Detecting the Area of Bovine Cumulus
Oocyte Complexes Using Deep Learning
and Semantic Segmentation
Georgios ATHANASIOU
a
, Jesus CERQUIDES
a
, Annelies RAES
b
,
Nima AZARI-DOLATABAD
b
, Daniel ANGEL-VELEZ
b
, Ann VAN SOOM
b
, and
Josep-Lluis ARCOS
a
a
Artificial Intelligence Research Institute (IIIA), CSIC, Campus UAB, 08193 Bellatera,
Spain
b
Department of Internal Medicine, Reproduction and Population Medicine, Ghent
University, 9820 Merelbeke, Belgium
Abstract. The cumulus-oocyte complex (COC) is an oocyte surrounded by spe-
cialized granulosa cells, called cumulus cells. The cumulus cells surrounding the
oocyte ensure healthy oocyte and embryo development. The maturity of COCs at
oocyte retrieval may be used as an indicator to predict outcome of assisted repro-
ductive technology (ART). Segmenting COCs is a preliminary step in many im-
age processing pipelines to evaluate maturity. However, acquiring well-annotated
bright-field microscopy image datasets remains a time-consuming and inaccurate
procedure, for most biological domains. Additionally, specialists often partially dis-
agree on their annotations, not only among each other, but also among their own an-
notations, leading to an inconsistent outcome. Despite the recent advancements in
deep learning and image segmentation tools for biological and biomedical images,
there is limited usage of them for having more accurate and automated procedures.
In this work, we propose an automated pipeline to segment bovine COCs in bright-
field microscopy images. The results of our evaluation show that our pipeline is
able to segment COCs with the same level of quality as provided by human experts.
Keywords. Deep Learning, Bright-Field Microscopy, Biomedical Imaging, Image
Segmentation, Image Analysis
1. Introduction
Infertility is defined as a failure to achieve clinical pregnancy of 12 months or more
of regular, unprotected intercourse, and is a big issue for medicine and society. Once
the disease is diagnosed, the treatments involve the techniques of Assisted Reproductive
Technology (ART). Methods of ART are considered the intracytoplasmic injection of
sperm (ICSI) and the in-vitro fertilization (IVF). These methods require several sub-
steps, among of which the characterization of morphological characteristics of oocyte
and embryo biology elements.
Cumulus expansion is a key element for characterizing the quality of mammalian
oocytes, for later use in in-vitro fertilization (IVF). There are several methods for mea-
Artificial Intelligence Research and Development
A. Cortés et al. (Eds.)
© 2022 The authors and IOS Press.
This article is published online with Open Access by IOS Press and distributed under the terms
of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0).
doi:10.3233/FAIA220346
249