Evaluation of outlier prevalence of density distribution in brain computed tomography: Comparison of kurtosis and quartile statistics

In Nea Wang, Hyun Ji Kim, Eun Ji Kim, Young Tak Kim, Dong Ju Kim

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The purpose of this study is to investigate the association between the morphology of the brain computed tomography (CT) density distribution and pathological change of the brain. We retrospectively analyzed CT images of 221 patients with acquired brain injury (normal subject=102 vs. abnormal subject=119), obtained during emergency department admission. The kurtosis and the length of the whisker of the quartile statistics in the density distribution were derived to assess the degree of outliers of the density distribution. Although both parameters showed significance with CT abnormality (p <0.001), the area under the curve (AUC) of length of the whisker was higher than the AUC of kurtosis (0.70, 0.65, respectively). In conclusion, the length of whisker in quartile statistics more reliably reflects the extent of hemorrhagic and edematous lesions than the kurtosis.

Original languageEnglish
Title of host publication2018 6th International Conference on Brain-Computer Interface, BCI 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1-2
Number of pages2
Volume2018-January
ISBN (Electronic)9781538625743
DOIs
Publication statusPublished - 2018 Mar 9
Event6th International Conference on Brain-Computer Interface, BCI 2018 - GangWon, Korea, Republic of
Duration: 2018 Jan 152018 Jan 17

Other

Other6th International Conference on Brain-Computer Interface, BCI 2018
CountryKorea, Republic of
CityGangWon
Period18/1/1518/1/17

Keywords

  • acquired head injury
  • computed tomography
  • density distribution
  • kurtosis
  • quartile statistics

ASJC Scopus subject areas

  • Artificial Intelligence
  • Human-Computer Interaction
  • Behavioral Neuroscience

Fingerprint Dive into the research topics of 'Evaluation of outlier prevalence of density distribution in brain computed tomography: Comparison of kurtosis and quartile statistics'. Together they form a unique fingerprint.

Cite this