Uncertainty and normalized glandular dose evaluations in digital mammography and digital breast tomosynthesis with a machine learning methodology.
Authors
Affiliations (5)
Affiliations (5)
- Department of Physics "A. Pontremoli", University of Milan & INFN sez. Milano, Milano, Italy.
- Department of Imaging and Pathology, Division of Medical Physics & Quality Assessment, KU Leuven, Leuven, Belgium. Electronic address: [email protected].
- INFN - Sez. Ferrara, Via Saragat 1, Ferrara 44122, Italy.
- Department of Imaging and Pathology, Division of Medical Physics & Quality Assessment, KU Leuven, Leuven, Belgium; Department of Radiology, UZ Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium.
- Department of Medical Equipment, Electronic and Information Technologies in Healthcare, Medical University of Varna, Varna, Bulgaria.
Abstract
To predict the normalized glandular dose (DgN) coefficients and the related uncertainty in mammography and digital breast tomosynthesis (DBT) using a machine learning algorithm and patient-like digital breast models. 126 patient-like digital breast phantoms were used for DgN Monte Carlo ground truth calculations. An Automatic Relevance Determination Regression algorithm was used to predict DgN from anatomical breast features. These features included compressed breast thickness, glandular fraction by volume, glandular volume, center of mass and standard deviation of the glandular tissue distribution in the cranio-caudal direction. An algorithm for data imputation was explored to account for avoiding the use of the latter two features. 5-fold cross validation showed that the predictive model provides an estimation of DgN with 1% average difference from the ground truth; this difference was less than 3% in 50% of the cases. The average uncertainty of the estimated DgN values was 9%. Excluding the information related to the glandular distribution increased this uncertainty to 17% without inducing a significant discrepancy in estimated DgN values, with half of the predicted cases differing from the ground truth by less than 9%. The data imputation algorithm reduced the estimated uncertainty, without restoring the original performance. Predictive performance improved by increasing tube voltage. The proposed methodology predicts the DgN in mammography and DBT for patient-derived breasts with an uncertainty below 9%. Predicting test evaluations reported 1% average difference from the ground truth, with 50% of the cohort cases differing by less than 5%.