Application of data mining techniques on EMG registers of hemiplegic patients

Authors:
Ana Aguilera;Alberto Subero;Ramón Mata-Toledo
Affiliations:
Center of Analysis, Modelling and Treatment of Data, CAMYTD, Facultad de Ciencias y Tecnología, Universidad de Carabobo, Valencia, Venezuela;Center of Analysis, Modelling and Treatment of Data, CAMYTD, Facultad de Ciencias y Tecnología, Universidad de Carabobo, Valencia, Venezuela;Computer Science Department, James Madison University, Harrisonburg, Virginia
Venue:
ICDM'13 Proceedings of the 13th international conference on Advances in Data Mining: applications and theoretical aspects
Year:
2013

Citing 10
Cited 0

C4.5: programs for machine learning

C4.5: programs for machine learning
Averaging over decision stumps

ECML-94 Proceedings of the European conference on machine learning on Machine Learning
Random Forests

Machine Learning
Multiclass Alternating Decision Trees

ECML '02 Proceedings of the 13th European Conference on Machine Learning
Functional Trees

Machine Learning
Kernel-Based Feature Extraction for Automated Gait Classification Using Kinetics Data

ICNC '08 Proceedings of the 2008 Fourth International Conference on Natural Computation - Volume 04
Gait classification in children with cerebral palsy by Bayesian approach

Pattern Recognition
Decision tree grafting from the all-tests-but-one partition

IJCAI'99 Proceedings of the 16th international joint conference on Artificial intelligence - Volume 2
A novel gait analysis system based on adaptive neuro-fuzzy inference system

Expert Systems with Applications: An International Journal
Mining gait pattern for clinical locomotion diagnosis based on clustering techniques

ADMA'06 Proceedings of the Second international conference on Advanced Data Mining and Applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

Gait analysis provides a very large data volume coming from kinematic, kinetic, electromyographic (EMG) registers and physical examinations. The analysis and treatment of these data is difficult and time consuming. This work applies and explores exhaustively different analysis methods from data mining on these gait data. This study aims to provide a classification system based in gait patterns obtained from EMG records in children with spastic hemiplegia. The methods studied from data mining specifically for the classification task include SVM, neural networks, decision trees, regression logistic models and others. Different techniques of feature extraction and selection have been also employed and combined with classifications methods. The LMT algorithm provides the best result with 97% of instances classified correctly taking into account the indicators for 2 legs. A qualitative and quantitative validation were performed on the data.