Characterization of mammographic masses based on level set segmentation with new image features and patient information

Shi, Junbo; Sahiner, Berkman; Chan, Heang Ping; Ge, Junbo; Hadjiiski, Lubomir M.; Helvie, Mark A.; Nees, Alexis V.; Wu, Yi; Wei, Jun; Zhou, Chuan; Zhang, Yiheng; Cui, Jing

doi:10.1118/1.2820630

Cited by 95 publications

(57 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We use the area overlap metric [9,10,25], the Hausdorff distance metric, and the average minimum Euclidean distance metric [26][27][28] to quantify the consistency between the segmented results and the ground truth provided by a radiologist's manually delineated outlines. The AOM metric is defined as the ratio of the intersection to the union of the two areas to be compared:…”

Section: Discussionmentioning

confidence: 99%

Marker-Controlled Watershed for Lesion Segmentation in Mammograms

Liu

Song

2011

J Digit Imaging

View full text Add to dashboard Cite

Lesion segmentation, which is a critical step in computer-aided diagnosis system, is a challenging task as lesion boundaries are usually obscured, irregular, and low contrast. In this paper, an accurate and robust algorithm for the automatic segmentation of breast lesions in mammograms is proposed. The traditional watershed transformation is applied to the smoothed (by the morphological reconstruction) morphological gradient image to obtain the lesion boundary in the belt between the internal and external markers. To automatically determine the internal and external markers, the rough region of the lesion is identified by a template matching and a thresholding method. Then, the internal marker is determined by performing a distance transform and the external marker by morphological dilation. The proposed algorithm is quantitatively compared to the dynamic programming boundary tracing method and the plane fitting and dynamic programming method on a set of 363 lesions (size range, 5-42 mm in diameter; mean, 15 mm), using the area overlap metric (AOM), Hausdorff distance (HD), and average minimum Euclidean distance (AMED). The mean ± SD of the values of AOM, HD, and AMED for our method were respectively 0.72 ± 0.13, 5.69 ± 2.85 mm, and 1.76 ± 1.04 mm, which is a better performance than two other proposed segmentation methods. The results also confirm the potential of the proposed algorithm to allow reliable segmentation and quantification of breast lesion in mammograms.

show abstract

Section: Discussionmentioning

confidence: 99%

Marker-Controlled Watershed for Lesion Segmentation in Mammograms

Liu

Song

2011

J Digit Imaging

View full text Add to dashboard Cite

show abstract

“…CAD is being explored as a way of improving the specifi city of breast lesion classifi cation without sacrifi cing sensitivity (23)(24)(25)(26)(27)(28)(29)(30)(31)(32)(33)(34)(35)(36)(37). A subset of CAD systems predicts biopsy outcomes with use of each radiologist's description of breast lesions instead of automated, computerextracted features (26)(27)(28).…”

Section: Methodsmentioning

confidence: 99%

Computer-aided Classification of Breast Masses: Performance and Interobserver Variability of Expert Radiologists versus Residents

Singh¹,

Maxwell²,

Baker³

et al. 2011

Radiology

View full text Add to dashboard Cite

Purpose:To evaluate the interobserver variability in descriptions of breast masses by dedicated breast imagers and radiology residents and determine how any differences in lesion description affect the performance of a computer-aided diagnosis (CAD) computer classifi cation system. Materials and Methods:Institutional review board approval was obtained for this HIPAA-compliant study, and the requirement to obtain informed consent was waived. Images of 50 breast lesions were individually interpreted by seven dedicated breast imagers and 10 radiology residents, yielding 850 lesion interpretations. Lesions were described with use of 11 descriptors from the Breast Imaging Reporting and Data System, and interobserver variability was calculated with the Cohen k statistic. Those 11 features were selected, along with patient age, and merged together by a linear discriminant analysis (LDA) classifi cation model trained by using 1005 previously existing cases. Variability in the recommendations of the computer model for different observers was also calculated with the Cohen k statistic. Results:A signifi cant difference was observed for six lesion features, and radiology residents had greater interobserver variability in their selection of fi ve of the six features than did dedicated breast imagers. The LDA model accurately classifi ed lesions for both sets of observers (area under the receiver operating characteristic curve = 0.94 for residents and 0.96 for dedicated imagers). Sensitivity was maintained at 100% for residents and improved from 98% to 100% for dedicated breast imagers. For residents, the computer model could potentially improve the specifi city from 20% to 40% ( P , .01) and the k value from 0.09 to 0.53 ( P , .001). For dedicated breast imagers, the computer model could increase the specifi city from 34% to 43% ( P = .16) and the k value from 0.21 to 0.61 ( P , .001). Conclusion:Among fi ndings showing a signifi cant difference, there was greater interobserver variability in lesion descriptions among residents; however, an LDA model using data from either dedicated breast imagers or residents yielded a consistently high performance in the differentiation of benign from malignant breast lesions, demonstrating potential for improving specifi city and decreasing interobserver variability in biopsy recommendations.q RSNA, 2010 1

show abstract

“…In contrast to the three approaches that we propose in this work, most of the work that can be found in literature follows the classic pattern recognition chain: segmentation of a mass from the background tissue, feature extraction and classification. Good examples are the work of Shi et al [3] who apply level-sets for the segmentation of masses and propose several features specifically designed for the task of mass characterization. Varela et al [4] have developed features that measure the degree of sharpness and microlobulation of the mass margin to improve their previously proposed CAD approach.…”

Section: Introductionmentioning

confidence: 99%

Model-Based Characterization of Mammographic Masses

Heidt

Elter

Wittenberg

et al. 2009

Bildverarbeitung Für Die Medizin 2009

View full text Add to dashboard Cite

Abstract. The discrimination of benign and malignant types of mammographic masses is a major challenge for radiologists. The classic eigenfaces method was recently adapted for the detection of masses in mammograms. In the work at hand we investigate if this method is also suited for the problem of distinguishing benign and malignant types of this mammographic lesion. We furthermore evaluate two extended versions of the eigenfaces approach (fisherface and eigenfeature regularization extraction) and compare the performance of all three methods on a public mammography database. Our results indicate that all three methods can be applied to discriminate benign and malignant types of mammographic masses. However, our ROC analysis shows that the methods still require combination with other features to allow for reliable classification.

show abstract

Characterization of mammographic masses based on level set segmentation with new image features and patient information

Cited by 95 publications

References 49 publications

Marker-Controlled Watershed for Lesion Segmentation in Mammograms

Marker-Controlled Watershed for Lesion Segmentation in Mammograms

Computer-aided Classification of Breast Masses: Performance and Interobserver Variability of Expert Radiologists versus Residents

Model-Based Characterization of Mammographic Masses

Contact Info

Product

Resources

About