The temporomandibular joint (TMJ) is the joint that via muscle action and jaw motion allows for necessary physiological performances such as mastication. Whereas mandible translates and rotates [1]. Estimation of activity of muscles acting at the TMJ provides a knowledge of activation pattern solely of a specific patient that an electromyography (EMG) examination was carried out [2]. In this work, a Self-Organising Maps (SOMs) based method was used to identify a hidden similarities in EMG data between different patients. Artificial Neural Network algorithm used for this work – Supervised Kohonen Network (SKN) was proposed and described in [3] and extended into a MATLAB toolbox by [4]. SKN algorithm was used to simulate the model, resulting in a chart called class profiles describing the calculated averages of the Kohonen weights of each variable. It shows how data similarity coming from different subjects is distributed. The obtained results let one to draw a conclusion about muscle significance during specific motion. This SOM based modelling method was intended to estimate TMJ instability and its muscle performance during jaw motions. The study was an attempt to identify and categorise patients with similarly possible disorders in TMJ area which can be evaluated from their muscles activation data.
Authors
Additional information
- Category
- Aktywność konferencyjna
- Type
- publikacja w wydawnictwie zbiorowym recenzowanym (także w materiałach konferencyjnych)
- Language
- angielski
- Publication year
- 2021
