https://doi.org/10.1140/epjs/s11734-025-01944-x
Regular Article
Development of an algorithm for recognition of Toxoplasma gondii brain cysts in microscopic images based on fractal dimension
1
Institute of Physics Belgrade, National Institute of Republic of Serbia, University of Belgrade, Pregrevica 118, 11080, Belgrade-Zemun, Serbia
2
National Reference Laboratory for Toxoplasmosis, Group for Microbiology and Parasitology, Centre of Excellence for Food- and Vector-Borne Zoonoses, Institute for Medical Research, National Institute of Republic of Serbia, University of Belgrade, Belgrade, Serbia
a
This email address is being protected from spambots. You need JavaScript enabled to view it.
b
This email address is being protected from spambots. You need JavaScript enabled to view it.
Received:
29
June
2024
Accepted:
8
September
2025
Published online:
19
September
2025
Abstract
Toxoplasma gondii is an obligate intracellular parasite infecting up to one-third of the human population. There are currently no treatment options which could eradicate the parasite from the infected host. A gold-standard, confirmatory diagnostic bioassay test for toxoplasmosis uses microscopic identification of T. gondii cysts in brain samples of mice inoculated with biological samples under investigation, which is time-consuming and requires experienced microscopist. We upgrade the bioassay by applying computational image analysis. Choosing fractal dimension (FD) for automated cyst recognition is convenient, as this parameter is relatively stable for different parasite strains, cyst age, or mice genetic background. It is also convenient to have FD immediately calculated for each cyst found. We employ a block-based multiscale analysis to detect regions in high-resolution microscopic images, fulfilling the conditions to be classified as T. gondii brain cysts. Block size cannot be too small, as (i) it could reduce the FD calculation accuracy and (ii) significantly increase the computational burden of analysis which is meant to be objective, accurate, but also computationally efficient. Rather, the robustness of cyst recognition stems from the proposed framework and also relies in part on the inherent properties of brain cysts and typical cyst surroundings. We evaluate our approach on a number of microscopic images and show the performance estimates. The method could also be used for other objects well characterized by FD or another differentiating parameter.
Copyright comment Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
© The Author(s), under exclusive licence to EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature 2025
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
