Supervised Machine Learning Classification of Human Sperm Head Based on Morphological Features

REVOLLO, NATALIA; REVOLLO, NOELIA; CLAUDIO DELRIEUX; HERRERA, MARCELA; GONZÁLEZ JOSÉ, ROLANDO

Trends and Advancements of Image Processing and Its Applications

Springer

Año: 2021; p. 177 - 191

Biomedical technologies are driving innovative research and development initiativesworldwide, with the general aim of improving human health. Image-baseddiagnosis, in particular, is triggering novel image interpretation methodologies inalmost all imaging modalities related to human health. Infertility is a condition thataffects approximately 48.5 million couples worldwide [1]. Semen analysis is consideredthe main diagnostic source of information in the evaluation of male reproductivecapacity [2, 3]. This analysis is routinely performed using images taken withoptic microscopy devices on carefully produced and prepared samples. The wholeimage interpretation procedure requires highly skilled specialists with specifictraining in andrology and related subjects. Notwithstanding the background andprior experience of the specialists, interpretation is commonly biased, due to intraandinter-subjective variance [4, 5].