An automatic tracking method to measure the mandibula movement during real time MRI

Area of Science:

• Biomedical Engineering
• Dental Imaging
• Biomechanics

Background:

• Mandibular movement is complex and varies between individuals due to temporomandibular joint (TMJ) anatomy.
• Current methods for analyzing TMJ movement in real-time MRI (rt-MRI) are often subjective and time-consuming.
• Patient-centered dental treatment requires accurate assessment of individual condylar function.

Purpose of the Study:

• To develop and validate an automatic tracking algorithm for mandibular movement using rt-MRI.
• To compare the accuracy and efficiency of the automatic algorithm against manual tracking methods.
• To improve the quantitative analysis of TMJ function for dental treatment planning.

Main Methods:

• An automatic least mean square registration (LMS) algorithm was developed for tracking mandibular condylar pathways in rt-MRI.
• rt-MRI data of mouth opening was acquired from ten participants with skeletal class I.
• The LMS algorithm was compared to manual tracking (MT) by analyzing tracking speed, superimposition error, and condylar pathway distance.

Main Results:

• The LMS automatic tracking was 76% faster than manual tracking.
• LMS tracking demonstrated significantly lower superimposition error (0.0289 ± 0.0058) compared to MT (0.059 ± 0.0145).
• Discrepancies between LMS and MT pathways exceeded 1 mm and 1° during one-third of the observed movement.

Conclusions:

• Automatic condylar movement tracking in rt-MRI offers significant improvements in speed and accuracy over manual methods.
• This automated approach provides a foundation for more objective and quantitative evaluation of TMJ dynamics.
• The findings support the integration of advanced imaging analysis for personalized dental care and treatment planning.