Treffer: Vertebral Age Estimation Data and Models

Model Files Each “.pkl” file contains a fitted scikit‐learn estimator object. The expected input to each model’s .predict() method is a NumPy array of shape (n_samples, 19), where the 19 columns correspond exactly to the DS values in this order: [DS_C7, DS_T1, DS_T2, DS_T3, DS_T4, DS_T5, DS_T6, DS_T7, DS_T8, DS_T9, DS_T10, DS_T11, DS_T12, DS_L1, DS_L2, DS_L3, DS_L4, DS_L5, DS_S1]– If any DS value is missing (e.g., vertebral level not available), you must impute or omit that row before prediction.– Models were trained on standardized data (each DS column was z‐normalized using the training‐set mean and standard deviation). The pickle files include the fitted scalers if needed. Example: Loading and Using a Model EXAMPLE.txt is a minimal code snippet demonstrating how to load and apply a model to new data. Replace “Male_RF.pkl” with “Female_KNN.pkl” or any other .pkl file as needed. If your workflow includes retraining or cross-validating, ensure that you apply the same scaling and column ordering. Dataset File (RESULTS.xlsx) • Verify that “Age” is stored as a numeric column (integer or float).• The DS_ columns should also be numeric (floating‐point) values.• If you wish to recalculate correlation coefficients or generate new scatterplots, simply read the DS columns and Age column into Python (pandas and numpy) or R. License All files are released under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. You may use, share, and adapt the data and models, provided appropriate credit is given, a link to the license is provided, and any changes made are indicated. Contact If you have questions about dataset content, model formats, or reproducibility, please contact:Y.E. – yasinetli@yyu.edu.tr Thank you for your interest. We hope this repository facilitates further research and validation in vertebral-based age estimation.