dp-BREATH: Heat maps and probabilistic classification assisting the analysis of abnormal lung regions

Cazzolato, Mirela T.; Scabora, Lucas C.; Nesso, Marcos R.; Milano-Oliveira, Luis F.; Costa, Alceu Ferraz; Kaster, Daniel S.; Koenigkam-Santos, Marcel; Azevedo-Marques, Paulo Mazzoncini de; Traina, Caetano; Traina, Agma J. M.

doi:10.1016/j.cmpb.2019.01.014

Cited by 7 publications

(10 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The semi-automatic segmentation can be used in several analysis (Figure 1). Quantitative measures can be extracted, such as semantic and agnostic features (8), consequently, machine learning techniques can be ap- plied for the classification of a given anomaly (9,10,11) or for Content-Based Image Retrieval (CBIR) (12,13). Interactive segmentation tools can be meaningful during the training and education of new radiologists (14).…”

Section: Introductionmentioning

confidence: 99%

3DBGrowth: Volumetric Vertebrae Segmentation and Reconstruction in Magnetic Resonance Imaging

Ramos

Cazzolato

Faiçal

et al. 2019

2019 IEEE 32nd International Symposium on Computer-Based Medical Systems (CBMS)

Self Cite

View full text Add to dashboard Cite

Segmentation of medical images is critical for making several processes of analysis and classification more reliable. With the growing number of people presenting back pain and related problems, the semi-automatic segmentation and 3D reconstruction of vertebral bodies became even more important to support decision making. A 3D reconstruction allows a fast and objective analysis of each vertebrae condition, which may play a major role in surgical planning and evaluation of suitable treatments. In this paper, we propose 3DBGrowth, which develops a 3D reconstruction over the efficient Balanced Growth method for 2D images. We also take advantage of the slope coefficient from the annotation time to reduce the total number of annotated slices, reducing the time spent on manual annotation. We show experimental results on a representative dataset with 17 MRI exams demonstrating that our approach significantly outperforms the competitors and, on average, only 37% of the total slices with vertebral body content must be annotated without losing performance/accuracy. Compared to the state-of-the-art methods, we have achieved a Dice Score gain of over 5% with comparable processing time. Moreover, 3DBGrowth works well with imprecise seed points, which reduces the time spent on manual annotation by the specialist.

show abstract

Section: Introductionmentioning

confidence: 99%

3DBGrowth: Volumetric Vertebrae Segmentation and Reconstruction in Magnetic Resonance Imaging

Ramos

Cazzolato

Faiçal

et al. 2019

2019 IEEE 32nd International Symposium on Computer-Based Medical Systems (CBMS)

Self Cite

View full text Add to dashboard Cite

show abstract

“…The works and (CAZZOLATO et al, 2019) report BREATH, dp-BREATH, and all related results regarding this contribution.…”

Section: Contribution 2 the Breath And Dp-breath Methodsmentioning

confidence: 98%

“…In this chapter, we approach the second research problem of this Ph.D. research, aimed at performing image mining of medical information considering the embedded uncertainty. The corresponding contributions were published, and are available in the works and (CAZZOLATO et al, 2019).…”

Section: Chapter 4 Probabilistic Region-based Image Classification Anmentioning

confidence: 99%

Conquering knowledge from images: improving image mining with region-based analysis and associated information

Cazzolato¹

View full text Add to dashboard Cite

show abstract

“…A intensidade é calculada a partir da probabilidade de ocorrer uma anormalidade pulmonar na região. Essas imagens adicionais foram obtidas utilizando o método dp-BREATH (CAZZOLATO et al, 2017;CAZZOLATO et al, 2019). A Figura 17 apresenta exemplos das imagens utilizadas, contendo (i) a imagem original, (ii) a imagem segmentada (máscara), e (iii) o mapa de calor (heatmap) gerado.…”

Section: Descrição Dos Dados E Organizaçãounclassified

Desenvolvimento de uma Técnica para Estender um SGBD Relacional com Consultas por Similaridade

Junior¹,

Roberto²

View full text Add to dashboard Cite

Database Management Systems (DBMS) based on the Relational Theory are designed to meet the needs of storing and retrieving large amounts of data. These data can be represented by numeric values, dates, and/or small strings, and are generically called "scalar data". With the evolution of information technology, it is increasingly necessary to organize, store, and retrieve other types of data. In this work, we call such data "complex data", such as images, videos, time series, and genetic sequences. Comparisons based on Identity Relations or Order Relations are useful for querying scalar data but are not suitable for complex data. For complex data, similarity queries have been the most studied option, although their availability in existing DBMS is still limited. Metric Access Methods (MAMs) usually are applied for indexing complex data to speed-up similarity queries. This Master's project aimed at incorporating MAM resources to a Relational DBMS, by proposing and implementing a technique for extending a widely used Relational DBMS. Thus, we implemented the existing Slim-Tree MAM into PostgreSQL, which is a Relational DBMS. This implementation resulted in RAFIKI, a prototype capable of outperforming its predecessor system KIARA, in terms of speed, in the task of performing similarity queries. The experimental analysis carried showed that RAFIKI is up to 6 times faster than KIARA. Further, using the proposed technique, it is possible to extend PostgreSQL to support other MAMs.

show abstract

dp-BREATH: Heat maps and probabilistic classification assisting the analysis of abnormal lung regions

Cited by 7 publications

References 15 publications

3DBGrowth: Volumetric Vertebrae Segmentation and Reconstruction in Magnetic Resonance Imaging

3DBGrowth: Volumetric Vertebrae Segmentation and Reconstruction in Magnetic Resonance Imaging

Conquering knowledge from images: improving image mining with region-based analysis and associated information

Desenvolvimento de uma Técnica para Estender um SGBD Relacional com Consultas por Similaridade

Contact Info

Product

Resources

About