Discriminative cue integration for medical image annotation

Tommasi, Tatiana; Orabona, Francesco; Caputo, Barbara

doi:10.1016/j.patrec.2008.03.009

Cited by 73 publications

(47 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In only one case there was no significant difference between G-DAS and low-level integration1 however, SVM-DAS still performed better than the other solutions. This is in agreement with the results reported by Tommasi et al (2008) and Nilsback and Caputo (2004) and can be explained by greater robustness of the high-level methods to noisy cues or sensory channels and the ability of different classifiers to adapt to the characteristics of each single cue.…”

Section: Analysis Of Cue Integration Schemessupporting

confidence: 92%

Multi-modal Semantic Place Classification

Pronobis

Mozos

Caputo

et al. 2009

The International Journal of Robotics Research

139

106

View full text Add to dashboard Cite

show abstract

Section: Analysis Of Cue Integration Schemessupporting

confidence: 92%

Multi-modal Semantic Place Classification

Pronobis

Mozos

Caputo

et al. 2009

The International Journal of Robotics Research

139

106

View full text Add to dashboard Cite

show abstract

“…Based on the results shown in Table 1, the Bag of Words (BoW) and Support Vector Machines (SVM) [8] are used as main image representation technique and classifier, respectively. In the past years, the BoW was successfully employed in various medical image retrieval and classification tasks [9][10][11][12][13][14][15][16][17]. Among other classifiers, SVMs have shown a better generalization performance in medical domain compared with other classification techniques.…”

Section: Methodsmentioning

confidence: 99%

Automatic Detection of Biomedical Compound Figure using Bag of Words

2017

IJCCIE

View full text Add to dashboard Cite

Biomedical Figures in medical articles contain a largevariety of information such as text, illustrations, and images produced from various modalities. These figures represent an important part of the knowledge stored in medical articles and can often be reused for learning, research, and clinical decision support. Exploiting these figures as part of a query in automatic retrieval of medical articles has become an increasingly important and active research area. The major challenge is that almost half of the biomedical figures in medical articles are compound figures, meaning that they contain more than a single figure type. These compound figures then need to be separated into sub-figures in order to be used as query image to retrieve similar cases or articles. Prior to this, a major step is to detect a compound figure. In this paper, a classification algorithm is proposed for compound figure detection. A Bag of Words (BoW) is employed as image representation technique followed by a Support Vector Machine (SVM) classifier to generate a classification model. This model is able to automatically detect if an image is compound figure. The best accuracy rate obtained by this model is 93.5%.

show abstract

“…Modified SIFT features have been applied to this problem in the 2-D case [32]; however, one extension would be to apply this to libraries of higher-dimensional images. …”

Section: B Future Workmentioning

confidence: 99%

$n$-SIFT: $n$-Dimensional Scale Invariant Feature Transform

Cheung

Hamarneh

2009

IEEE Trans. on Image Process.

159

View full text Add to dashboard Cite

Abstract-We propose the -dimensional scale invariant feature transform ( -SIFT) method for extracting and matching salient features from scalar images of arbitrary dimensionality, and compare this method's performance to other related features. The proposed features extend the concepts used for 2-D scalar images in the computer vision SIFT technique for extracting and matching distinctive scale invariant features. We apply the features to images of arbitrary dimensionality through the use of hyperspherical coordinates for gradients and multidimensional histograms to create the feature vectors. We analyze the performance of a fully automated multimodal medical image matching technique based on these features, and successfully apply the technique to determine accurate feature point correspondence between pairs of 3-D MRI images and dynamic CT data.Index Terms-Biomedical image processing, difference of Gaussian, feature extraction, image matching, medical images, scale invariant feature transform (SIFT).

show abstract

Discriminative cue integration for medical image annotation

Cited by 73 publications

References 32 publications

Multi-modal Semantic Place Classification

Multi-modal Semantic Place Classification

Automatic Detection of Biomedical Compound Figure using Bag of Words

$n$-SIFT: $n$-Dimensional Scale Invariant Feature Transform

Contact Info

Product

Resources

About