Document type
Journal articles
Document subtype
Full paper
Title
Case study – apply a deep-learning algorithm to exosomes detection with online resources
Participants in the publication
M. C. Proença (Author)
MARE – Marine and Environmental Sciences Centre, Universidade de Lisboa
Dep. Física
A. P. Alves de Matos (Author)
EGAS MONIZ - COOPERATIVA DE ENSINO SUPERIOR, CRL
Summary
Abstract - exosomes are membrane vesicles part of the extracellular vesicle structures (EVs) that constitute a mode of intercellular communication. Extracellular vesicles are promising biomarkers for many diseases. Their participation in pathological processes, with relevance to non-communicable diseases (NCDs) is supported by mounting evidence and their study will surely drive new therapeutic and diagnostic strategies. Electron microscopy is currently used to identify this kind of particles in transmission electron microscopy images, but their morphology can span a wide range of vesicle structures and sizes. This preliminary work shows how a recent deep-learning algorithm available online can be applied to exosome images to localize all the instances of the objects of interest, in any scale and several backgrounds, allowing an in-depth analysis of their composition, and adding a novel instrument for their morphological classification.
Date of Publication
2021-12-24
Institution
FACULDADE DE CIÊNCIAS DA UNIVERSIDADE DE LISBOA
Where published
Annals of Antivirals and Antiretrovirals
Publication Identifiers
ISSN - 2692-4625
Publisher
Peertechz Publications Private Limited
Number of pages
3
Starting page
033
Last page
035
Document Identifiers
DOI -
https://doi.org/10.17352/aaa.000014
URL -
http://dx.doi.org/10.17352/aaa.000014
Keywords
Exosomes; TEM images; Deep-learning algorithms; Detection of small objects