[HIMAWAN-3] PCA-Based Kinematic Classification of Mandibular Movements from TMD Patients
[Abstract]
[Objective]
This retrospective study aimed to kinematically classify mandibular movements collected during Temporomandibular Disorders (TMD) treatment, employing Principal Component Analysis (PCA) and cluster analysis (k-means), and to investigate their correlation with symptoms of pain-related TMD.
[Method]
This study was conducted with the approval of the Ethics Committee of Tohoku University (No: 33541). Informed consent regarding the clinical research and the handling of personal information was obtained from all participants prior to data collection. The study included five TMD participants (age: 39 to 86 years, with an SD of 18.96) and three healthy participants (age: 32 to 42 years, with an SD of 5.13) with no stomatognathic problems. The TMD participants underwent tailored treatment for their symptoms, and their maximum unassisted mouth opening (MAMO) was recorded randomly with a motion capture system (ARCUS digma II, Kavo, Biberach, Germany) at multiple time points. MAMO for healthy participants served as control. All trials were categorized by three labels: "Symptomatic", "Asymptomatic" and "Control". The dataset comprised 28 trials, transferred to the musculoskeletal simulation 1) (AnyBody Modeling System, AnyBody Technology, Aalborg, Denmark) for extraction of joint angle time series, which were then transformed into Fourier series. Subsequently, PCA and cluster analysis (k-means) were conducted to classify all trials. Within the scatter plots generated through PCA and cluster analysis, MAMO pathways for each TMD participant at multiple time points were observed.
[Results and Discussion]
All MAMO trials were categorized into two clusters (Cluster 1 and Cluster 2). Cluster 1 exhibited a higher prevalence of 'Symptomatic' trials, while Cluster 2 had a higher prevalence of 'Asymptomatic' trials (Fig.1). Additionally, the trials of each TMD participant with pain-related symptoms revealed MAMO pathways between classified two clusters in the scatter plots. This systematic approach enabled an understanding of how specific patterns of MAMO correlate with pain-related symptoms in TMD. Although the small dataset, these results show promise for a novel functional assessment for TMD based on the kinematic features.
[References]
1) Damsgaard M,et al. Analysis of musculoskeletal systems in the AnyBody Modeling System. Simulation Modelling Practice and Theory. 2006; 14(8):1100-11
[Objective]
This retrospective study aimed to kinematically classify mandibular movements collected during Temporomandibular Disorders (TMD) treatment, employing Principal Component Analysis (PCA) and cluster analysis (k-means), and to investigate their correlation with symptoms of pain-related TMD.
[Method]
This study was conducted with the approval of the Ethics Committee of Tohoku University (No: 33541). Informed consent regarding the clinical research and the handling of personal information was obtained from all participants prior to data collection. The study included five TMD participants (age: 39 to 86 years, with an SD of 18.96) and three healthy participants (age: 32 to 42 years, with an SD of 5.13) with no stomatognathic problems. The TMD participants underwent tailored treatment for their symptoms, and their maximum unassisted mouth opening (MAMO) was recorded randomly with a motion capture system (ARCUS digma II, Kavo, Biberach, Germany) at multiple time points. MAMO for healthy participants served as control. All trials were categorized by three labels: "Symptomatic", "Asymptomatic" and "Control". The dataset comprised 28 trials, transferred to the musculoskeletal simulation 1) (AnyBody Modeling System, AnyBody Technology, Aalborg, Denmark) for extraction of joint angle time series, which were then transformed into Fourier series. Subsequently, PCA and cluster analysis (k-means) were conducted to classify all trials. Within the scatter plots generated through PCA and cluster analysis, MAMO pathways for each TMD participant at multiple time points were observed.
[Results and Discussion]
All MAMO trials were categorized into two clusters (Cluster 1 and Cluster 2). Cluster 1 exhibited a higher prevalence of 'Symptomatic' trials, while Cluster 2 had a higher prevalence of 'Asymptomatic' trials (Fig.1). Additionally, the trials of each TMD participant with pain-related symptoms revealed MAMO pathways between classified two clusters in the scatter plots. This systematic approach enabled an understanding of how specific patterns of MAMO correlate with pain-related symptoms in TMD. Although the small dataset, these results show promise for a novel functional assessment for TMD based on the kinematic features.
[References]
1) Damsgaard M,et al. Analysis of musculoskeletal systems in the AnyBody Modeling System. Simulation Modelling Practice and Theory. 2006; 14(8):1100-11