Automatic segmentation of the thumb trapeziometacarpal joint using parametric statistical shape modelling and random forest regression voting

• Published in: Computer methods in biomechanics and biomedical engineering. Vol. 7; no. 3; pp. 297 - 301
• Main Authors: Schneider, Marco T. Y., Zhang, Ju, Crisco, Joseph J., Weiss, Arnold-Peter C., Ladd, Amy L., Nielsen, Poul M. F., Besier, Thor
• Format: Journal Article
• Language: English
• Published: England Taylor & Francis 04.05.2019
• Subjects: Automatic segmentation; model generation; random forest; regression voting; statistical shape model; trapeziometacarpal; random forest; trapeziometacarpal; automatic segmentation; model generation; regression voting; statistical shape model
• ISSN: 2168-1163; 2168-1171
• DOI: 10.1080/21681163.2018.1501765

We propose an automatic pipeline for creating shape modelling suitable parametric meshes of the trapeziometacarpal (TMC) joint from clinical CT images for the purpose of batch processing and analysis. The method uses 3D random forest regression voting with statistical shape model segmentation. The method was demonstrated in a validation experiment involving 65 CT images, 15 of which were randomly selected to be excluded from the training set for testing. With mean root mean squared errors of 1.066 and 0.632 mm for the first metacarpal and trapezial bones, respectively, and a segmentation time of ~2 min per CT image, the preliminary results showed promise for providing accurate 3D meshes of TMC joint bones for batch processing.