Naslov
Medical data annotation and json to dataset conversion using LabelMe and Python

Autori
Severinski, Karlo ; Cvija, Tajana

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proceedings of the International Scientific Student Conference RI-STEM-2021 / Lorencin, Ivan ; Baressi Šegota, Sandi - Rijeka, 2021, 27-32

ISBN
978-953-8246-22-7

Skup
International Student Scientific Conference (Ri-STEM 2021)

Mjesto i datum
Rijeka, Hrvatska, 10.06.2021. - 11.06.2021

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Annotation ; artificial intelligence ; dataset ; json ; LabelMe ; Python ; semantic segmentation ; U-net

Sažetak
This paper will discuss the use of the LabelMe tool for annotation of radiological data. Annotations will be made of the prostate and the bladder that will thus be prepared for the use of U-net semantic segmentation. The goal of this paper is to show how to prepare annotated data for processing using U-net and their conversion from jpeg format to json data and the conversion of json data into an annotation mask for later use in semantic segmentation. The annotation tool used is the already mentioned LabelMe as part of the anaconda package and the programming language Python as support for method implementation. The obtained results are presented in detail in the paper. The results provide the necessary data for later use in semantic segmentation.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika, Interdisciplinarne tehničke znanosti, Temeljne medicinske znanosti

POVEZANOST RADA