Performance evaluation of sperm concentration, motility, and morphological analysis for GSA-810 series of sperm quality analysis system

Abstract

Background: The performance evaluation of each computer-assisted sperm analysis (CASA) system may provide a basis for the interpretation of clinical results and further improvement of the CASA system.

Methods: The accuracy of the GSA-810 CASA system was evaluated by detecting latex bead quality control products. The precision of sperm concentration, morphology, and percentages of progressively motile sperm (PR) were evaluated by coefficient of variation (CV). Three samples with sperm concentration of about 100 × 10⁶ /mL were diluted to evaluate the linear range.

Results: The detection values of latex beads were within the range of target values. The CVs of sperm concentration and PR were significantly and negatively correlated with sperm concentration (r = -0.561, p = 0.001) and PR value (r = -0.621, p < 0.001), respectively. The R² values of the linear range of sperm concentration were ≥0.99. There was no significant difference in sperm motility and PR within 1-10 min at 36.5°C ± 0.5°C. The coincidence rates of sperm morphology and sperm head morphology for 36 semen samples analyzed by the GSA-810 system and manual method were 99.40% and 99.67%, respectively. The CVs of the percentage of sperm with abnormal morphology and percentage of sperm with abnormal head morphology were less than 5%.

Conclusion: The GSA-810 system can accurately analyze normal semen samples, but the repeatability of the results is poor for oligozoospermia and asthenozoospermia samples. The future CASA system for analyzing sperm morphology should focus on recognizing the middle and tail segments of a spermatozoon.

Keywords: computer-assisted sperm analysis system; performance evaluation; sperm concentration; sperm morphology; sperm motility.

FIGURE 1

Linear range of sperm concentration detected by the GSA‐810 system. Three semen samples with sperm concentration of about 100 × 10⁶/mL were diluted with their own seminal plasma for 2, 4, 7, 10, 20, and 50 times, respectively. Each concentration point of each sample was detected three times by the GSA‐810 system, and the mean value was calculated. A correlation curve was drawn based on the measured value and theoretical value, and the R ² value was calculated. All R ² values were ≥0.99.

FIGURE 2

Correlations of the CV values of sperm concentration and PR with the mean values of sperm concentration and PR. CV, coefficients of variation; PR, percentage of progressively motile sperm. Sperm concentration and PR of 30 fresh semen samples were analyzed 10 times repeatedly for each sample using the GSA‐810 system. The mean value and CV for each sample were calculated. The correlations of the mean value of sperm concentration with the CV values of sperm concentration and PR were analyzed. (A) The CV of sperm concentration was significantly and negatively correlated with sperm concentration (r = −0.561, p = 0.001). (B) The CV of sperm PR was significantly and negatively correlated with sperm concentration (r = −0.439, p = 0.017). (C) The CV of sperm PR was significantly and negatively correlated with sperm PR value (r = −0.621, p < 0.001).