Examination of semen characteristics is routinely performed for fertility status investigation of the male partner of an infertile couple as well as for evaluation of the sperm donor candidate. A useful tool for preliminary assessment of semen characteristics might be an artificial neural network. Thus, the aim of the present study was to construct an artificial neural network, which could be used for predicting the result of semen analysis based on the basic questionnaire data. On the basis of eleven survey questions two models of artificial neural networks to predict semen parameters were developed. The first model aims to predict the overall performance and profile of semen. The second network was developed to predict the concentration of sperm. The network to evaluate sperm concentration proved to be the most efficient. 92.93% of the patients in the learning process were properly qualified for the group with a correct or incorrect result, while the result for the test set was 85.71%. This study suggests that an artificial neural network based on eleven survey questions might be a valuable tool for preliminary evaluation and prediction of the semen profile.
- MeSH
- Semen Analysis * methods instrumentation MeSH
- Humans MeSH
- Sperm Motility MeSH
- Infertility, Male MeSH
- Neural Networks, Computer * MeSH
- Sperm Count methods instrumentation MeSH
- Surveys and Questionnaires MeSH
- Semen * MeSH
- Sperm Banks MeSH
- Spermatozoa * abnormalities growth & development ultrastructure MeSH
- Check Tag
- Humans MeSH
Background and Objectives: Semen analysis plays a vital role in understanding the healthy state of the sperm in men. The computer aided semen quantification technique quantifies the quality of the sperm from the semen sample which is digitally sampled and processed using digital image processing technique. Methods: The semen samples were collected from 402 infertile men aged between 25-50 years. Similarly 25 samples were collected from the age matched healthy fertile men (control group) as per the diagnostic report from the physician. A total of 427 samples used in this study were analyzed using traditional manual method (ground truth) and the proposed automated method based on the image processing algorithm. Results: Conventional semen analysis procedure was performed manually after liquefaction of the samples. The parameters such as morphology, sperm count and motility types were determined and compared between manual and automated methods. We have achieved a significant repeatability and reproducibility of the results using the automated method. Automated method has demonstrated to be computationally efficient and it required less amount of time to process any given field of view. It is also less susceptible to any rater bias for the analyzed field of view and the results were comparable with the manual method. Conclusions: In this article we describe the developmental stages involved in the semen analysis, custom built automated image analysis protocol and the report generation based on the parameters involving sperm count and motility types.
