2014
DOI: 10.1016/j.jbi.2014.02.018
|View full text |Cite
|
Sign up to set email alerts
|

A hierarchical knowledge-based approach for retrieving similar medical images described with semantic annotations

Abstract: Computer-assisted image retrieval applications could assist radiologist interpretations by identifying similar images in large archives as a means to providing decision support. However, the semantic gap between low-level image features and their high level semantics may impair the system performances. Indeed, it can be challenging to comprehensively characterize the images using low-level imaging features to fully capture the visual appearance of diseases on images, and recently the use of semantic terms has … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
14
0

Year Published

2014
2014
2020
2020

Publication Types

Select...
5
2

Relationship

1
6

Authors

Journals

citations
Cited by 32 publications
(14 citation statements)
references
References 54 publications
0
14
0
Order By: Relevance
“…thorax CT). The challenge of successfully detecting and selecting purely visual biomarkers for general medical retrieval is still an unsolved problem in the literature [9].…”
Section: Resultsmentioning
confidence: 99%
“…thorax CT). The challenge of successfully detecting and selecting purely visual biomarkers for general medical retrieval is still an unsolved problem in the literature [9].…”
Section: Resultsmentioning
confidence: 99%
“…The common specificity is used to consider that lower level pairs of term nodes are more similar than higher level pairs. In a previous study (Kurtz et al, 2014), we have extended this definition to normalize it and to give an equal weight to the path length and the common specificity features. The proposed definition becomes: dΘfalse(xi,xjfalse)=log0.16667em(minppathpfalse(xi,xjfalse)·CSfalse(xi,xjfalse)+γ)ωitalicnormitalicΘ where ωnormΘ=maxi,j(log0.16667em(minppathpfalse(xi,xjfalse)·CSfalse(xi,xjfalse)+γ)) is a normalization factor evaluating the maximal dissimilarity value between two terms and CS ( x i , x j ) = D c − depth ( LCA ( x i , x j )) is the common specificity of the terms.…”
Section: Methodsmentioning
confidence: 99%
“…, the dissimilarities) between the elements of the vectors had to be manually defined. Recently, this distance has been extended to the comparison of images described with high-dimensional vectors of semantic terms (Kurtz et al, 2014). In this generic extension, the dissimilarity between semantic terms can be provided by any term dissimilarity measure.…”
Section: Image Retrieval Based On Semantic Featuresmentioning
confidence: 99%
See 1 more Smart Citation
“…In this paper, we have proposed a hybrid technique which uses color, shape and texture feature of the image and use these features for classification. www.ijarai.thesai.org [40] in 2014. The semantic gap between the low-level image features and their high level semantics has always ruined the retrieval quality.…”
Section: Ontology and Image Analysismentioning
confidence: 99%