Computational Forensics & Machine Learning: Leveraging AI and Machine Learning for Intergenerational Analysis of Craniofacial Heritability of the Indian families using photographs

Background: Facial features are known to be highly heritable, exhibiting remarkable resemblance within families across generations. This inheritance patternhas signi򟿿cant implications in򟿿elds such as forensics, where reconstructing facial characteristics from imited ncestral data can aid in identi򟿿cation and investigation. Aims: This study aims to leverage arti򟿿cial intelligence (AI) and machine learning techniques o onduct a comprehensive computational analysis of craniofacial heritability within Indian families. Methods: A dataset comprising facial photographs of three generations (grandparents, parents and children) from 51 Indian families were compiled. Computer vision algorithms were employed to extract precise anthropometric measurements from these images. Various statistical methods, including Pearson correlation, hypothesis testing (T-tests, ANOVA, chi-square) and imensionality reduction techniques (PCA, PCoA), were applied to quantify intergenerational relationships. Furthermore, machine learning models, such as linear regression and random forest regression, were developed to predict descendant facial features from ancestral data. Results: Pearson Correlation Analysis rev