Articles

<b>Background:</b><b> </b>Currently, the DRL quantity in mammography are evaluated in terms of mean glandular dose (MGD). Since the MGD cannot be measured directly, it can be obtained by calculation using the equation (D=K*g*c*s). In previous studies, the conversion factor g was calculated by Monte Carlo simulation and is not reported from actual measurements. In this study, we focused on the g-factor, which is a conversion factor to the MGD at 50% glandularity, and attempted to measure it using a nanoDot dosimeter to see if it can be used in mammography. <b>Methods: </b>The nanoDot dosimeters were inserted in a PMMA phantom at depths ranging from 0 cm to 6 cm in 1 cm increments, and measurements were made in three HVLs of 0.3 mmAl, 0.35 mmAl, and 0.4 mmAl HVL. The g-factor was calculated from the nanoDot dosimeter values using a conversion equation. <b>Results and Discussion:</b> The measured g-factors for all the HVLs were in close agreement with those of Dance et al. The values of the previous studies did not include the backscatter factor, which may have underestimated the MGD. The difference was smaller for the 0.4 mm Al. Compared to the other HVLs, the 0.4 mm Al was measured without a compression plate, which may have been influenced by the presence or absence of a compression plate. <b>Conclusion:</b> The nanoDot dosimeters were used to calculate g-factors. The results agreed with those of previous studies within uncertainty. This indicates that nanoDot dosimeters can be used in the mammography field.

Cell division that is not controlled leads to cancer, an incurable condition. An early diagnosis has the potential to lower death rates from breast cancer, the most frequent disease in women worldwide. Imaging studies of the breast may help doctors find the disease and diagnose it. This study explores an effectiveness of DL and ML models in a classification of mammography images for breast cancer detection, utilizing the publicly available CBIS-DDSM dataset, which comprises 5,000 images evenly divided between benign and malignant cases. To improve diagnostic accuracy, models such as Gaussian Naïve Bayes (GNB), CNNs, KNN, and MobileNetV2 were assessed employing performance measures including F1-score, recall, accuracy, and precision. The methodology involved data preprocessing techniques, including transfer learning and feature extraction, followed by data splitting for robust model training and evaluation. Findings indicate that MobileNetV2 achieved a highest accuracy99.4%, significantly outperforming GNB (87.2%), CNN (96.7%), and KNN (91.2%). The outstanding capacity of MobileNetV2 to identify between benign and malignant instances was shown by the investigation, which also made use of confusion matrices and ROC curves to evaluate model performance.