The prediction of breast cancer biopsy outcomes using two CAD approaches that both emphasize an intelligible decision process

Abstract

Mammography is the most effective method for breast cancer screening available today. However, the low positive predictive value of breast biopsy resulting from mammogram interpretation leads to approximately 70% unnecessary biopsies with benign outcomes. To reduce the high number of unnecessary breast biopsies, several computer-aided diagnosis (CAD) systems have been proposed in the last several years. These systems help physicians in their decision to perform a breast biopsy on a suspicious lesion seen in a mammogram or to perform a short term follow-up examination instead. We present two novel CAD approaches that both emphasize an intelligible decision process to predict breast biopsy outcomes from BI-RADS findings. An intelligible reasoning process is an important requirement for the acceptance of CAD systems by physicians. The first approach induces a global model based on decison-tree learning. The second approach is based on case-based reasoning and applies an entropic similarity measure. We have evaluated the performance of both CAD approaches on two large publicly available mammography reference databases using receiver operating characteristic (ROC) analysis, bootstrap sampling, and the ANOVA statistical significance test. Both approaches outperform the diagnosis decisions of the physicians. Hence, both systems have the potential to reduce the number of unnecessary breast biopsies in clinical practice. A comparison of the performance of the proposed decision tree and CBR approaches with a state of the art approach based on artificial neural networks (ANN) shows that the CBR approach performs slightly better than the ANN approach, which in turn results in slightly better performance than the decision-tree approach. The differences are statistically significant (p value < 0.001). On 2100 masses extracted from the DDSM database, the CRB approach for example resulted in an area under the ROC curve of A(z) = 0.89 +/- 0.01, the decision-tree approach in A(z) = 0.87 +/- 0.01, and the ANN approach in A(z) = 0.88 +/- 0.01.

Using BI-RADS Descriptors and Ensemble Learning for Classifying Masses in Mammograms

“…Kim et al [5] used BI-RADS-based features, and applied a Support Vector Machine based on recursive feature elimination (SVM-RFE) for classifying abnormalities in mammogram images. Elter et al [6] presented two CAD systems which use decision-tree learning and case-based reasoning for the prediction of breast cancer from BI-RADS attributes.…”

Reliable Probabilistic Prediction for Medical Decision Support

“…-Mammographic Mass, which is concerned with the discrimination between benign and malignant mammographic masses based on 3 BI-RADS attributes (mass shape, margin and density) and the patient's age [4]. It consists of 961 cases of which 516 are benign and 445 are malignant.…”

“…The resulting dataset consisted of 830 examples described by 10 attributes each. The NNs used consisted of 4 hidden units for the Mammographic Mass data, as this was the number of units used in [4], and 10 for the Pima Indians Diabetes data, as this seemed to have the best performance with the original NN classifier. All NNs were trained with the scaled conjugate gradient algorithm minimizing cross-entropy error and early stopping based on a validation set consisting of 20% of the corresponding training set.…”

A fine-grained Random Forests using class decomposition: an application to medical diagnosis

“…All these sets are from the medical domain as can be seen in Table 1, and most of it have been downloaded from the UCI repository [5]. These include (1) Breast Cancer Wisconsin (Original) Data Set [25], (2) Heart Disease [28], (3) Lung Cancer [19], (4) Mammographic Mass Data Set [11], (5) Parkinsons Data Set [23], (6) Pima Indians Diabetes [5], (7) Thyroid 1 . Table 1 summarises the main Charasteristics of these datasets.…”