Synthetic Image Sequence Generation for Endothelium in Situ Simulator

Synthetic image generation (SIG) has played a relevant role in the cell physiology field to validate image processing algorithms, and its usage may be extended to simulator development, helping to consolidate knowledge, reduce the usage of experimental animals, and allow research progress without hi...

Full description

Saved in:
Bibliographic Details
Published in2023 Congress in Computer Science, Computer Engineering, & Applied Computing (CSCE) pp. 517 - 521
Main Authors Sanchez-Tecuatl, Marcial, Moccia, Francesco, Berra-Romani, Roberto, Martinez-Carballido, Jorge F.
Format Conference Proceeding
LanguageEnglish
Published IEEE 24.07.2023
Subjects
Calcium
Computational modeling
endothelial injury
in situ endothelial cells
Instruments
intracellular calcium
MIMICs
phantom
Phantoms
Protocols
synthetic image
Wounds
Online AccessGet full text

Cover

More Information
Summary:Synthetic image generation (SIG) has played a relevant role in the cell physiology field to validate image processing algorithms, and its usage may be extended to simulator development, helping to consolidate knowledge, reduce the usage of experimental animals, and allow research progress without highly specialized and costly instrumentation. Endothelial cells (EC) control critical functions of the entire cardiovascular system through a complex mixture of sensing and communication capabilities, and there is vast evidence that ECs functionality is highly governed by their intracellular calcium concentration ([Ca 2+ ]i), thus its impairment may lead to life-threatening complications. We present an approach for SIG of ECs towards a simulator that mimics the variability of [Ca 2+ ]i at in in situ endothelium injury scenarios. Our SIG is composed of three sequential stages: 1) feature extraction measurements from in situ ECs; 2) randomized phantom generation; and 3) behavioral model assignment. Three different instances of behavioral models are shown. The embryonic approach of our SIG was demonstrated to be functional and suitable for any given behavioral model representation.
DOI:10.1109/CSCE60160.2023.00090