Age Estimation Using Machine Learning Algorithms with Parameters Obtained from X-ray Images of the Calcaneus

Area of Science:

• Forensic science
• Radiology
• Biomedical engineering

Background:

• Bone age determination is crucial for forensic, surgical, and scientific applications.
• Accurate age estimation aids in legal investigations, medical diagnoses, and anthropological studies.

Purpose of the Study:

• To estimate age with high accuracy and precision using Machine Learning (ML) algorithms.
• To analyze parameters derived from calcaneus (heel bone) X-ray images of healthy individuals.

Main Methods:

• Retrospective analysis of foot X-ray images from 341 individuals aged 18-65 years.
• Measurement of calcaneus parameters: maximum width (MW), body width (BW), maximum length (MAXL), minimum length (MINL), facies articularis cuboidea height (FACH), maximum height (MAXH), and tuber calcanei width (TKW).
• Application of ML models to age estimation using grouped measurements (20-45, 46-64, 65+ years).

Main Results:

• The Extra Tree Classifier algorithm achieved an accuracy of 0.85 in age estimation.
• Other ML algorithms demonstrated accuracy rates ranging from 0.78 to 0.82.
• The maximum height (MAXH) parameter of the calcaneus showed the highest contribution to age estimation, as determined by SHAP analysis.

Conclusions:

• The calcaneus bone provides a reliable basis for accurate and precise age estimations.
• ML algorithms applied to calcaneus measurements offer a promising approach for age determination.