ARTIFICIAL INTELLIGENCE TO EVALUATE THE SHORT-TERM PROGRESS OF DEVICE ASSISTED SCOLIOSIS THERAPY ON THE EXAMPLE OF FED METHOD | ||||
Delta University Scientific Journal | ||||
Article 4, Volume 3, Issue 2, September 2020, Page 33-45 PDF (1.14 MB) | ||||
Document Type: Review articles | ||||
DOI: 10.21608/dusj.2020.205877 | ||||
View on SCiNiTO | ||||
Authors | ||||
Paula Schumann1; Andreas Heinke1; Thurid Jochim1; Tilman Lieberknecht1; Jenny Nisser2; Steffen Derlien2; Zbigniew Śliwiński3; Hagen Malberg1; Grzegorz Śliwiński1 | ||||
1Institute of Biomedical Engineering, Technische Universität Dresden, Germany | ||||
2Institute of physiotherapy, University Hospital Jena, Germany | ||||
3Institute of physiotherapy, UJK Kielce, Poland | ||||
Abstract | ||||
X-Ray or video raster stereography are used for the progress control of the FED therapy butapplied only at intervals of months. A short-term evaluation would allow to adjust the therapyparameters based on the individual therapy progression and could also provide a direct feedbackfor patient. Therefore, this study aims to isolate parameters for a short-term progressionmonitoring by applying machine learning algorithms on a set of 130 posture characteristics. Ameasuring procedure using the DIERS formetric 4D optical measuring system was developedand validated on six patients. The measuring procedure was repeated eight times (four days,each morning and afternoon). Eight parameters were evaluated. The Wilcoxon signed rank testand the Friedman test were used to verify the statistical significance. In order to identify smallchanges in posture correlating with the applied treatment a hierarchical cluster analysis wasperformed. The evaluation shows that the parameters pelvic tilt, kyphosis angle and lordosisangle changed significantly between the individual measuring points, but not across all eightparameters. The data is highly dependent on the daily form and cooperation of the patient. Thecluster classification is not determined on the basis of the four measurement points, but on thebasis of patient individuality. Hierarchical clustering can classify new patients to match themwith successful treatment plans of similar cases. By further optimizing the setting parameters abetter cluster result should be achieved. More measurements will be made to expand thedatabase. In order to obtain a short-term patient monitoring, other methods of artificialintelligence especially neural networks will be considered | ||||
Statistics Article View: 228 PDF Download: 146 |
||||