Related Experiment Video
Updated: Aug 14, 2026
08:41
Mouse Round Spermatid Injection
Published on: January 26, 2024
Real-time artificial-intelligence-assisted rare sperm searching: a prospective, multicentre, paired sister-oocyte
Dale M Goss1, Steven A Vasilescu2, Phillip A Vasilescu3
1IVF Australia, Alexandria, Sydney, Australia; NeoGenix Fertility Pty Ltd, Sydney, Australia; University of Technology Sydney, Sydney, Australia.
Reproductive Biomedicine Online
|August 12, 2026
Summary
Artificial intelligence (AI)-assisted sperm searching significantly reduces search time and increases sperm recovery in male infertility cases. This technology shows promise for improving outcomes in assisted reproductive technologies.
Area of Science:
- Reproductive Medicine
- Medical Technology
- Infertility Research
Background:
- Severe male factor infertility presents significant challenges in sperm retrieval for assisted reproductive technologies.
- Traditional sperm searching methods can be time-consuming and may yield suboptimal results.
Purpose of the Study:
- To evaluate the efficacy of real-time artificial intelligence (AI)-assisted sperm searching in severe male factor infertility.
- To compare AI-assisted sperm searching with unassisted searching regarding search time and sperm recovery.
- To investigate the impact of AI-assisted sperm searching on downstream embryological outcomes.
Main Methods:
- Prospective, multicentre, within-patient paired split-dish pilot study involving 40 patients with non-obstructive azoospermia or severe oligozoospermia.
- Matched samples were searched with and without AI assistance to compare primary outcomes: time per spermatozoon and time per dish.
- Secondary outcomes included sperm recovery per dish and embryological parameters (fertilization, usable embryos) in a subset of patients.
Main Results:
- AI-assisted sperm searching significantly reduced median time per dish (35.80 min to 21.80 min) and increased median sperm recovery per dish (0.00 to 0.75).
- Median time per spermatozoon decreased from 10.85 min to 5.50 min when sperm were successfully retrieved by both methods.
- AI assistance enabled sperm identification for intracytoplasmic sperm injection (ICSI) in 8 out of 40 cases.
- Exploratory analysis in a subset showed improved median fertilization rates (0.0% to 36.7%) and usable embryo rates (0.00 to 0.21) with AI assistance.
Conclusions:
- AI-assisted sperm localization is effective in reducing search time and enhancing sperm recovery in patients with severe male factor infertility.
- While preliminary, embryological outcomes suggest potential benefits, warranting larger studies for confirmation.
Keywords:
Artificial intelligenceIntracytoplasmic sperm injectionNon-obstructive azoospermiaSevere oligozoospermiaSperm recoverySperm searching
