Neural networks method for identification of the objects behind the screen

Author

Ratajewicz-Mikolajczak, Elzbieta ; Sikora, Jan

Author_Institution

Inst. of Electr. Eng. & Electrotechnologies, Lublin Tech. Univ., Poland

Volume

21

Issue

6

fYear

2002

fDate

6/1/2002 12:00:00 AM

Firstpage

613

Lastpage

619

Abstract

The possibilities of identification of intraventicular haemorrhage with the aid of electrical impedance tomography are presented in this paper. On the base of the tomographic image of a human brain a simplified two-dimensional mathematical model of the human head is proposed. The mathematical model is relatively poor, but the main purpose of this paper is to investigate the possibility and precision of the identification procedure. This task is particularly difficult because the object is hidden behind the skull, which becomes a low-conductivity screen when the scalp electrodes are used.

Keywords

blood; brain models; electric impedance imaging; finite element analysis; medical image processing; neural nets; electrical impedance tomography; human brain; human head; identification procedure precision; intraventicular haemorrhage identification; low-conductivity screen; medical diagnostic imaging; scalp electrodes; simplified two-dimensional mathematical model; skull; Biological neural networks; Head; Hemorrhaging; Humans; Impedance; Mathematical model; Neural networks; Scalp; Skull; Tomography; Algorithms; Brain; Computer Simulation; Electric Impedance; Electrodes; Finite Element Analysis; Head; Hematoma, Subdural; Humans; Models, Biological; Neural Networks (Computer); Reproducibility of Results; Sensitivity and Specificity; Stochastic Processes; Tomography;

fLanguage

English

Journal_Title

Medical Imaging, IEEE Transactions on

Publisher

ieee

ISSN

0278-0062

Type

jour

DOI

10.1109/TMI.2002.800581

Filename

1021927