Automated classification of clustered microcalcifications into malignant and benign types
- PMID: 11128313
- DOI: 10.1118/1.1318221
Automated classification of clustered microcalcifications into malignant and benign types
Abstract
The objectives in this study were to design and test a fully automated method for classification of microcalcification clusters into malignant and benign types, and to compare the method's performance with that of radiologists. A novel aspect of the approach is that the relative location and orientation of clusters inside the breast was taken into account for feature calculation. Furthermore, correspondence of location of clusters in mediolateral oblique (MLO) and cranio-caudal (CC) views, was used in feature calculation and in final classification. Initially, microcalcifications were automatically detected by using a statistical method based on Bayesian techniques and a Markov random field model. To determine malignancy or benignancy of a cluster, a method based on two classification steps was developed. In the first step, classification of clusters was performed and in the second step a patient based classification was done. A total of 16 features was used in the study. To identify meaningful features, a feature selection was applied, using the area under the receiver operating characteristic (ROC) curve (Az value) as a criterion. For classification the k-nearest-neighbor method was used in a leave-one-patient-out procedure. A database of 192 mammograms with 280 true positive detected microcalcification clusters was used for evaluation of the method. The set consisted of cases that were selected for diagnostic work up during a 4 year period of screening in the Nijmegen region (The Netherlands). Because of the high positive predictive value in the screening program (50%), this set did not contain obvious benign cases. The method's best patient-based performance on this set corresponded with Az = 0.83, using nine features. A subset of the data set, containing mammograms from 90 patients, was used for comparing the computer results to radiologists' performance. Ten radiologists read these cases on a light-box and assessed the probability of malignancy for each patient. All participants had experience in clinical mammography and participated in our observer study during the last 2 days of a 2-week training session leading to screening mammography certification. Results on the subset showed that the method's performance (Az = 0.83) was considerably higher than that of the radiologists (Az = 0.63).
Similar articles
-
Image segmentation feature selection and pattern classification for mammographic microcalcifications.Comput Med Imaging Graph. 2005 Sep;29(6):419-29. doi: 10.1016/j.compmedimag.2005.03.002. Comput Med Imaging Graph. 2005. PMID: 16002263
-
Digital mammography: observer performance study of the effects of pixel size on the characterization of malignant and benign microcalcifications.Acad Radiol. 2001 Jun;8(6):454-66. doi: 10.1016/S1076-6332(03)80616-5. Acad Radiol. 2001. PMID: 11394537
-
Optimizing parameters for computer-aided diagnosis of microcalcifications at mammography.Acad Radiol. 2000 Jun;7(6):406-12. doi: 10.1016/s1076-6332(00)80380-3. Acad Radiol. 2000. PMID: 10845399
-
Improvement of radiologists' characterization of mammographic masses by using computer-aided diagnosis: an ROC study.Radiology. 1999 Sep;212(3):817-27. doi: 10.1148/radiology.212.3.r99au47817. Radiology. 1999. PMID: 10478252
-
Computerized analysis of mammographic microcalcifications in morphological and texture feature spaces.Med Phys. 1998 Oct;25(10):2007-19. doi: 10.1118/1.598389. Med Phys. 1998. PMID: 9800710
Cited by
-
Exploring CNN potential in discriminating benign and malignant calcifications in conventional and dual-energy FFDM: simulations and experimental observations.J Med Imaging (Bellingham). 2021 May;8(3):033501. doi: 10.1117/1.JMI.8.3.033501. Epub 2021 May 13. J Med Imaging (Bellingham). 2021. PMID: 34002162 Free PMC article.
-
The importance of early detection of calcifications associated with breast cancer in screening.Breast Cancer Res Treat. 2018 Jan;167(2):451-458. doi: 10.1007/s10549-017-4527-7. Epub 2017 Oct 17. Breast Cancer Res Treat. 2018. PMID: 29043464 Free PMC article.
-
Modified Bat Algorithm for Feature Selection with the Wisconsin Diagnosis Breast Cancer (WDBC) Dataset.Asian Pac J Cancer Prev. 2017 May 1;18(5):1257-1264. doi: 10.22034/APJCP.2017.18.5.1257. Asian Pac J Cancer Prev. 2017. PMID: 28610411 Free PMC article.
-
Fuzzy technique for microcalcifications clustering in digital mammograms.BMC Med Imaging. 2014 Jun 24;14:23. doi: 10.1186/1471-2342-14-23. BMC Med Imaging. 2014. PMID: 24961885 Free PMC article.
-
Analysis of perceived similarity between pairs of microcalcification clusters in mammograms.Med Phys. 2014 May;41(5):051904. doi: 10.1118/1.4870959. Med Phys. 2014. PMID: 24784383 Free PMC article.
Publication types
MeSH terms
LinkOut - more resources
Full Text Sources
Medical