Accurate Landmark Localization for Medical Images Using Perturbations

Recently, various studies have been proposed to learn the rich representations of images during deep learning. In particular, the perturbation method is a simple way to learn rich representations that has shown significant success. In this study, we present effective perturbation approaches for medi...

Full description

Bibliographic Details
Main Authors: Junhyeok Kang, Kanghan Oh, Il-Seok Oh
Format: Article
Language:English
Published: MDPI AG 2021-11-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/11/21/10277
_version_ 1797512813812383744
author Junhyeok Kang
Kanghan Oh
Il-Seok Oh
author_facet Junhyeok Kang
Kanghan Oh
Il-Seok Oh
author_sort Junhyeok Kang
collection DOAJ
description Recently, various studies have been proposed to learn the rich representations of images during deep learning. In particular, the perturbation method is a simple way to learn rich representations that has shown significant success. In this study, we present effective perturbation approaches for medical landmark localization. To this end, we report an extensive experiment that uses the perturbation methods of erasing, smoothing, binarization, and edge detection. The hand X-ray dataset and the ISBI 2015 Cephalometric dataset are used to evaluate the perturbation effect. The experimental results show that the perturbation method forces the network to extract richer representations of an image, leading to performance increases. Moreover, in comparison with the existing methods that lack any complex algorithmic change of network, our methods with specific perturbation methods achieve superior performance.
first_indexed 2024-03-10T06:06:58Z
format Article
id doaj.art-dcdbf490715e4247b847d7e4ef8de245
institution Directory Open Access Journal
issn 2076-3417
language English
last_indexed 2024-03-10T06:06:58Z
publishDate 2021-11-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj.art-dcdbf490715e4247b847d7e4ef8de2452023-11-22T20:30:26ZengMDPI AGApplied Sciences2076-34172021-11-0111211027710.3390/app112110277Accurate Landmark Localization for Medical Images Using PerturbationsJunhyeok Kang0Kanghan Oh1Il-Seok Oh2Division of Computer Science and Engineering, Jeonbuk National University, Jeonju 54896, KoreaDepartment of Computer and Software Engineering, Wonkwang University, Iksan 54538, KoreaDivision of Computer Science and Engineering, Jeonbuk National University, Jeonju 54896, KoreaRecently, various studies have been proposed to learn the rich representations of images during deep learning. In particular, the perturbation method is a simple way to learn rich representations that has shown significant success. In this study, we present effective perturbation approaches for medical landmark localization. To this end, we report an extensive experiment that uses the perturbation methods of erasing, smoothing, binarization, and edge detection. The hand X-ray dataset and the ISBI 2015 Cephalometric dataset are used to evaluate the perturbation effect. The experimental results show that the perturbation method forces the network to extract richer representations of an image, leading to performance increases. Moreover, in comparison with the existing methods that lack any complex algorithmic change of network, our methods with specific perturbation methods achieve superior performance.https://www.mdpi.com/2076-3417/11/21/10277artificial intelligencelandmark localizationcontext feature learningimage perturbation
spellingShingle Junhyeok Kang
Kanghan Oh
Il-Seok Oh
Accurate Landmark Localization for Medical Images Using Perturbations
Applied Sciences
artificial intelligence
landmark localization
context feature learning
image perturbation
title Accurate Landmark Localization for Medical Images Using Perturbations
title_full Accurate Landmark Localization for Medical Images Using Perturbations
title_fullStr Accurate Landmark Localization for Medical Images Using Perturbations
title_full_unstemmed Accurate Landmark Localization for Medical Images Using Perturbations
title_short Accurate Landmark Localization for Medical Images Using Perturbations
title_sort accurate landmark localization for medical images using perturbations
topic artificial intelligence
landmark localization
context feature learning
image perturbation
url https://www.mdpi.com/2076-3417/11/21/10277
work_keys_str_mv AT junhyeokkang accuratelandmarklocalizationformedicalimagesusingperturbations
AT kanghanoh accuratelandmarklocalizationformedicalimagesusingperturbations
AT ilseokoh accuratelandmarklocalizationformedicalimagesusingperturbations