Does a CBIR system really impact decisions of physicians in a clinical environment?

Ponciano-Silva, Marcelo; Souza, Juliana P.; Bugatti, Pedro Henrique; Bedo, Marcos V. N.; Kaster, Daniel S.; Braga, Rosana Teresinha Vaccare; Bellucci, Ângela Delete; Azevedo-Marques, Paulo Mazzoncini de; Traina, Caetano; Traina, Agma J. M.

doi:10.1109/cbms.2013.6627762

Cited by 14 publications

(14 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, those data as a whole are barely used and the PACS has all of its greatness reduced just to per patient system [2]. Moreover, PACS require improvements [3] to aid in decision making and support queries according to the image content [4].…”

Section: Introduction and Related Workmentioning

confidence: 99%

See 1 more Smart Citation

MedInject: A General-Purpose Information Retrieval Framework Applied in a Medical Context

Carvalho

Seraphim

et al. 2014

2014 IEEE 27th International Symposium on Computer-Based Medical Systems

Self Cite

View full text Add to dashboard Cite

The continuous improvement of medical software and instrumentation have contributed to generate large amounts of medical image data. Thus, plenty of ContentBased Image Retrieval systems have emerged in order to index and retrieve images according to similarity criteria. Some of those systems are applied in very specific domains, such as mammography, lung or spine exams. Others, however, are general-purpose applications that can be adopted in a medical environment. In such context, we realized those specific systems could benefit from the facilities brought by generic frameworks and propose our solution. This article presents a novel information retrieval core framework that performs both indexing and similarity search operations over medical image data sets. The framework follows a modular architecture based on Design Patterns and can be easily extended, allowing to other system developers to take advantages of its functions by using the provided interfaces. We performed extensive experiments evaluating several of its properties and target abstractions using medical real data, and show that it allows the implementation to achieve proper similarity retrieval and significant performance improvements in relation to the existing alternatives.

show abstract

Section: Introduction and Related Workmentioning

confidence: 99%

“…Those pieces of information are collected by using feature extractors and must describe relevant aspects of the data. The most employed features considered in a clinical environment are color, shape and texture descriptors, as well as the extractors built for a specific domain [4].…”

Section: Introductionmentioning

confidence: 99%

MedInject: A General-Purpose Information Retrieval Framework Applied in a Medical Context

Carvalho

Seraphim

et al. 2014

2014 IEEE 27th International Symposium on Computer-Based Medical Systems

Self Cite

View full text Add to dashboard Cite

show abstract

“…Manually setting these parameters is a heavy burden for experts and, sometimes, unfeasible. This factor contributes to the lack of CBMIR use in clinical practice [13,14]. Therefore, to increase CBMIR accuracy and usability automatically endowing a CBMIR with proper perceptual parameters is a topic of major interest [7,13].…”

Section: Introductionmentioning

confidence: 99%

Endowing a Content-Based Medical Image Retrieval System with Perceptual Similarity Using Ensemble Strategy

et al. 2015

Self Cite

View full text Add to dashboard Cite

Content-based medical image retrieval (CBMIR) is a powerful resource to improve differential computer-aided diagnosis. The major problem with CBMIR applications is the semantic gap, a situation in which the system does not follow the users' sense of similarity. This gap can be bridged by the adequate modeling of similarity queries, which ultimately depends on the combination of feature extractor methods and distance functions. In this study, such combinations are referred to as perceptual parameters, as they impact on how images are compared. In a CBMIR, the perceptual parameters must be manually set by the users, which imposes a heavy burden on the specialists; otherwise, the system will follow a predefined sense of similarity. This paper presents a novel approach to endow a CBMIR with a proper sense of similarity, in which the system defines the perceptual parameter depending on the query element. The method employs ensemble strategy, where an extreme learning machine acts as a meta-learner and identifies the most suitable perceptual parameter according to a given query image. This parameter defines the search space for the similarity query that retrieves the most similar images. An instance-based learning classifier labels the query image following the query result set. As the concept implementation, we integrated the approach into a mammogram CBMIR. For each query image, the resulting tool provided a complete second opinion, including lesion class, system certainty degree, and set of most similar images. Extensive experiments on a large mammogram dataset showed that our proposal achieved a hit ratio up to 10% higher than the traditional CBMIR approach without requiring external parameters from the users. Our database-driven solution was also up to 25% faster than content retrieval traditional approaches.

show abstract

“…O sistema CBMIR possui uma interface para coletar as informações solicitadas pelo especialista (Ponciano- Silva et al, 2013). Uma vez que estas estão definidas, ele submete uma consulta com SQL estendido para o shell, que então retorna com o conjunto solução.…”

Section: Considerações Finaisunclassified

“…A base de dados já traz as classificações entre (1) massa e (2) calcificação, além das classificações BI-RADS (Ponciano- Silva et al, 2013) de composição e malignidade. O conjunto resultante consiste de 2892 imagens.…”

Section: Dados De Mamografiaunclassified

Incluindo funções de distância e extratores de características para suporte a consultas por similaridade

Bêdo¹

View full text Add to dashboard Cite

Does a CBIR system really impact decisions of physicians in a clinical environment?

Cited by 14 publications

References 14 publications

MedInject: A General-Purpose Information Retrieval Framework Applied in a Medical Context

MedInject: A General-Purpose Information Retrieval Framework Applied in a Medical Context

Endowing a Content-Based Medical Image Retrieval System with Perceptual Similarity Using Ensemble Strategy

Incluindo funções de distância e extratores de características para suporte a consultas por similaridade

Contact Info

Product

Resources

About