Segmentation of the corpus luteum on Doppler ultrasound images using CNN in order to enable the early diagnosis of pregnancy in cows

• Thiago M. dos Santos Dias Goiano Federal Institute
• Cristiane de F. dos Santos Goiano Federal Institute
• Hugo Jaime M. C. Peron Goiano Federal Institute
• Fabrício C. dos Santos Goiano Federal Institute

DOI: https://doi.org/10.5753/erigo.2021.18441

Abstract

The early determination of pregnancy in cows using ultrasound can generate great savings and increase herd production. Observing the evidence that perfusion in the corpus luteum provides a good parameter to determine early pregnancy, this study aims to perform the segmentation of images from Doppler applied to the reproductive system of bovines, in order to enable early diagnosis of pregnancy, since the calculation of the perfusion area in Doppler ultrasound is trivial in face of a segmented corpus luteum. To perform the segmentation task, a CNN with UNet architecture was developed, resulting in an accuracy of 0.8847 in the set of test images.

References

Cai, Y., Li, Q., Fan, Y., Zhang, L., Huang, H., and Ding, X. (2021). An automatic trough line identification method based on improved unet. Atmospheric Research, 264:105839.

Cheung, W., Stevenson, G. N., de Melo Tavares Ferreira, A. E. G., Alphonse, J., and Welsh, A. W. (2020). Feasibility of image registration and fusion for evaluation of structure and perfusion of the entire second trimester placenta by three-dimensional power doppler ultrasound. Placenta, 94:13–19.

de Freitas, D. S., da Silva Camargo, S., Comin, H. B., Domingues, R., Gaspar, E. B., and Cardoso, F. F. (2019). Reconhecimento da ceratoconjuntivite infecciosa bovina utilizando imagens termográficas e redes neurais convolucionais. Revista Brasileira de Computação Aplicada, 11(3):133–145.

Embrapa (2019). Diagnóstico precoce de prenhez com ultrassom Doppler em vacas ajuda a aumentar produção.

Ma, J., Fan, X., Yang, S. X., Zhang, X., and Zhu, X. (2017). Contrast Limited Adaptive Histogram Equalization Based Fusion for Underwater Image Enhancement. Preprints, (March):1–27.

Medeiros, M., Farinasso, A., Nery, I., Souza, H., Vasconcelos, J., Trindade, K., Batista, A., Manso, H., Carneiro, G., and Wischral, A. (2017). Estudo da vascularização fo-licular e do corpo lúteo de éguas cíclicas tratadas com extrato de pituitaria equina utilizando ultrassom doppler colorido. Arquivo Brasileiro de Medicina Veterinária e Zootecnia, 69:1089–1096.

Pozor, M. and McDonnell, S. (2004). Color doppler ultrasound evaluation of testicular blood flow in stallions. Theriogenology, 61(5):799–810.

Pugliesi, G., Rezende, R. G., Silva, J. d., Lopes, E., Nishimura, T. K., Baruselli, P. S., Madureira, E. H., and Binelli, M. (2017). Uso da ultrassonografia doppler em programas de iatf e tetf em bovinos. Rev Bras Reprod Anim, 41(1):140–150.

Rodríguez Alvarez, J., Arroqui, M., Mangudo, P., Toloza, J., Jatip, D., Rodríguez, J. M.,Teyseyre, A., Sanz, C., Zunino, A., Machado, C., and Mateos, C. (2018). Body condition estimation on cows from depth images using convolutional neural networks. Computers and Electronics in Agriculture, 155:12–22.

Rollim, V. M. (2017). Uso da ultrassonografia na reprodução de bovinos.

Ronneberger, O., Fischer, P., and Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. In International Conference on Medical image computing and computer-assisted intervention, pages 234–241. Springer.

Santos, A. M. d., Seixas, J. M. d., Pereira, B. d. B., and Medronho, R. d. A. (2005). Usando redes neurais artificiais e regressão logística na predição da hepatite a. Revista Brasileira de Epidemiologia, 8:117–126.

Trevisol, E, F. J. D. F. A. J. M. C. F. J. (2011). EFFECT ON CORPUS LUTEUM VOLUME AND VASCULARITY IN COWS AFTER OVULATION INDUCTION WITH DIFFERENT GnRH ANALOGUES.