A newly engineered AI system offers hope to couples struggling with male infertility by identifying sperm that conventional methods miss. The Sperm Tracking and Recovery (STAR) platform combines advanced imaging, AI analysis, and microfluidic techniques to uncover ultra-scarce sperm in men diagnosed with azoospermia—a condition where no sperm appear in semen samples.
Study findings, presented by Hemant Suryawanshi, PhD, of Columbia University Fertility Center in New York, reveal that STAR successfully pinpointed sperm in every one of 10 test samples where sperm had been artificially introduced, whereas a trained clinician only detected sperm in three of those same samples. Across all trials, STAR identified 167 sperm cells, compared to just four found through manual inspection.
When applied to 175 men previously classified as sperm-free, STAR uncovered at least one sperm in 26% of cases. The data suggests that many of these patients actually produce sperm in trace amounts, but existing detection methods lack the sensitivity to reveal them, Suryawanshi explained. What this data shows is that a substantial number of these patients have rare sperm, and the limitation is not the biological absence of sperm for these cases, it’s actually our lack of ability to detect the sperm — and systems like STAR are trying to overcome that limitation.
Male infertility factors into roughly half of all infertility cases, with 10% to 15% of affected men suffering from azoospermia. For couples seeking biological conception when the male partner has this condition, standard procedures like testicular sperm extraction (TESE) or microsurgical TESE, which surgically retrieve sperm from testicular tissue, are often attempted.
Read Also: Study Challenges Risks of Very Low Protein Diet for Kidney
To evaluate STAR’s effectiveness, researchers introduced 40 sperm into azoospermic samples, then split each into two portions, one analyzed by STAR, the other by a human expert. Given 60 minutes per sample, STAR consistently located all sperm across 10 trials, while the clinician found sperm in only three instances. The system’s detection rate exceeded manual methods by 41 times, with STAR identifying 167 sperm versus just four through traditional inspection.
Among the 175 patients referred for STAR analysis, all initially deemed sperm-negative, the technology detected at least one sperm in 26% of samples. Notably, 29% of patients who had previously undergone TESE but remained sperm-negative were found to have recoverable sperm via STAR. The system has already contributed to three live births, with retrieved sperm used in intracytoplasmic sperm injection (ICSI) or preserved for future use.
One breakthrough case involved a man whose ejaculate tested positive for sperm after two failed TESE procedures. Despite 19 years of infertility, the couple achieved a live birth in December 2025 using STAR-detected sperm. Two additional births resulted from STAR-assisted conceptions in July and the current month, according to Suryawanshi.