The IRMA code for unique classification of medical images

Lehmann, Thomas; Schubert, Henning; Keysers, Daniel; Kohnen, Michael; Wein, B.

doi:10.1117/12.480677

Cited by 122 publications

(87 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…All images are described by their recording date and the body region that is visualized in the image. For precise localization, the body part terminology has been defined according to the image retrieval in medical applications (IRMA) code for medical images, i.e., a monohierarchical multiaxial classification scheme [28]. Figure 5 (left) visualizes the BLOB module integrated into the German Calciphylaxis Registry.…”

Section: Instantiation Of the Blob Modulementioning

confidence: 99%

Integrated Image Data and Medical Record Management for Rare Disease Registries. A General Framework and its Instantiation to the German Calciphylaxis Registry

et al. 2014

View full text Add to dashboard Cite

Especially for investigator-initiated research at universities and academic institutions, Internet-based rare disease registries (RDR) are required that integrate electronic data capture (EDC) with automatic image analysis or manual image annotation. We propose a modular framework merging alpha-numerical and binary data capture. In concordance with the Office of Rare Diseases Research recommendations, a requirement analysis was performed based on several RDR databases currently hosted at Uniklinik RWTH Aachen, Germany. With respect to the study management tool that is already successfully operating at the Clinical Trial Center

show abstract

Section: Instantiation Of the Blob Modulementioning

confidence: 99%

Integrated Image Data and Medical Record Management for Rare Disease Registries. A General Framework and its Instantiation to the German Calciphylaxis Registry

et al. 2014

View full text Add to dashboard Cite

show abstract

“…The aim of the automatic image annotation task is to classify images into a set of classes, according to the IRMA code [7]. The labels are hierarchical therefore, errors in the annotation are counted depending on the level at which the error is done and on the number of possible choices.…”

Section: Single Cue Image Annotationmentioning

confidence: 99%

Cue Integration for Medical Image Annotation

Tommasi

Orabona

Caputo

2008

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. This paper presents the algorithms and results of our participation to the image annotation task of ImageCLEFmed 2007. We proposed a multi-cue approach where images are represented both by global and local descriptors. These cues are combined following two SVMbased strategies. The first algorithm, called Discriminative Accumulation Scheme (DAS), trains an SVM for each feature, and considers as output of each classifier the distance from the separating hyperplane. The final decision is taken on a linear combination of these distances. The second algorithm, that we call Multi Cue Kernel (MCK), uses a new Mercer kernel which can accept as input different features while keeping them separated. The DAS algorithm obtained a score of 29.9, which ranked fifth among all submissions. The MCK algorithm with the one-vs-all and with the one-vs-one multiclass extensions of SVM scored respectively 26.85 and 27.54. These runs ranked first and second among all submissions.

show abstract

“…This paper describes the medical annotation task of ImageCLEF 2008 [1]. The objective of this task is to provide the IRMA (Image Retrieval in Medical Applications) code [2] for each image of a given set of previously unseen medical (radiological) images. 12,076 classified training images are provided to be used in any way to train a classifier.…”

Section: Introductionmentioning

confidence: 99%

“…The results of the classification step can be used for multilingual image annotations as well as for DICOM standard header corrections. According to the IRMA code [2], a total of 197 classes are defined. The IRMA coding system consists of four axes with three to four positions, each in {0,…,9,a,…,z}, where "0" denotes "unspecified" to determine the end of a path along an axis:…”

Section: Introductionmentioning

confidence: 99%

Hierarchical annotation of medical images

Dimitrovski

Kocev

Loskovska³

et al. 2011

Pattern Recognition

145

View full text Add to dashboard Cite

In this paper, we describe an approach for the automatic medical annotation task of the 2008 CLEF cross-language image retrieval campaign (ImageCLEF). The data comprise 12076 fully annotated images according to the IRMA code. This work is focused on the process of feature extraction from images and hierarchical multi-label classification. To extract features from the images we used a technique called: local distribution of edges. With this techniques each image was described with 80 variables. The goal of the classification task was to classify an image according to the IRMA code. The IRMA code is organized hierarchically. Hence, as classifer we selected an extension of the predictive clustering trees (PCTs) that is able to handle this type of data. Further more, we constructed ensembles (Bagging and Random Forests) that use PCTs as base classifiers.

show abstract

The IRMA code for unique classification of medical images

Cited by 122 publications

References 11 publications

Integrated Image Data and Medical Record Management for Rare Disease Registries. A General Framework and its Instantiation to the German Calciphylaxis Registry

Integrated Image Data and Medical Record Management for Rare Disease Registries. A General Framework and its Instantiation to the German Calciphylaxis Registry

Cue Integration for Medical Image Annotation

Hierarchical annotation of medical images

Contact Info

Product

Resources

About