New perspectives for computer-aided discrimination of Parkinson's disease and essential tremorPetra Povalej Bržan
, J.A. Gallego
, J. P. Romero
, Vojko Glaser
, E. Rocon
, Julián Benito-León
, Félix Bermejo-Pareja
, Ignacio Posada
, Aleš Holobar
, 2017, original scientific article
Abstract: Pathological tremor is a common but highly complex movement disorder, affecting ~5% of population older than 65 years. Different methodologies have been proposed for its quantification. Nevertheless, the discrimination between Parkinson's disease tremor and essential tremor remains a daunting clinical challenge, greatly impacting patient treatment and basic research. Here, we propose and compare several movement-based and electromyography-based tremor quantification metrics. For the latter, we identified individual motor unit discharge patterns from high-density surface electromyograms and characterized the neural drive to a single muscle and how it relates to other affected muscles in 27 Parkinson's disease and 27 essential tremor patients. We also computed several metrics from the literature. The most discriminative metrics were the symmetry of the neural drive to muscles, motor unit synchronization, and the mean log power of the tremor harmonics in movement recordings. Noteworthily, the first two most discriminative metrics were proposed in this study. We then used decision tree modelling to find the most discriminative combinations of individual metrics, which increased the accuracy of tremor type discrimination to 94%. In summary, the proposed neural drive-based metrics were the most accurate at discriminating and characterizing the two most common pathological tremor types.
Keywords: Parkinson's disease, essential tremor, electromyography, wrist movements, motor units, muscular excitation, decision tree
Published in DKUM: 03.11.2017; Views: 1179; Downloads: 369
Full text (3,31 MB)
This document has many files! More...
Comprehensive decision tree models in bioinformaticsGregor Štiglic
, Simon Kocbek
, Igor Pernek
, Peter Kokol
, 2012, original scientific article
Abstract: Purpose Classification is an important and widely used machine learning technique in bioinformatics. Researchers and other end-users of machine learning software often prefer to work with comprehensible models where knowledge extraction and explanation of reasoning behind the classification model are possible. Methods This paper presents an extension to an existing machine learning environment and a study on visual tuning of decision tree classifiers. The motivation for this research comes from the need to build effective and easily interpretable decision tree models by so called one-button data mining approach where no parameter tuning is needed. To avoid bias in classification, no classification performance measure is used during the tuning of the model that is constrained exclusively by the dimensions of the produced decision tree. Results The proposed visual tuning of decision trees was evaluated on 40 datasets containing classical machine learning problems and 31 datasets from the field of bioinformatics. Although we did notexpected significant differences in classification performance, the resultsdemonstrate a significant increase of accuracy in less complex visuallytuned decision trees. In contrast to classical machine learning benchmarking datasets, we observe higher accuracy gains in bioinformatics datasets. Additionally, a user study was carried out to confirm the assumptionthat the tree tuning times are significantly lower for the proposed method in comparison to manual tuning of the decision tree. Conclusions The empirical results demonstrate that by building simple models constrained by predefined visual boundaries, one not only achieves good comprehensibility, but also very good classification performance that does not differ from usually more complex models built using default settings of the classical decision tree algorithm. In addition, our study demonstrates the suitability of visually tuned decision trees for datasets with binary class attributes anda high number of possibly redundant attributes that are very common in bioinformatics.
Keywords: decision tree models, machine learning technique, visual tuning, bioinformatics
Published in DKUM: 05.06.2012; Views: 1722; Downloads: 314
Full text (524,39 KB)
This document has many files! More...