Briefly, the semen samples were collected at a room near the laboratory or at home and handled according to the WHO guidelines Furthermore, statistical significance was tested by a corrected paired t-test, where a p-value below or equal to 0. One limitation of these algorithms is that they are only able to predict one value at a time, meaning we had to run them once for each of the three sperm motility variables. Using data from 85 participants and three-fold cross-validation, we observe that the initial results are promising. CASA was introduced during the s after the digitization of images made it possible to analyze images using a computer. Andersen and Oliwia Witczak. Similar to the machine learning algorithms, all methods which combined the participant data with the videos performed worse than those without, leading to the same conclusion as previously discussed. Since sperm concentration is an important confounding variable when assessing sperm motility by CASA, we performed additional experiments using the two best-performing algorithms to investigate whether or not it had any influence.