Quantitation of specific parameters of motility in large numbers of human sperm by digital image processing.

The CellSoft computer-assisted digital image analysis system was validated for quantitating specific motility parameters in large numbers of human sperm. Motility patterns ranging from linear head trajectories (Type 1) to nonlinear, asymmetric patterns with overlapping trajectory (Type 5) were subjectively identified in semen and washed samples prepared for in vitro fertilization. A representative of each type was used for optimizing the digital imaging set-up parameters, tracking rate, and frequency. Each cell type was also characterized according to the following motility parameters: curvilinear velocity (Vcl), straight line velocity (Vsl), linearity of forward progression (Lin), maximum and mean lateral head amplitude (maxLHA; mean LHA), and beat cross frequency (BCF). Comparison of all parameters that could be determined both digitally and manually (Vcl, Vsl, Lin, and BCF) indicated no differences (p greater than 0.05) in Vcl, Lin, or BCF and only slight differences (5-6%) in Vsl measurements. After validation of the digital imaging technique, populations of seminal and washed cells were studied. Replicate analysis of the same sample demonstrated no significant intraassay variability. A comparison of semen and washed cells from 10 different donors indicated that all of the motility parameters, with the exception of Lin, were significantly higher (p less than 0.05) in washed cells. It was concluded that the digital imaging system can adequately and rapidly quantitate a large number of cells with heterogeneous motility patterns. This technique may prove to be useful in defining motility characteristics associated with capacitation, the acrosome reaction, and fertility of human sperm.