapt its use. In this sense, the intention has been to create a tool that is accessible, easy to install and use by non-experts, but also adaptable and extensible for anyone who wants to develop new solutions on top of it.
In addition to the institutions mentioned at the beginning of the article, researchers from the University of Seville, the Severo Ochoa Molecular Biology Centre in Madrid, and the MRC Laboratory of Molecular Biology in Cambridge in the United Kingdom, among other institutions, have also participated in this work.
+info: desarrollan-herramienta-de-codigo-abierto-para-detectar-patrones-ocultos-en-tejidos-celulares