Statistical approach to boar semen evaluation using intracellular intensity distribution of head images

Area of Science:

• Veterinary Science
• Biotechnology
• Image Analysis

Background:

• Accurate assessment of sperm viability is crucial in animal reproduction.
• Current methods for boar sperm classification may have limitations in precision and efficiency.

Purpose of the Study:

• To develop and validate a novel method for classifying boar sperm heads based on intracellular intensity distributions.
• To improve the accuracy of estimating the fraction of live sperm cells in boar semen samples.

Main Methods:

• Image pre-processing including segmentation and normalization of boar sperm heads.
• Defining a model intensity distribution for live boar sperm heads using expert-annotated images.
• Calculating deviation from the model distribution to classify sperm as alive or dead.

Main Results:

• The proposed method achieves minimal error in estimating the fraction of alive sperm cells.
• For samples of 100 sperm heads, the standard deviation of the fraction estimation error is less than 0.04 in the 0.7-1.0 range.
• The method's estimations are within 8% of veterinary expert estimations 95% of the time.

Conclusions:

• The developed method provides a reliable and accurate approach for boar sperm head classification.
• This technique meets the practical requirements of veterinary practice for sperm viability assessment.
• The image analysis method offers a significant advancement in boar fertility diagnostics.