Initialization of active contours for segmentation of breast cancer via fusion of ultrasound, Doppler, and elasticity images

Keatmanee, Chadaporn; Chaumrattanakul, Utairat; Kotani, Kazuhiko; Makhanov, Stanislav S.

doi:10.1016/j.ultras.2017.12.008

Cited by 19 publications

(12 citation statements)

References 80 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…According to the advantages of combining conventional US, elasticity, and Doppler images in breast cancer diagnosis, therefore, this fusion idea was applied not only in image segmentation (Lui et al, 2006, Keatmanee et al, 2019 but also in classification (Bhatia et al, 2016;Chang et al, 2016;Hung et al, 2009;Shiao et al, 2009;Chang et al, 2012;Lai et al, 2013).…”

Section: Feature Extracting With Fmmentioning

confidence: 99%

A Breast Cancer Contour Detection With Level Sets and SVM Model

Keatmanee

Thongvigitmanee

Chaumrattanakul

et al. 2022

International Journal of Knowledge and Systems Science

Self Cite

View full text Add to dashboard Cite

Level sets have been widely used to isolate features of breast tumors in ultrasound images. However, region-based methods always produce multiple contours. Since tumors are regularly undefined from the shadows and muscular regions in breast ultrasound images, computerized tumors location and arrangement is significantly difficult. Therefore, the authors introduce a breast cancer contour detection model using support vector machine (SVM) as a binary classification. Features of the binary SVM model were extracted from level sets and FM method (the fusion of ultrasound, elasticity, and Doppler images). The model was accurately able to predict a correct breast tumor contour from false contours which were segmented by region-based level sets. The proposed method was evaluated on 60 datasets collected by professional radiologists at the Thammasat University Hospital of Thailand. From the experimental results, the breast cancer contours were detected correctly with high accuracy. The percentage of correct detection was 93%.

show abstract

Section: Feature Extracting With Fmmentioning

confidence: 99%

A Breast Cancer Contour Detection With Level Sets and SVM Model

Keatmanee

Thongvigitmanee

Chaumrattanakul

et al. 2022

International Journal of Knowledge and Systems Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…Für die Segmentierung, ein entscheidender Schritt für die Leistungsfähigkeit der KI, wurde der Einsatz der Elastografie neu vorgeschlagen. Ein Algorithmus zur automatischen Segmentierung von Brusttumoren wurde durch Bildfusion von B-Modus, Power-Doppler-und SWE-Elastografie vorgeschlagen [30]. In einer retrospektiven Studie wurden SWEund B-Mode-Bilder segmentiert, analysiert und die Daten wurden in einen Lern-und Klassifizierungs-Algorithmus eingegeben.…”

Section: Elastografie In Der Brustbildgebungunclassified

“…New use of elastography has been proposed for segmentation, which is a crucial step for AI performance. An algorithm for automatic segmentation of breast tumors was proposed by fusing B-mode, Power Doppler and SWE elastography images [30]. In a retrospective study, SWE and B-mode images were segmented and analyzed, and data were fed into an algorithm for learning and classification.…”

mentioning

confidence: 99%

Elastography in Breast Imaging

Carlsen¹,

Hansen²,

Ewertsen³

et al. 2019

Ultraschall in Med

View full text Add to dashboard Cite

“…The contour active method is not accurate if the initial initialization is incorrect (far from the target limit). So, the literature [13] proposed a new initialization method based on image fusion. Segmentation of a pair of images together is called co-segmentation.…”

Section: Introductionmentioning

confidence: 99%

The Use of Pre and Post Processing to Enhance Mandible Segmentation using Active Contours on Dental Panoramic Radiography Images

Nafi’iyah

Fatichah

Astuti

et al. 2020

2020 3rd International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)

View full text Add to dashboard Cite

Mandibular segmentation is indispensable to support the automation of the gender detection system based on the dental panoramic radiography image. However, the dental panoramic radiography image has low image contrast, the gray intensity value inhomogeneous, and the gray intensity value between the teeth and mandibular bone is almost indistinguishable. So, a good segmentation method is required to separate the mandible and teeth properly. This study aims to analyze the effect of the use of preprocessing and post-processing to enhance mandible segmentation on dental panoramic radiography images properly. In the preprocessing, we use contrast enhancement and Gaussian filters to make the mandibular area more prominent. Meanwhile, in the post-processing, we use erosion and opening morphology to remove the tooth area attached to the mandible. The mandibular segmentation uses the Active Contours method with predefined contour initialization. The dataset used is 86 dental panoramic radiographic images and the segmentation evaluation method uses Jaccard similarity. The experimental results show that the mandibular segmentation with preprocessing and postprocessing obtain Jaccard similarity values are 0.31 and 0.34, on average. Meanwhile, the results of mandibular segmentation with post-processing achieve the Jaccard similarity values are 0.51 and 0.52, on average.

show abstract

Initialization of active contours for segmentation of breast cancer via fusion of ultrasound, Doppler, and elasticity images

Cited by 19 publications

References 80 publications

A Breast Cancer Contour Detection With Level Sets and SVM Model

A Breast Cancer Contour Detection With Level Sets and SVM Model

Elastography in Breast Imaging

The Use of Pre and Post Processing to Enhance Mandible Segmentation using Active Contours on Dental Panoramic Radiography Images

Contact Info

Product

Resources

About