Title Magnetinio lauko duomenų apdorojimo metodų tyrimas žandikaulio judėjimui sekti /
Translation of Title Investigation of magnetic field data processing methods for jaw motion tracking.
Authors Dranginis, Jokūbas
Full Text Download
Pages 52
Keywords [eng] permanent magnet ; magnetic field ; spatial position ; localization ; jaw tracking
Abstract [eng] The biomechanics of the jaw and its movements play an important role in many physiological functions, including speech, chewing and breathing. Thus, accurate monitoring of jaw movements is of great clinical importance, providing potential insights into health problems such as temporomandibular joint disorders, sleep apnea, and various speech disorders. This paper discusses the use of the magnetic field of a permanent magnet to track jaw movements, focusing on the data processing techniques used to determine spatial position. The permanent magnet enables an unobtrusive and non-invasive technique, providing the opportunity to continuously monitor the jaw without disturbing its natural movements. However, the complexity of the magnetic field equations requires sophisticated data processing techniques to extract a meaningful spatial position from the magnetic field data. This paper examines four such methods: least-squares trust region reflective, particle swarm optimization, look-up table, and neural network. Each technique offers unique data processing advantages and allows accurate, fast tracking of jaw movements. The research empirically evaluated the speed and accuracy of each method in determining the spatial position. The results showed distinct performance characteristics of each method with different trade-offs between computational speed and spatial accuracy. The most accurate spatial position was determined from the experimentally recorded data with the least squares trust region reflective method with a root mean square error of 0.703 mm, the neural network method proved to be the fastest by determining the spatial position in 0.256 s. These findings reveal the potential application of each technique in different clinical scenarios, considering the necessary balance between processing speed and positional accuracy for various diagnostic and therapeutic purposes. Such a study aims to identify possible directions for improving the hardware and software for tracking jaw movements, thus contributing to the improvement of related clinical practice.
Dissertation Institution Kauno technologijos universitetas.
Type Master thesis
Language Lithuanian
Publication date 2023