Estimation of glandular dose in mammography based on artificial neural networks

Autor(es)

Trevisan Massera, Rodrigo ^[1] ; Tomal, Alessandra ^[1]

Afiliação do(s) autor(es)

^[1] Univ Estadual Campinas, Inst Fis Gleb Wataghin, BR-13083859 Campinas - Brazil

• Tipo de documento: Artigo Científico
• Fonte: Physics in Medicine and Biology; v. 65, n. 9 MAY 7 2020.
• Citações Web of Science: 0

Resumo

This work proposes using artificial neural networks (ANNs) for the regression of the dosimetric quantities employed in mammography. The data were generated by Monte Carlo (MC) simulations using a modified and validated version of the PENELOPE (v. 2014) + penEasy (v. 2015) code. A breast model of a homogeneous mixture of adipose and glandular tissue was adopted. The ANNs were constructed using the Keras and scikit-learn libraries for mean glandular dose (MGD) and air kerma (K-air) regressions, respectively. In total, seven parameters were considered, including the incident photon energies (from 8.25 to 48.75 keV), breast geometry, breast glandularity and K-air acquisition geometry. Two ensembles of five ANNs each were formed to calculate MGD and K-air. The normalized glandular dose coefficients (DgN) were calculated using the ratio of the ensemble outputs for MGD and K-air. Polyenergetic DgN values were calculated by weighting monoenergetic values by the spectrum bin probabilities. The results indicate a very good ANN prediction performance when compared to the validation data, with median errors on the order of the average simulation uncertainties (\& x224d; 0.2%). Moreover, the predicted DgN values are in good agreement compared with previously published works, with mean (maximum) differences up to 2.2% (9.4%). Therefore, it is shown that ANNs could be a complementary or alternative technique to tables, parametric equations and polynomial fits to estimate DgN values obtained via MC simulations. (AU)

• Processo FAPESP: 16/15366-9 - Otimização dos parâmetros de exposição em mamografia digital: estudos experimentais e por Simulação Monte Carlo
• Beneficiário: Rodrigo Trevisan Massera
• Modalidade de apoio: Bolsas no Brasil - Mestrado
• Processo FAPESP: 15/21873-8 - Desenvolvimento e implementação de metodologias para otimização de técnicas de imagens em radiologia digital
• Beneficiário: Alessandra Tomal
• Modalidade de apoio: Auxílio à Pesquisa - Regular