This paper presents a semi-autonomous AI-based platform designed for the efficient management and quantitative analysis of human spermatozoa. Addressing the limitations of manual semen analysis, this system integrates advanced image processing and analytical techniques to offer a high-throughput diagnostic solution. During operation, the proposed system autonomously performs a precise quantitative assessment of sperm concentration, accurately tracks individual sperm motility patterns, and systematically classifies morphological abnormalities. The result is a comprehensive sperm analysis report, meticulously generated according to the latest established World Health Organization (WHO) guidelines for concentration, motility, and morphology. A distinguishing feature of this system is the ability to yield reliable preliminary results even with minimally pre-processed clinical samples, thereby enhancing diagnostic objectivity, efficiency, and reliability in male reproductive health assessments.
Trinh et al. (Tue,) studied this question.