D 2020

The Black-White Pixels Ratio in Medial Temporal Lobe Brain Structure in Transcranial B-Images as a Measurable Marker of Alzheimer's Disease Probability: The Reproducibility Overview

BLAHUTA, Jiří, Tomáš SOUKUP and Lukáš PAVLÍK

Basic information

Original name

The Black-White Pixels Ratio in Medial Temporal Lobe Brain Structure in Transcranial B-Images as a Measurable Marker of Alzheimer's Disease Probability: The Reproducibility Overview

Authors

BLAHUTA, Jiří (203 Czech Republic, guarantor, belonging to the institution), Tomáš SOUKUP (203 Czech Republic, belonging to the institution) and Lukáš PAVLÍK (203 Czech Republic)

Edition

Croatia, 2020 International Conference on Software, Telecommunications and Computer Networks (SoftCOM), p. 1-6, 6 pp. 2020

Publisher

Institute of Electrical and Electronics Engineers Inc.

Other information

Language

English

Type of outcome

Stať ve sborníku

Field of Study

20206 Computer hardware and architecture

Country of publisher

United States of America

Confidentiality degree

není předmětem státního či obchodního tajemství

Publication form

electronic version available online

RIV identification code

RIV/47813059:19240/20:A0000675

Organization unit

Faculty of Philosophy and Science in Opava

ISBN

978-1-7281-7538-6

Keywords in English

Alzheimer's Disease; B-image echogenicity; black-white ratio; medial temporal lobe echogenicity; MTL atrophy

Tags

Tags

International impact, Reviewed
Změněno: 2/12/2020 11:11, Mgr. Kamil Matula, Ph.D.

Abstract

V originále

The objective of the presented study is to introduce analysis the black-white pixels ratio in medial temporal lobe (MTL) brain structure in digitized transcranial B-images using our developed software B-MODE Assist. This ratio could be a significant marker related to the echogenicity grade. In B-image, the MTL is stable anechogenic structure but increased echogenicity of surrounding cortex is detectable in the case of MTL atrophy; the atrophy predicts Alzheimer's Disease. Within this study, a set of 97 B-images is used. There are two key goals of this pilot study; verify the reproducibility of the ratio between two observers and to show that the ratio is not much different in the case of original image in comparison with a magnified segment and using constant brightness increasing; the ratio should be independent on image resolution and brightness. To select medial temporal lobe structure, a 20mm square-shaped Region of Interest is used. The correlation coefficient r=0.69 was achieved between two independent, non-experienced observers. The error does not exceed 28 % between 2 non-experienced observers and does not exceed 12 % between the measurements from the same observer. Also, the variance and the range were computed. Achieved results have been judged by an experienced sonographer. In the future, the ratio could be an effective value to determine the MTL atrophy based on the ratio value and possibility to create a rule-based expert system which should help evaluate probability of Alzheimer's Disease based on the cortex echogenicity grade. In general, this approach could be a reliable way to detect risk of Alzheimer's Disease using neurosonological imaging.