Development of an age estimation method for bones based on machine learning using post-mortem computed tomography images of bones

•We developed an age estimation method for adult bone using machine learning.•The vertebral body, ischial tuberosity, iliac crest, and femur were analyzed.•High-frequency components were extracted from images as feature variables.•Partial least squares regression (PLS) was applied for dimensionality reduction.•Support vector regression with PLS components effectively estimated ages.