Modelling semantics from image data: opportunities from LIDC

Raicu, Daniela; Varutbangkul, Ekarin; Furst, Jacob; Armato, Samuel G.

doi:10.1504/ijbet.2010.029653

Cited by 15 publications

(14 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Reeves et al 1 found very high interobserver variation in lung nodule boundaries marked by radiologists. Similarly, in our previous work, 2 we showed high uncertainty and low levels of agreement between radiologist annotations when attempting to map semantic characteristics to lung nodule image content. Ochs et al 3 showed the importance of enforcing agreement between radiologists when creating a reference standard for computer-aided diagnosis (CAD) systems.…”

Section: Introductionsupporting

confidence: 76%

“…The particular decision rule presented in Table 4 was based on our previous research. 2 Table 5 summarizes our results; specifically, it contains all LIDC terms with matches in RadLex™. Items denoted as shared indicated that the same RadLex™ term is shared by two or more LIDC terms.…”

Section: Discussionmentioning

confidence: 99%

“…11 In comparison to radiologist ratings and annotations, which are subject to high levels of variability, lung nodule image features provide a quantitative and objective way to capture information about lung nodule images. 2 radiologists' subjective, high-level interpretations and objective, low-level image features is known as a semantic gap. 21 Figure 2 provides an example of an LIDC radiologist's ratings for an image and associated low-level image features captured from the same image based on nodule boundaries marked by the radiologists.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Mapping LIDC, RadLex™, and Lung Nodule Image Features

et al. 2010

Self Cite

View full text Add to dashboard Cite

Ideally, an image should be reported and interpreted in the same way (e.g., the same perceived likelihood of malignancy) or similarly by any two radiologists; however, as much research has demonstrated, this is not often the case. Various efforts have made an attempt at tackling the problem of reducing the variability in radiologists’ interpretations of images. The Lung Image Database Consortium (LIDC) has provided a database of lung nodule images and associated radiologist ratings in an effort to provide images to aid in the analysis of computer-aided tools. Likewise, the Radiological Society of North America has developed a radiological lexicon called RadLex. As such, the goal of this paper is to investigate the feasibility of associating LIDC characteristics and terminology with RadLex terminology. If matches between LIDC characteristics and RadLex terms are found, probabilistic models based on image features may be used as decision-based rules to predict if an image or lung nodule could be characterized or classified as an associated RadLex term. The results of this study were matches for 25 (74%) out of 34 LIDC terms in RadLex. This suggests that LIDC characteristics and associated rating terminology may be better conceptualized or reduced to produce even more matches with RadLex. Ultimately, the goal is to identify and establish a more standardized rating system and terminology to reduce the subjective variability between radiologist annotations. A standardized rating system can then be utilized by future researchers to develop automatic annotation models and tools for computer-aided decision systems.

show abstract

Section: Introductionsupporting

confidence: 76%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Mapping LIDC, RadLex™, and Lung Nodule Image Features

et al. 2010

Self Cite

View full text Add to dashboard Cite

show abstract

“…Up to four radiologists marked these images using ratings on a scale of 1 to 5 to assess nine semantic characteristics: lobulation, spiculation, sphericity, calcification, texture, internal structure, malignancy, margin, subtlety. Internal structure and calcification are not considered because radiologists gave each image the same rating, 'soft tissue' and 'no calcification' respectively [4]. A set of 64 image features contained in 4 categories (shape, size, intensity, and texture) are extracted from each image ( Table 1).…”

Section: Datasetmentioning

confidence: 99%

“…Our previous research [4] on building CAD systems has focused on using extracted images features to predict radiologist ratings of semantic characteristics from the LIDC [5]. In neither case do these CAD systems take into account the differences in CT scanners.…”

Section: Introductionmentioning

confidence: 99%

A statistical analysis of the effects of CT acquisition parameters on low-level features extracted from CT images of the lung

Wantroba

Raicu

Furst

2009

2009 16th IEEE International Conference on Image Processing (ICIP)

Self Cite

View full text Add to dashboard Cite

RadSem: Semantic Annotation and Retrieval for Medical Images

Möller

Regel

Sintek

2009

Lecture Notes in Computer Science

View full text Add to dashboard Cite

We present a tool for semantic medical image annotation and retrieval. It leverages the MEDICO ontology which covers formal background information from various biomedical ontologies such as the Foundational Model of Anatomy (FMA), terminologies like ICD-10 and RadLex and covers various aspects of clinical procedures. This ontology is used during several steps of annotation and retrieval: (1) We developed an ontology-driven metadata extractor for the medical image format DI-COM. Its output contains, e. g., person name, age, image acquisition parameters, body region, etc. (2) The output from (1) is used to simplify the manual annotation by providing intuitive visualizations and to provide a preselected subset of annotation concepts. Furthermore, the extracted metadata is linked together with anatomical annotations and clinical findings to generate a unified view of a patient's medical history. (3) On the search side we perform query expansion based on the structure of the medical ontologies. (4) Our ontology for clinical data management allows us to link and combine patients, medical images and annotations together in a comprehensive result list. (5) The medical annotations are further extended by links to external sources like Wikipedia to provide additional information. 1 1 This research has been supported in part by the THESEUS Program in the MEDICO Project, which is funded by the German Federal Ministry of Economics and Technology under the grant number 01MQ07016. The responsibility for this publication lies with the authors.

show abstract

Modelling semantics from image data: opportunities from LIDC

Cited by 15 publications

References 54 publications

Mapping LIDC, RadLex™, and Lung Nodule Image Features

Mapping LIDC, RadLex™, and Lung Nodule Image Features

A statistical analysis of the effects of CT acquisition parameters on low-level features extracted from CT images of the lung

RadSem: Semantic Annotation and Retrieval for Medical Images

Contact Info

Product

Resources

About