Sex estimation from posterior cranial fossa with machine learning algorithms

The aim of the study is to determine whether there is a significant difference between male and female in terms of morphometric measurements of anatomical structures in the fossa cranii posterior using machine learning algorithms. A total of 200 individuals, consisting of 100 men and 100 women who met the inclusion and exclusion criteria for the BAIBU Education and Research Hospital, were included in the study. Measurements were taken with 26 parameters using volume rendering technique on the images. Logistic Regression, Random Forest, Decision Tree, K-Nearest Neighbour, Gaussian Naive Bayes, Linear Discriminant Analysis, and Support Vector Machine algorithms were employed in the study. As considering the accuracy score, it is observed that the most effective algorithm is Random Forest, with a value of 0.894 for the training set and 0.875 for the test set. It has been concluded that machine learning algorithms can be used to determine gender by utilizing morphometric measurements of anatomical structures in the posterior cranial fossa.