D 2015

Analysis of Neurosurgery Data Using Statistical and Data Mining Methods

BERKA, Petr, J. JABLONSKÝ, Luboš MAREK and Michal VRABEC

Basic information

Original name

Analysis of Neurosurgery Data Using Statistical and Data Mining Methods

Authors

BERKA, Petr (203 Czech Republic, guarantor, belonging to the institution), J. JABLONSKÝ (203 Czech Republic), Luboš MAREK (203 Czech Republic) and Michal VRABEC (203 Czech Republic)

Edition

9414. vyd. Switzerland, Advances in Artificial Intelligence and Its Applications. 14th. Mexican Int. Conference on Artificial Intelligence, p. 310-321, 12 pp. 2015

Publisher

Springer International Publishing

Other information

Language

English

Type of outcome

Stať ve sborníku

Field of Study

10201 Computer sciences, information science, bioinformatics

Country of publisher

Switzerland

Confidentiality degree

není předmětem státního či obchodního tajemství

Publication form

printed version "print"

RIV identification code

RIV/04274644:_____/15:#0000087

Organization unit

University of Finance and Administration

ISBN

978-3-319-27100-2

ISSN

UT WoS

000367681400023

Keywords in English

Engineering controlled terms; Artificial intelligence; Decision trees; Neurosurgery; Patient monitoring; Patient treatment; Surgery; Trees (mathematics)

Tags

International impact, Reviewed
Změněno: 29/3/2017 14:52, Ing. Dominika Moravcová

Abstract

V originále

The data concerning the outcomes of surgical clipping and endovascular treatment in acute aneurysmal subarachnoid hemorrhage (SAH) patients have been analyzed to reveal relations between subjective neuropsychological assessments, measurable characteristics of the patient and the disease, and the type of treatment the patient had undergone one year before. We build upon results of previous analyses where have been found that the differences in neuropsychological assessment of the patients treated by either coiling or clipping was small and slightly in favor of surgical group. Using this data, we compare the “classical” statistical and data mining approach. While statistics offers techniques based on contingency tables, where the compared variables have to be manually selected, data mining methods like association rules, decision rules or decision trees offer the possibility to generate and evaluate a number of more complex hypotheses about the hidden relationships. We used SAS JMP to perform the statistical analysis and LISp-Miner system for the data mining experiments. © Springer International Publishing Switzerland 2015.