EUVIMEDEUVIMED entdecken
Crossref

Evaluation of a computational model for mycetoma-causative agents identification

Hyam Omar Ali, Romain Abraham, Guillaume Desoubeaux iD, Ahmed H Fahal, Clovis Tauber iD

Transactions of The Royal Society of Tropical Medicine and Hygiene · 06.04.2024 · 118 (4) · 253-263

DOI: 10.1093/trstmh/trad057 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Transactions of The Royal Society of Tropical Medicine and Hygiene, online erschienen am 13.12.2023. Verfasst von Hyam Omar Ali und 4 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Mathematical Sciences, University of Khartoum, 11111, Khartoum, Sudan. Hauptthemen laut MEDLINE: Mycetoma. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 30 Literaturangaben. Gefördert von: Ministry of Higher Education and Scientific Research, Republic of Sudan, L'Oréal-UNESCO for Women in Science, European Mathematical Society.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

The therapeutic strategy for mycetoma relies heavily on the identification of the causative agents, which are either fungal or bacterial. While histopathological examination of surgical biopsies is currently the most used diagnostic tool, it requires well-trained pathologists, who are lacking in most rural areas where mycetoma is endemic. In this work we propose and evaluate a machine learning approach that semi-automatically analyses histopathological microscopic images of grains and provides a classification of the disease as eumycetoma or actinomycetoma.

Methods

The computational model is based on radiomics and partial least squares. It is assessed on a dataset that includes 890 individual grains collected from 168 patients originating from the Mycetoma Research Centre in Sudan. The dataset contained 94 eumycetoma cases and 74 actinomycetoma cases, with a distribution of the species among the two causative agents that is representative of the Sudanese distribution.

Results

The proposed model achieved identification of causative agents with an accuracy of 91.89%, which is comparable to the accuracy of experts from the domain. The method was found to be robust to a small error in the segmentation of the grain and to changes in the acquisition protocol. Among the radiomics features, the homogeneity of mycetoma grain textures was found to be the most discriminative feature for causative agent identification.

Conclusion

The results presented in this study support that this computational approach could greatly benefit rural areas with limited access to specialized clinical centres and also provide a second opinion for expert pathologists to implement the appropriate therapeutic strategy.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

MycetomaHumansBiopsySudan

Stichwörter der Autor:innen

diagnosishistologymachine learningmycetoma

Autor:innen und Institutionen

  • Faculty of Mathematical Sciences, University of Khartoum, 11111, Khartoum, Sudan.
  • Mycetoma Research Centre, University of Khartoum, 11111, Khartoum, Sudan.
  • Boulevard Tonnellé, University of Tours, Inserm, 37032, Tours, France.
  • CNRS U7013, rue de Chartres, University of Orleans, 45067, Orleans, France.
  • Parasitology and Mycology Department, Bretonneau Hospital, 37032, Tours, France.
  • Boulevard Tonnellé, University of Tours, 37032, Tours, France.

ORCID-Profile: Guillaume Desoubeaux, Clovis Tauber

Förderung

  • Ministry of Higher Education and Scientific Research, Republic of Sudan
  • L'Oréal-UNESCO for Women in Science
  • European Mathematical Society

Zitieren

APA

Omar Ali, H., Abraham, R., Desoubeaux, G., Fahal, A. H., & Tauber, C. (2024). Evaluation of a computational model for mycetoma-causative agents identification. Transactions of The Royal Society of Tropical Medicine and Hygiene, 118(4), 253-263. https://doi.org/10.1093/trstmh/trad057

Vancouver

Omar Ali H, Abraham R, Desoubeaux G, Fahal AH, Tauber C. Evaluation of a computational model for mycetoma-causative agents identification. Transactions of The Royal Society of Tropical Medicine and Hygiene. 2024;118(4):253-263. doi:10.1093/trstmh/trad057

BibTeX
@article{OmarAli2024,
  author = {Omar Ali, Hyam and Abraham, Romain and Desoubeaux, Guillaume and Fahal, Ahmed H and Tauber, Clovis},
  title = {Evaluation of a computational model for mycetoma-causative agents identification},
  journal = {Transactions of The Royal Society of Tropical Medicine and Hygiene},
  year = {2024},
  volume = {118},
  number = {4},
  pages = {253--263},
  doi = {10.1093/trstmh/trad057}
}
Kostenlos starten →