The Matlab toolbox SciXMiner is designed for the visualization and analysis of time series and features with a special focus to classification problems. It was developed at the Institute of Applied Computer Science of the Karlsruhe Institute of Technology (KIT), a member of the Helmholtz Association of German Research Centres in Germany. The aim was to provide an open platform for the development and improvement of data mining methods and its applications to various medical and technical problems.SciXMiner bases on Matlab (tested for the version 2017a). Many functions do not require additional standard toolboxes but some parts of Signal, Statistics and Wavelet toolboxes are used for special cases. The decision to a Matlab-based solution was made to use the wide mathematical functionality of this package provided by The Mathworks Inc. MATLAB R and Simulink R are registered trademarks of The MathWorks, Inc.SciXMiner is controlled by a graphical user interface (GUI) with menu items and control elements like popup lists, checkboxes and edit elements. This makes it easier to work with SciXMiner for inexperienced users. Furthermore, an automatization and batch standardization of analyzes is possible using macros. The standard Matlab style using the command line is also available.

The accuracy of many regression models suffers from inhomogeneous data coverage. Models loose accuracy because they are unable to locally adapt the model complexity. This article develops and evaluates an automated design process for the generation of hybrid regression models from arbitrary submodels. For the first time, these submodels are weighted by a One-Class Support Vector Machine, taking local data coverage into account. Compared to reference regression models, the newly developed hybrid models achieve significant better results in nine out of ten benchmark datasets. To enable straightforward usage in data science, an implementation is integrated in the open source MATLAB toolbox SciXMiner.

The hACS therefore offers an alternative solution for simultaneous and proportional myoelectric control of two degrees of freedom that avoids several robustness issues related to machine learning based approaches.

No abstract

ZusammenfassungBioelektrische Signale werden oft für Steuerungsaufgaben in der Rehabilitationstechnik eingesetzt. So stellt die antagonistische myoelektrische Steuerung den de-facto Standard für die Ansteuerung von Handprothesen dar. Neuerdings wird auch die Navigation eines Elektrorollstuhls über die Kontraktionen zweier Ohrmuskeln erforscht. Charakteristisch für diese Anwendungen sind individuelle Amplituden und unbeabsichtigte Koaktivierungen, die eine direkte Interpretation der Intention des Anwenders erschweren. Dieser Artikel diskutiert Einflussgrößen auf die Qualität der myoelektrischen Signale und stellt eine Signalverarbeitungskette zur Bereinigung des Signals bzw. zur Intentionsschätzung vor. Zur individuellen Anpassung der Parameter werden standardisierte Trainingsparadigmen eingeführt. Mit Hilfe von Regressionsmodellen sollen Fehlerquellen wie Koaktivierungen eliminiert werden. Die Funktionalität der Verfahren wird anhand simulierter und realer Daten von zweikanaligen myoelektrischen Messungen von Unterarm und Ohr nachgewiesen.