Toward a standard reference database for computer-aided mammography

Oliveira, Júlia Epischina Engrácia de; Gueld, Mark Oliver; Araújo, Arnaldo de Albuquerque; Ott, Bastian; Deserno, Thomas M.

doi:10.1117/12.770325

Cited by 59 publications

(32 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Figure 5 gives an overview of the selection process. Articles were excluded mainly because they did not address CADe of masses and microcalcifications or breast cancer CADx (104), or because they did not use the DDSM for testing (47), or used less than 100 test items (23) in the evaluation studies presented. All included articles together contained 68 CADe studies and 38 CADx studies, that is a total of 106 evaluation studies.…”

Section: Literature Evaluation Resultsmentioning

confidence: 99%

“…Various publications have addressed the issue of image database creation for mammography CADe and CADx [43][44][45][46][47][48]. Well-known challenges include the design of databases, the choice of cases which must avoid biases, the acquisition of ground truth (image annotations, patient meta-data) in a satisfactory quality, the database size, and the evaluation methodology.…”

Section: Recommendations For Future Evaluation Studiesmentioning

confidence: 99%

See 1 more Smart Citation

Needs assessment for next generation computer-aided mammography reference image databases and evaluation studies

Horsch

Hapfelmeier²,

Elter

2011

Int J CARS

View full text Add to dashboard Cite

Overall, both the use of the DDSM as data source for evaluation of mammography CAD systems, and the application of statistical evaluation methods were found highly diverse. Results reported from different studies are therefore hardly comparable. Drawbacks of the DDSM (e.g. varying quality of lesion annotations) may contribute to the reasons. But larger bias seems to be caused by authors' own decisions upon study design. RECOMMENDATIONS/CONCLUSION: For future evaluation studies, we derive a set of 13 recommendations concerning the construction and usage of a test database, as well as the application of statistical evaluation methods.

show abstract

Section: Literature Evaluation Resultsmentioning

confidence: 99%

Section: Recommendations For Future Evaluation Studiesmentioning

confidence: 99%

Needs assessment for next generation computer-aided mammography reference image databases and evaluation studies

Horsch

Hapfelmeier²,

Elter

2011

Int J CARS

View full text Add to dashboard Cite

show abstract

“…The steps of the proposed method for reducing the features set are as follows: The database used to evaluate the proposed method was retrieved from the image retrieval in medical applications (IRMA) project [20] and included 3,676 mammogram patches that were categorized into three classes (normal, benign, and malignant) according to ground-truth information. The patches came from four different databases (DDSM, MIAS, LLNL, and RWTH), and each patch had a size of 128 by 128 pixels.…”

Section: Experimental Evaluation and Discussionmentioning

confidence: 99%

A novel feature reduction framework for digital mammogram image classification

Alharbi

Falzon

Kwan

2015

2015 3rd IAPR Asian Conference on Pattern Recognition (ACPR)

View full text Add to dashboard Cite

The visual similarity between normal breast tissues and abnormal lesions in digital mammogram images makes computer-aided diagnosis of breast cancer using automatically detected features a highly error-prone task. Our contribution in this paper is a novel feature reduction framework for selecting the most discriminative features that achieves both efficiency and classification accuracy. Our approach applies five individual feature-ranking methods including Fisher score, minimum redundancymaximum relevance, relief-f, sequential forward feature selection, and genetic algorithm for sorting the extracted features and selecting the features with highest ranking to setup a classifier. Our method achieves an accuracy of 94.27% and a sensitivity of 98.36% with a specificity of 99.27% on a set of 1,100 mammogram patches taken from image retrieval in medical applications database using a neural network classifier, which competes with state-ofthe-art classification accuracy 93.11%. Furthermore, we demonstrate that only 49 out of the 119 extracted features are sufficient to achieve the reported accuracy of normal vs. abnormal classification.

show abstract

“…Also, we convert the rectangular ROIs to polar coordinates to emphasize on some special patterns in mammograms. The database for validation involves a set of ROIs extracted from the IRMA version [10] the Digital Database for Screening Mammography (DDSM) database [11]. Comparison of the obtained results with pervious works illustrates the significant improvement in the performance of architectural distortion recognition.…”

mentioning

confidence: 89%

A textural approach for recognizing architectural distortion in mammograms

Mohammadi

Fatemizadeh

Sheikhzadeh

et al. 2013

2013 8th Iranian Conference on Machine Vision and Image Processing (MVIP)

View full text Add to dashboard Cite

Breast cancer is considered as the most important cause of death among women. Architectural distortions are very important signs of breast cancer and early detection of them is a rewarding work. In this paper we propose a method to recognize architectural distortion from normal parenchyma. In our proposed method, appropriate features are extracted by the analysis of oriented textures with the application of orientation component of recent the state-ofthe-art local texture descriptor called Monogenic Binary Coding (MBC). In addition, we transform Region of Interests (ROIs) to polar coordinates in order to highlight some specific patterns in mammograms. Various classifiers are used over a set of mammograms from Digital Database for Screening Mammography (DDSM). The results show that proposed method is very encouraging. The best performance achieved is 91.25% in terms of the average accuracy using the Nearest Neighbor classifier.

show abstract

Toward a standard reference database for computer-aided mammography

Cited by 59 publications

References 19 publications

Needs assessment for next generation computer-aided mammography reference image databases and evaluation studies

Needs assessment for next generation computer-aided mammography reference image databases and evaluation studies

A novel feature reduction framework for digital mammogram image classification

A textural approach for recognizing architectural distortion in mammograms

Contact Info

Product

Resources

About