A Modified HSIFT Descriptor for Medical Image Classification of Anatomy Objects

Modeling low level features to high level semantics in medical imaging is an important aspect in filtering anatomy objects. Bag of Visual Words (BOVW) representations have been proven effective to model these low level features to mid level representations. Convolutional neural nets are learning sys...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Sumeer Ahmad Khan, Yonis Gulzar, Sherzod Turaev, Young Suet Peng
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	medical images classification feature extraction deep learning feature discrimination Mathematics QA1-939
Acceso en línea:	https://doaj.org/article/4655ef7b4db9439bb9576e0dd8a3085d
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:4655ef7b4db9439bb9576e0dd8a3085d
record_format	dspace
spelling	oai:doaj.org-article:4655ef7b4db9439bb9576e0dd8a3085d2021-11-25T19:05:48ZA Modified HSIFT Descriptor for Medical Image Classification of Anatomy Objects10.3390/sym131119872073-8994https://doaj.org/article/4655ef7b4db9439bb9576e0dd8a3085d2021-10-01T00:00:00Zhttps://www.mdpi.com/2073-8994/13/11/1987https://doaj.org/toc/2073-8994Modeling low level features to high level semantics in medical imaging is an important aspect in filtering anatomy objects. Bag of Visual Words (BOVW) representations have been proven effective to model these low level features to mid level representations. Convolutional neural nets are learning systems that can automatically extract high-quality representations from raw images. However, their deployment in the medical field is still a bit challenging due to the lack of training data. In this paper, learned features that are obtained by training convolutional neural networks are compared with our proposed hand-crafted HSIFT features. The HSIFT feature is a symmetric fusion of a Harris corner detector and the Scale Invariance Transform process (SIFT) with BOVW representation. The SIFT process is enhanced as well as the classification technique by adopting bagging with a surrogate split method. Quantitative evaluation shows that our proposed hand-crafted HSIFT feature outperforms the learned features from convolutional neural networks in discriminating anatomy image classes.Sumeer Ahmad KhanYonis GulzarSherzod TuraevYoung Suet PengMDPI AGarticlemedical imagesclassificationfeature extractiondeep learningfeature discriminationMathematicsQA1-939ENSymmetry, Vol 13, Iss 1987, p 1987 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	medical images classification feature extraction deep learning feature discrimination Mathematics QA1-939
spellingShingle	medical images classification feature extraction deep learning feature discrimination Mathematics QA1-939 Sumeer Ahmad Khan Yonis Gulzar Sherzod Turaev Young Suet Peng A Modified HSIFT Descriptor for Medical Image Classification of Anatomy Objects
description	Modeling low level features to high level semantics in medical imaging is an important aspect in filtering anatomy objects. Bag of Visual Words (BOVW) representations have been proven effective to model these low level features to mid level representations. Convolutional neural nets are learning systems that can automatically extract high-quality representations from raw images. However, their deployment in the medical field is still a bit challenging due to the lack of training data. In this paper, learned features that are obtained by training convolutional neural networks are compared with our proposed hand-crafted HSIFT features. The HSIFT feature is a symmetric fusion of a Harris corner detector and the Scale Invariance Transform process (SIFT) with BOVW representation. The SIFT process is enhanced as well as the classification technique by adopting bagging with a surrogate split method. Quantitative evaluation shows that our proposed hand-crafted HSIFT feature outperforms the learned features from convolutional neural networks in discriminating anatomy image classes.
format	article
author	Sumeer Ahmad Khan Yonis Gulzar Sherzod Turaev Young Suet Peng
author_facet	Sumeer Ahmad Khan Yonis Gulzar Sherzod Turaev Young Suet Peng
author_sort	Sumeer Ahmad Khan
title	A Modified HSIFT Descriptor for Medical Image Classification of Anatomy Objects
title_short	A Modified HSIFT Descriptor for Medical Image Classification of Anatomy Objects
title_full	A Modified HSIFT Descriptor for Medical Image Classification of Anatomy Objects
title_fullStr	A Modified HSIFT Descriptor for Medical Image Classification of Anatomy Objects
title_full_unstemmed	A Modified HSIFT Descriptor for Medical Image Classification of Anatomy Objects
title_sort	modified hsift descriptor for medical image classification of anatomy objects
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/4655ef7b4db9439bb9576e0dd8a3085d
work_keys_str_mv	AT sumeerahmadkhan amodifiedhsiftdescriptorformedicalimageclassificationofanatomyobjects AT yonisgulzar amodifiedhsiftdescriptorformedicalimageclassificationofanatomyobjects AT sherzodturaev amodifiedhsiftdescriptorformedicalimageclassificationofanatomyobjects AT youngsuetpeng amodifiedhsiftdescriptorformedicalimageclassificationofanatomyobjects AT sumeerahmadkhan modifiedhsiftdescriptorformedicalimageclassificationofanatomyobjects AT yonisgulzar modifiedhsiftdescriptorformedicalimageclassificationofanatomyobjects AT sherzodturaev modifiedhsiftdescriptorformedicalimageclassificationofanatomyobjects AT youngsuetpeng modifiedhsiftdescriptorformedicalimageclassificationofanatomyobjects
_version_	1718410274413740032

A Modified HSIFT Descriptor for Medical Image Classification of Anatomy Objects

Ejemplares similares