Unsupervised Machine Learning in Classification of Neurobiological Data

Konrad Andrzej Ciecierski , Tomasz Mandat

Abstract

In many cases of neurophysiological data analysis, the best results can be obtained using supervised machine learning approaches. Such very good results were obtained in detection of neurophysiological recordings recorded within Subthalamic Nucleus (STN) during deep brain stimulation (DBS) surgery for Parkinson disease. Supervised machine learning methods relay however on external knowledge provided by an expert. This becomes increasingly di�cult if the subject's domain is highly specialized as is the case in neurosurgery. The proper computation of features that are to be used for classi�cation without good domain knowledge can be diffcult and their proper construction heavily in uences quality of the �nal classi�cation. In such case one might wonder whether, how much and to what extent the unsupervised methods might become useful. Good result of unsupervised approach would indicate presence of a natural grouping within recordings and would also be a further con�rmation that features selected for classi�cation and clustering provide good basis for discrimination of recordings recorded within Subthalamic Nucleus (STN). For this test, the set of over 12 thousand of brain neurophysiological recordings with precalculated attributes were used. This paper shows comparison of results obtained from supervised - random forest based - method with those obtained from unsupervised approaches, namely K-Means and Hierarchical clustering approaches. It is also shown, how inclusion of certain types of attributes in uences the clustering based results.
Author Konrad Andrzej Ciecierski (FEIT / IN)
Konrad Andrzej Ciecierski,,
- The Institute of Computer Science
, Tomasz Mandat - Institute of Psychiatrics and Neurology
Tomasz Mandat,,
-
Pages203-233
Publication size in sheets1.5
Book Bembenik Robert, Skonieczny Łukasz, Protaziuk Grzegorz M., Kryszkiewicz Marzena, Rybiński Henryk (eds.): Intelligent Methods and Big Data in Industrial Applications, Studies in Big Data, vol. 40, 2018, Springer International Publishing, ISBN 978-3-319-77603-3, [978-3-319-77604-0], 376 p., DOI:10.1007/978-3-319-77604-0
Keywords in EnglishSTN, DBS, DWT4 decomposition, Signal power, Unsu- pervised learning, K-Means Clustering, Hierarchical Clustering
DOIDOI:10.1007/978-3-319-77604-0_15
URL https://www.springer.com/la/book/9783319776033
Languageen angielski
File
20170184.pdf 2.45 MB
Score (nominal)15
ScoreMinisterial score = 15.0, 26-06-2018, BookChapterSeriesAndMatConf
Ministerial score (2013-2016) = 15.0, 26-06-2018, BookChapterSeriesAndMatConf
Citation count*
Cite
Share Share

Get link to the record


* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.
Back