Rapid and label-free identification of normal spermatozoa based on image analysis and micro-Raman spectroscopy.

Semen analysis is performed for evaluation of fertility disorders, however it is susceptible to subjectivity of investigators, and lacking of objective criterion for sperm cell quality remains a problem. There is an ongoing debate on which criteria should be employed to define normal spermatozoa. Here, the aim of our study is to evaluate the possibility of label-free and rapid identification of normal sperm cell through the use of image analysis combined with micro-Raman spectroscopy. By using a smooth-surfaced and hydrophobic substrate, spermatozoa smear was rapidly prepared for microscopic imaging and acrosome area can be clearly visualized without any extra stains; then a self-written image analysis program was utilized to segment sperm head and acrosome area and automatically calculates morphological indices. Most important, intensity ratio of 1055 cm(-1) to 1095 cm(-1) from the obtained Raman spectra is found to indicate a potential biomarker for assessing the sperm DNA integrity. Our preliminary study demonstrates that micro-Raman spectroscopy combined with image analysis can be a potentially useful tool for rapid and label-free identification of normal sperm cell by providing both morphological and biochemical information.