pone.0281236.g001.tif (982.38 kB)

Procedure steps of the proposed approach.

figure

posted on 2023-02-06, 18:38 authored by Gizem Dursun, Dunja Bijelić, Neşe Ayşit, Burcu Kurt Vatandaşlar, Lidija Radenović, Abdulkerim Çapar, Bilal Ersen Kerman, Pavle R. Andjus, Andrej Korenić, Ufuk Özkaya

The entire Ca²⁺ time-lapse imaging recordings (videos) were projected into a single image using different projection methods. Segmentation was performed by using a multi-level thresholding, and obtained binary mask was compared to the ground truths (ROIs marked by researcher). Finally, a new matrix called feature image was generated from extracted traces and used for assessing classification accuracy of various classifiers.

History

Usage metrics

PLOS ONE

Categories

Keywords

three main steps tested threshold levels namely maximum intensity lapse image recordings encompassed cell somas biomedical signals obtained achieving better scores 14 %, interestingly automated screening system crafted training parameters control experimental groups automated screening systems manually annotating data linear scaling projection div >< p 2 +classification accuracy reached classification procedure based 33 imaging videos assessing segmentation parameters classification accuracy training purposes healthcare systems control vs based methods based approach classification performance automated analysis various classifiers standard deviation series ).segmentation results segmentation f projection methods pixels within necessarily correspond mean value machine learning level thresholding feature images essential part crafting features combined segmentation calcium imaging autonomous tool 80 throughout 21 belonged

Licence

Exports

RefWorks

BibTeX

Ref. manager

Endnote

DataCite

NLM

DC