Analysis of clustered microcalcifications by using a single numeric classifier extracted from mammographic digital images

Rationale and Objectives. The authors prospectively tested the performance of a single numeric classifier constructed from a discriminative analysis c lassification system based on automatic computer-extracted quantitative fea tures of clustered microcalcifications. Materials and Methods. Mammographically detected clustered microcalcificati ons in patients who had been referred for biopsy were digitized at 600 dpi with an 8-bit gray scale. A software program was developed to extract featu res automatically from digitized images to describe the clustered microcalc ifications quantitatively. The significance of these features was evaluated by using the Wilcoxon test, the Welch modified two-sample t test, and the two-sample Kolmogorov-Smirnov test. A discriminant analysis pattern recogni tion system was constructed to generate a single numeric classifier for eac h case, based on the extracted features. This system was trained on 137 arc hival known reference cases and its performance tested on 24 unknown prospe ctive cases. The results were evaluated by using receiver operating charact eristic analysis. Results. Thirty-seven extracted parameters demonstrated a statistically sig nificant difference between the values for the benign and for the malignant lesions. Seven independent factors were selected to construct the classifi er and to evaluate the unknown prospective cases. The area under the receiv er operating characteristic curve for the prospective cases was 0.88. Conclusion. A pattern recognition classifier based on quantitative features for clustered microcalcifications at screen-film mammography was found to perform satisfactorily. The software may be of value in the interpretation of mammographically detected microcalcifications.