To establish population-specific age estimation models in adults from costal cartilage for contemporary Chinese by using three-dimensional volume-rendering technique. Five hundred and twelve individuals (254 females and 258 males) with documented ages between 20 and 85 years were retrospectively included. Their clinical CT examinations (1 mm slice thickness) were used to develop the sex-specific age prediction model. A validation sample comprising 26 female and 24 male individuals was then used to test the predictive accuracy of the established models. Simple linear regression (SLR), multiple linear regression (MLR), gradient boosting regression (GBR), support vector machine (SVM), and decision tree regression (DTR) were utilized to build the age diagnosis models from calibration samples. By comparison, the decision tree regression was the relatively more accurate age prediction model for male, with mean absolute error = 5.31 years, least absolute error = 0.10 years, correct percentage within 5 years = 54%, and the correct percentage within 10 years = 88%. The stepwise multiple linear regression equations was the relatively more accurate one for female, with mean absolute error = 6.72 years, least absolute error = 0.68 years, correct percentage within 5 years = 42%, and correct percentage within 10 years = 77%. Our results indicated that the present established age estimation model can be applied as an additional guidance for age estimation in adults.