Advancing Neurological Healthcare: Machine Learning Applications for Gait Disorders and Fall Detection

This dissertation explores machine learning approaches for the automatic classification of neurological gait disorders and the detection of falls in older adults. Using EMG and wearable sensor data, the study evaluates multiple machine learning algorithms and demonstrates their effectiveness in supporting objective diagnosis, monitoring, and patient safety in neurological healthcare.:2 Introduction .............................................................................. 5 2.1 Background and Motivation ........................................................ 5 2.1.1 Overview of Abnormalities in Walking Patterns ............................ 5 2.1.2 Clinical Significance of Gait Analysis and Fall Detection in Neurological Healthcare ... 8 2.2 Machine Learning ............................................................... 10 2.2.1 Support Vector Machine (SVM) .......................................... 12 2.2.2 K-Nearest Neighbors (kNN) ............................................ 13 2.2.3 Random Forest (RF) ................................................... 15 2.2.4 Convolutional Neural Network (CNN) ................................... 17 2.3 Research Goals and Overview of the Thesis ...................................... 22 2.3.1 Development and Evaluation of Machine Learning Algorithms for Automatic Classification of Gait Disorders ......................................................... 22 2.3.2 Application of Machine Learning in Fall Detection Systems for Elderly Patients ....................................................................................... 23 3 Study 1 .................................................................................. 26 4 Study 2 .................................................................................. 38 5 Summary .................................................................................. 53 6 References ............................................................................... 57 7 Declaration on the Independent Completion of the Work ............................. 61 7.1 Publications ................................................................. 62 7.2 Acknowledgements ............................................................. 63