Denoise pre-training for segmentation neural networks

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26220%2F19%3APU132911" target="_blank" >RIV/00216305:26220/19:PU132911 - isvavai.cz</a>

Result on the web

—

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

Denoise pre-training for segmentation neural networks

Original language description

This paper proposes a method for pre-training segmentation neural networks on small datasets using unlabelled training data with added noise. The pre-training process helps the network with initial better weights settings for the training itself and also augments the training dataset when dealing with small labelled datasets especially in medical imaging. The experiment comparing results of pre-trained and not pre-trained networks on MRI brain segmentation task has shown that the denoise pre-training helps the network with faster training convergence without overfitting and achieving better results in all compared metrics even on very small datasets.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

—

Continuities

S - Specificky vyzkum na vysokych skolach

Publication year

2019

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

Proceedings of the 25th Conference STUDENT EEICT 2019

ISBN

978-80-214-5735-5

ISSN

—

e-ISSN

—

Number of pages

5

Pages from-to

739-744

Publisher name

Vysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních technologií

Place of publication

Brno

Event location

Brno

Event date

Apr 25, 2019

Type of event by nationality

CST - Celostátní akce

UT code for WoS article

—