Medical Image Data Upscaling with Generative Adversarial Networks

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F62690094%3A18450%2F20%3A50017071" target="_blank" >RIV/62690094:18450/20:50017071 - isvavai.cz</a>

Výsledek na webu

<a href="https://link.springer.com/chapter/10.1007%2F978-3-030-45385-5_66" target="_blank" >https://link.springer.com/chapter/10.1007%2F978-3-030-45385-5_66</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-030-45385-5_66" target="_blank" >10.1007/978-3-030-45385-5_66</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Medical Image Data Upscaling with Generative Adversarial Networks

Popis výsledku v původním jazyce

Super-resolution is one of the frequently investigated methods of image processing. The quality of the results is a constant problem in the methods used to obtain high resolution images. Interpolation-based methods have blurry output problems, while non-interpolation methods require a lot of training data and high computing power. In this paper, we present a supervised generative adversarial network system that accurately generates high resolution images from a low resolution input while maintaining pathological invariance. The proposed solution is optimized for small sets of input data. Compared to existing models, our network also provides faster learning. Another advantage of our approach is its versatility for various types of medical imaging methods. We used peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) as the output image quality evaluation method. The results of our test show an improvement of 5.76% compared to optimizer Adam used in the original paper [10]. For faster training of the neural network model, calculations on the graphic card with the CUDA architecture were used. © Springer Nature Switzerland AG 2020.

Název v anglickém jazyce

Medical Image Data Upscaling with Generative Adversarial Networks

Popis výsledku anglicky

Super-resolution is one of the frequently investigated methods of image processing. The quality of the results is a constant problem in the methods used to obtain high resolution images. Interpolation-based methods have blurry output problems, while non-interpolation methods require a lot of training data and high computing power. In this paper, we present a supervised generative adversarial network system that accurately generates high resolution images from a low resolution input while maintaining pathological invariance. The proposed solution is optimized for small sets of input data. Compared to existing models, our network also provides faster learning. Another advantage of our approach is its versatility for various types of medical imaging methods. We used peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) as the output image quality evaluation method. The results of our test show an improvement of 5.76% compared to optimizer Adam used in the original paper [10]. For faster training of the neural network model, calculations on the graphic card with the CUDA architecture were used. © Springer Nature Switzerland AG 2020.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

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

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2020

Kód důvěrnosti údajů

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

Název statě ve sborníku

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ISBN

978-3-030-45384-8

ISSN

0302-9743

e-ISSN

—

Počet stran výsledku

11

Strana od-do

739-749

Název nakladatele

Springer

Místo vydání

Cham

Místo konání akce

Granada, Spain

Datum konání akce

6. 5. 2020

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—