Automatic Detection of Macrophages in Canine Bone Marrow Smears Using Deep Learning

• Rafael Luz Araújo UFPI / IFPI
• Viviane B. L. Dias UFPI
• Armando Luz Borges UFPI
• Lucas B. M. de Souza UFPI
• Kawan Sousa Dias UFPI
• Clara E. S. Sátiro UFPI
• Clésio de A. Gonçalves IFPI
• Ana C. L. Pacheco UFPI
• Romuere R. V. e Silva UFPI

DOI: https://doi.org/10.5753/sbcas.2025.7650

Abstract

Canine visceral leishmaniasis (CVL) is a neglected zoonosis affecting dogs and humans, posing significant public health challenges due to diagnostic difficulties. This study aimed to enhance the detection of macrophages in microscopic images of canine bone marrow aspirates by employing deep learning techniques. The YOLOv8 model was utilized in conjunction with data augmentation strategies, including color adjustments and geometric transformations. The results were promising, achieving a recall of 0,79 and an mAP50 of 0,85, indicating high sensitivity and precision in detection. This step is crucial for CVL diagnosis, as accurate identification of macrophages facilitates the subsequent detection of amastigotes.

References

Álvaro, C. I. L. S. (2022). Relatório de estágio e monografia intitulada”leishmaniose canina: Sintomas e tratamentos”. Master’s thesis, Universidade de Coimbra (Portugal).

Antunes, T., Godoy, K., Oliveira, G., Silveira, A., Ramos, C., and Souza, A. (2018). Técnicas de citologia aspirativa, biópsia e citobloco de medula óssea para identificação e determinação de intensidade parasitária na leishmaniose visceral canina. Arquivo Brasileiro de Medicina Veterinária e Zootecnia, 70:1362–1368.

Aziz, S., Bilal, M., Khan, M. U., and Amjad, F. (2020). Deep learning-based automatic morphological classification of leukocytes using blood smears. In 2020 International Conference on Electrical, Communication, and Computer Engineering (ICECCE), pages 1–5. IEEE.

Buslaev, A., Iglovikov, V. I., Khvedchenya, E., Parinov, A., Druzhinin, M., and Kalinin, A. A. (2020). Albumentations: Fast and flexible image augmentations. Information, 11(2):125.

Cannet, A., Simon-Chane, C., Histace, A., Akhoundi, M., Romain, O., Souchaud, M., Jacob, P., Sereno, D., Volf, P., Dvorak, V., et al. (2023). Species identification of phle-botomine sandflies using deep learning and wing interferential pattern (wip). Scientific Reports, 13(1):21389.

Coelho, M. L., França, T., Mateus, N. L. F., Junior, M. S. d. C. L., Cena, C., and do Nascimento Ramos, C. A. (2023). Canine visceral leishmaniasis diagnosis by uv spectroscopy of blood serum and machine learning algorithms. Photodiagnosis and Photodynamic Therapy, 42:103575.

CVAT.ai Corporation (2023). Computer Vision Annotation Tool (CVAT).

da Saúde, M. (2025). Painel epidemiológico. [link].

Eder, K., Kutscher, T., Marzi, A., Barroso, Á., Schnekenburger, J., and Kemper, B. (2021). Automated detection of macrophages in quantitative phase images by deep learning using a mask region-based convolutional neural network. In Label-free Biomedical Imaging and Sensing (LBIS) 2021, volume 11655, pages 88–94. SPIE.

Fleiss, J. L., Levin, B., and Paik, M. C. (2013). Statistical methods for rates and proportions. john wiley & sons.

Gonçalves, C., Andrade, A. L., Dias, V. B. L., de Andrade, N. B., Aguiar, B. G. A., Veloso, R. R., et al. (2022). Método automático para detecção de leishmaniose visceral em humanos. In Congresso Brasileiro de Automática-CBA, volume 3.

Gonçalves, C., Andrade, N., Borges, A., Rodrigues, A., Veras, R., Aguiar, B., and Silva, R. (2023). Automatic detection of visceral leishmaniasis in humans using deep learning. Signal, Image and Video Processing, 17(7):3595–3601.

Górriz, M., Aparicio, A., Raventós, B., Vilaplana, V., Sayrol, E., and López-Codina, D. (2018). Leishmaniasis parasite segmentation and classification using deep learning. In Articulated Motion and Deformable Objects: 10th International Conference, AMDO 2018, Palma de Mallorca, Spain, July 12-13, 2018, Proceedings 10, pages 53–62. Springer.

Isaza-Jaimes, A., Bermúdez, V., Bravo, A., Castrillo, J. S., Lalinde, J. D. H., Fossi, C. A., Flórez, A., and Rodríguez, J. E. (2020). A computational approach for leishmania genus protozoa detection in bone marrow samples from patients with visceral leishmaniasis. AVFT–Archivos Venezolanos de Farmacología y Terapéutica, 39(7).

Jocher, G., Chaurasia, A., and Qiu, J. (2023). Ultralytics yolov8.

Koutinas, A. and Koutinas, C. (2014). Pathologic mechanisms underlying the clinical findings in canine leishmaniosis due to leishmania infantum/chagasi. Veterinary pathology, 51(2):527–538.

Kumar, R. and Nylén, S. (2012). Immunobiology of visceral leishmaniasis. Frontiers in immunology, 3:251.

Larios, G., Ribeiro, M., Arruda, C., Oliveira, S. L., Canassa, T., Baker, M. J., Marangoni, B., Ramos, C., and Cena, C. (2021). A new strategy for canine visceral leishmaniasis diagnosis based on ftir spectroscopy and machine learning. Journal of Biophotonics, 14(11):e202100141.

Marcondes, M. and Day, M. J. (2019). Current status and management of canine leishmaniasis in latin america. Research in Veterinary Science, 123:261–272.

Neves, J. C., Castro, H., Tomás, A., Coimbra, M., and Proença, H. (2014). Detection and separation of overlapping cells based on contour concavity for leishmania images. Cytometry Part A, 85(6):491–500.

PAHO (2017). Epidemiological Report of the Americas. Leishmaniases. PAHO Washington, DC, USA.

Redmon, J., Divvala, S., Girshick, R., and Farhadi, A. (2015). You only look once: Unified, real-time object detection. arXiv preprint arXiv:1506.02640.

Sadeghi, A., Sadeghi, M., Fakhar, M., Zakariaei, Z., Sadeghi, M., and Bastani, R. (2024). A deep learning-based model for detecting leishmania amastigotes in microscopic slides: a new approach to telemedicine. BMC Infectious Diseases, 24(1):551.

Silva, F. S. (2007). Patologia e patogênese da leishmaniose visceral canina. Rev Trop Cienc Agr Biol, 1(1):20–31.

Silva, J. M., Zacarias, D. A., Figueirêdo, L. C. D., Soares, M. R. A., Ishikawa, E. A., Costa, D. L., and Costa, C. H. (2014). Bone marrow parasite burden among patients with new world kala-azar is associated with disease severity. American Journal of Tropical Medicine and Hygiene, 90:621–626.