© N.A. Alomairy et al., Published by EDP Sciences 2026

1 Introduction

Breast cancer (BC) is the most prevalent female malignancy, accounting for 10.4% of all female cancers worldwide, and is the second most common and fifth highest mortality among all cancers in the world (Samaila & Rilwanu, 2024; Sharma et al., 2010; Suliman et al., 2023). In Saudi Arabia, it is the most common cause of cancer death among women with a 29.7% incidence rate in 2018 (Abusanad, 2022; Alqahtani et al., 2020; Basudan, 2022). Therefore, Early detection significantly improves survival and can be achieved through breast self-examination, clinical tests, and mammography (Alahmad et al., 2023; Suliman et al., 2023).

Mammography is a low-energy X-ray imaging technique used for breast screening and diagnosis (Josephine et al., 2020; Suliman et al., 2023). Standard screening follows international guidelines and includes craniocaudal (CC) and mediolateral oblique (MLO) views, which enhance lesion detection and reduce recall rates (Alahmad et al., 2023; Lekatou et al., 2019). Breast compression is used to minimize breast thickness. This improves picture quality while also reducing the dispersed radiation and patient radiation dose (Josephine et al., 2020). Despite its benefits in early cancer detection, mammography uses ionizing radiation, which can damage DNA and increase cancer risk (Kosar, 2022; Pereira et al., 2021; Tahiri et al., 2021). Given the breast tissue’s high radiosensitivity, exposure must be kept as low as reasonably achievable (Grimm et al., 2022; Tamam et al., 2021).

International bodies have set diagnostic reference levels (DRLs) to optimize patient dose while ensuring image quality. The International Atomic Energy Agency (IAEA) recommends a normal breast dose of 3 mGy for standard breast imaging examinations (Di Maria et al., 2024; Schiabel et al., 2023; Uniyal et al., 2024). The European Commission (EC) guidelines specify a DRL of 2.5 mGy for a standard compressed breast thickness (CBT) of 53 mm (Dalah et al., 2024; Garba et al., 2021; Lekatou et al., 2019; Nassar et al., 2023). In Saudi Arabia, a DRL of 1.44 mGy (1.36 mGy for CC view and 1.54 mGy for MLO view) for a ranged CBT of (40–50 mm) was set by the Saudi Food and Drug Authority (SFDA, 2022).

Recent reports on mean glandular dose (MGD) in mammography from some countries have revealed scattered values. In Sudan, MGD was 1.54 mGy (CC) and 1.58 mGy (MLO) (Abdallah et al., 2021); in Greece, 1.2 mGy (CC) and 1.5 mGy (MLO) (Tsapaki et al., 2008); and in Malaysia, 1.54 mGy (CC) and 1.82 mGy (MLO) (Jamal et al., 2003). Factors affecting MGD include CBT, beam quality, and projection angle, with no correlation to ethnicity, BMI, or age. These findings emphasize the requirement for rigorous quality control in mammography to achieve optimum image quality at the lowest possible radiation doses.

The aim of this study is to evaluate MGD in mammography, identify dose optimization factors, and compare CC and MLO views with international to ensure compliance with radiation protection guidelines. The goal is to improve Saudi Arabian mammography techniques in accordance with the ALARA (As Low As Reasonably Achievable) principle and breast cancer screening practice to ensure better patient safety and outcomes.

2 Materials and methods

2.1 Study design and setting

In this study, the mammography system at a university hospital was utilized. This study was a retrospective cross-sectional analysis conducted on a sample of 325 patients, all of whom were evaluated after machine installation. The sample size was chosen based on the machine’s usage history and to facilitate comparisons with previous national and international studies. The patients were scanned using two main views, CC and MLO projections. The data was collected using the Picture Archiving Communication System (PACS) over the period of January 2022 to October 2024. The study sample was comprised of female patients aged 40 yr and older who were undergoing routine screening or diagnostic mammography. Patients with a history of breast surgeries that could affect dose estimation, those who had previously had breast radiation therapy, pregnant or breastfeeding women, and images with missing information were excluded.

2.2 Ethical considerations

To ensure transparency, all ethical considerations were adhered to, including obtaining approval from the committee of Standing Committee Publication and Research Ethics (Reference number: REC-46/06/1244). In addition, all patient data underwent anonymization to maintain confidentiality and privacy. The study adhered to the principles outlined in the Declaration of Helsinki, thus ensuring that the rights and well-being of patients took precedence during the investigative procedure. Because of the retrospective nature of the data collection, the individual patient consent requirement was waived for this study.

2.3 Machine specification

The mammography machine used at the University Hospital is the Fujifilm AMULET Innovality (Model FDR MS-3500). It produces high-resolution images with a minimal dose includes dual-mode tomosynthesis and comfort features. Key output specifications include a maximum tube voltage of 49 kV at 450 mAs (102 mA), and a maximum current of 200 mA at 35 kV with 90 mAs. The unit also delivers up to 600 mAs at 30 kV with 166 mA. The source-to-image distance (SID) is 65 cm, and the maximum field size is 24 cm × 30 cm.

This machine employs a Tungsten (W) target with a Rhodium (Ru) filter. The machine apparatus utilized were subjected to quality control assessments by a specialized local team, with the most recent quality control taking place in February 2024.

2.4 Dosimetric assessment

The Mean Glandular Dose (MGD) is used as the dosimetric measurement in mammography, defined as the mean dose to the glandular tissue of the breast. To gather relevant parameters, a data sheet was used, collecting parameters such as compressed breast thickness (CBT), target and filter material, maximum electrical voltage (kVp), time-current product (mAs), and half-value layer (HVL). Additionally, CBT was estimated using data from the reference tube voltage (kVp), electrical current (mAs), and the target/filter combination. The MGD was calculated indirectly from the entrance surface dose (K) and half-value layer (HVL). The last quality control records indicated that beam quality, measured by the half-value layer (HVL), was 0.5 mm, which was used in the MGD calculations. The conversion coefficients used for this calculation are derived from Monte Carlo simulations of typical breast projections, assuming a 50% distribution between adipose (fatty) and glandular tissues. In this study, the conversion coefficients provided by Dance et al. (2009) were employed to extrapolate the MGD. The estimation of the MGD was carried out using an indirect method, which involved inputting specific parameters into a mathematical equation (1), as specified by Dance (Dance et al., 2000).

$$\text{MGD}\left(\text{mGy}\right)=\text{Kxgxcxs},$$(1)

where K is the Entrance Surface dose reported in the machine; g is the conversion factor for the 50% glandular breast based on thickness and HVL (Dance et al., 2000); c is the factor that corrects for differences in compression of a typical breast compared to the 50% glandular size (Dance et al., 2000); and s= 1.042 is the factor that corrects for differences due to the choice of X-ray spectrum (Dance et al., 2000).

2.5 Statistical analysis

The data was collected and organized in a Microsoft Excel spreadsheet. It underwent statistical analysis, incorporating both descriptive and inferential statistics, using the SPSS software (Version 26, SPSS Inc., Chicago, IL, USA). Pearson’s correlation testing was performed to assess the relationships between patient age, compressed breast thickness (CBT), exposure parameters, and radiation dose metrics. Multiple linear regression was conducted to identify independent predictors of the mean glandular dose (MGD). The dependent variable was MGD (in mGy) and the independent variables were patient age, compressed breast thickness (CBT in cm), tube voltage (kVp), tube current-time product (mAs), and entrance skin dose (ESD in mGy). Linearity, normality, and homoscedasticity assumptions were checked before using the model. Statistical significance was set at a P-value threshold of less than 0.05.

3 Results

3.1 Assessment of mean glandular dose (MGD) in mammography

The study sample consisted of 325 patients in total who underwent X-ray mammography at the university hospital. The descriptive statistics for patient dose and technical parameters during mammography, including tube voltage (kVp), tube current (mAs), and compressed breast thickness (CBT), are presented in Table 1. Across all projections, the mean tube voltage was 29.14 kVp (SD = 1.47, range: 26–39 kVp), the mean tube current was 63.03 mAs (SD = 18.40, range: 30–177 mAs), and the mean compressed breast thickness was 46.20 mm (SD = 15.91, range: 10–100 mm). For the cranio-caudal (CC) projection, the mean tube voltage was 28.55 kVp (SD = 1.14, range: 26–39). The mean tube current was 56.16 mAs (SD = 12.94, range: 30–123 mAs), and the mean compressed breast thickness was 39.70 mm (SD = 10.63, range: 15–72 mm). In the mediolateral oblique (MLO) projection, the mean tube voltage was 29.72 kVp (SD = 1.53, range: 27–39 kVp), with a higher mean tube current of 69.90 mAs (SD = 20.37, range: 37–177 mAs), and a mean compressed breast thickness of 52.70 mm (SD = 17.57, range: 10–100 mm).

Table 2 shows the descriptive statistical distribution of ESD and MGD values in both types of examination. For ESD, the average value was 1.06 mGy, while the median was 1 mGy. The distribution of ESD around the average and median is quite regular; the first and third quartiles range from 0.87 to 1.22 mGy. For the CC examination, the average exposure was 0.96 mGy, while the MLO examination is higher with an average of 1.18 mGy. The mean and median are very close for the CC examination, so the distribution is rather symmetrical. The range from the first quartile of 0.80 mGy to the third quartile of 1.13 mGy is moderately variable. The mean and median for the MLO examination are slightly higher compared to the CC examination, indicating that the doses in this examination tend to be somewhat higher overall. The quartile range is also a little broader than in the CC examination, with a range of 0.94 to 1.30 mGy. In the case of MGD, the mean of all is 0.35 mGy with a median of 0.34 mGy, and the interval between the 1st and 3rd quartiles is from 0.30 to 0.42 mGy. The MGD distributions for the CC and MLO examinations are comparable, with the CC having a mean of 0.32 mGy and the MLO slightly higher at 0.36 mGy. Overall, the data shows modest variability in ESD and MGD, with slightly higher ESD values observed during the MLO examination.

3.2 Analysis of the factors influencing MGD

The correlation analysis between several key mammographic variables, including age, tube voltage (kVp), tube current-time (mAs), compressed breast thickness (CBT), entrance skin dose (ESD), and mean glandular dose (MGD) are shown in Table 3. The results indicate that age is weakly negatively associated with ESD (r = −0.16, p = 0.02) and MGD (r = −0.17, p = 0.01). Tube voltage (kVp) is positively related to tube current-time (mAs) by r = 0.61, p < 0.001, CBT by r = 0.745, p < 0.001, and ESD by r = 0.54, p < 0.001. However, there is no significant correlation with MGD. Tube current-time (mAs) also significantly correlates positively with CBT, r = 0.56, p < 0.001; ESD, r = 0.92, p < 0.001; and MGD, r = 0.45, p < 0.001. Finally, CBT shows a positive correlation with ESD (r = 0.30, p < 0.001) and a negative correlation with MGD (r = −0.42, p < 0.001).

Multiple linear regression was conducted to determine the independent predictors of MGD. As shown in Table 4, the model was significant (F(5, 306) = 338.8, p < 0.001), explaining approximately 84.7% of the variance in MGD (Adjusted R² = 0.844). Compressed breast thickness (CBT) was a strong negative predictor of MGD (B = −0.0499, p < 0.001), and tube voltage (kVp) and entrance skin dose (ESD) were positive predictors. Noticeably, mAs had a strong inverse correlation with MGD (B = −0.0021, p < 0.001). Age did not appear to be a predictor in this model.

3.3 Comparison with international studies

The comparison of the MGD and CBT values of the current study with those reported in previous studies, as shown in Table 5, shows significant differences in radiation dose and breast thickness among various populations. The mean CBT values in the current study (39.70 mm for CC and 52.70 mm for MLO) are in the lower to mid-range, with Sudan and Iran having the highest CBT values. Similarly, the MGD values in this study (0.32 mGy for CC and 0.36 mGy for MLO) are significantly lower than the values reported in international studies, in which the MGD values range from 1.21 to 2.40 mGy.

4 Discussion

4.1 Assessment of mean glandular dose (MGD) in mammography

The justification and optimization of the radiation dose to patients are the main principles of radiation protection. Since there is a lack of sufficient data for the breast and due to the possibility of giving a dose higher than the limit, the calculation of MGD for patients undergoing mammography is very important when considering optimization in dose delivery. In this study, the values of MGD obtained for MLO projections were higher than those for CC projections, which can be reasonably explained by the increased breast thickness in the MLO view. MLO projection includes the pectoralis muscle that is not so obviously included in the CC projection. Additional thickness and density of the breast tissue from the pectoralis muscle results in greater attenuation of the beams (Cheddad et al., 2014; Suliman et al., 2023). Attenuation happens in beams upon the absorption or scattering of an X-ray beam through the breast tissue it is passing through, decreasing the intensity of the radiation reaching the detector. Since more X-ray energy is needed to penetrate the thicker tissue in the MLO view, a higher MGD would result in adequately imaging the breast tissue, especially the deeper structures (Di Maria et al., 2022). It is possible to attribute several radiation exposure factors to this increase in MGD for the MLO projections. The ESD, a key component of MGD, depends on a variety of parameters including exposure factors (such as tube voltage (kVp) and tube current-time (mAs)), the quality of the X-ray beam (which is influenced by the half-value layer (HVL)), the combination of target filter and filter materials, and breast thickness (Suliman et al., 2023). All of these play a significant role in determining the overall radiation dose and influences the differences in MGD between the two kinds of projection.

4.2 Analysis of factors influencing MGD

A pivotal observation in the study was the significant negative correlation between age and MGD. This suggests that as age increases, patients may have less glandular tissue and doses tend to slightly decrease, although the association is not very strong. Additionally, a strong positive correlation exists between tube voltage and both tube current (r = 0.612, p < 0.001) and ESD (r = 0.537, p < 0.001). These findings indicate that higher kVp settings are associated with increased mAs, greater ESD, and thicker breast compression, all required to maintain image quality.

The tube current had a highly significant correlation with ESD (r = 0.923 and p < 0.001), indicating that as one increases the tube current, the entrance surface dose also goes up directly, thus the optimization of this factor becomes necessary. CBT was significantly positively correlated with tube voltage, current, and ESD (r = 0.745, p < 0.001; r = 0.562, p < 0.0; r = 0.30, p < 0.001) and negatively correlated with MGD (r = −0.420, p < 0.001), suggesting that while thicker breasts result in higher ESDs, they are associated with lower MGDs. This inverse relationship is due to the operation of modern automatic exposure control (AEC) systems. They actively adjust exposure settings based on patient anatomy to attain the most optimum dose. In our set-up, the Fujifilm AMULET Innovality uses W/Rh anode-filter combinations and optimized beam qualities, which allow for reduced MGD even with denser breasts without sacrificing image quality. The system software is expected to decrease glandular dose by way of enhanced filtration and beam quality change with the rise in breast thickness. Although the opposite trend is not extensively documented in recent literature, similar findings can be seen in recent research involving digital systems with AEC optimization and dose modulation strategies (Di Maria et al., 2024; Suliman et al., 2023).

Overall, the analysis indicates that the tube current-time, mAs, is the most influential variable and is strongly related to the doses, ESD and MGD, while tube voltage, kVp, is another important variable that influences ESD and breast thickness. On the other hand, the weak negative correlation of age with both doses might indicate the less direct effect on dose-related variables. This could imply that as CBT increases, so does the possibility that MGD might be inversely related to the fact that thicker breasts require higher ESD to obtain adequate imaging, while the overall glandular dose may be reduced. These correlations indicate a complex interaction between patient factors and technical parameters that requires an imaging technique adjusted to individual profiles in order to optimize radiation exposure.

The multivariate regression analysis also confirmed that compressed breast thickness was independently associated with a decrease in MGD, even after other exposure parameters were accounted for. This supports the theory that new advanced AEC systems and beam filtration lower the glandular dose in thicker breasts. These findings are in agreement with recent studies by Suliman et al. (2023) and Di Maria et al. (2024). The positive influences of ESD and kVp reported in our regression model are both in line with our univariate correlation results and previous research, evidencing their dominant status in regulating glandular dose. The negative correlation between mAs and MGD may be indicative of system-specific exposure compensation settings made by the digital mammography machine. These conclusions highlight the importance of employing multivariate techniques to untangle the complex relationship between the technical and patient-related factors determining radiation dose in mammography.

4.3 Comparison with international studies

The MGDs for CC and MLO projections were 0.32 and 0.36 mGy respectively. These values are considerably lower compared to most of the international literature and benchmarks, as reflected in Tamam et al. (2021), Alahmad et al. (2023), Lekatou et al. (2019), and Josephine et al. (2020), and are below the international benchmarks sets by the IAEA (Di Maria et al., 2024; Uniyal et al., 2024), EC (Lekatou et al., 2019), and SFDA (SFDA, 2022). This low MGD is significant because of the high radiosensitivity of breast tissue, emphasizing the importance of adhering to diagnostic reference levels (DRLs). Additionally, the range of the MGD values and quartile distribution suggests that the majority of patients received doses within a narrow range, indicating consistency in the quality of mammography performed. Compared to studies in other countries, this study’s CBT values were relatively lower. The Saudi CBT values (39.7 mm for CC and 52.7 mm for MLO) were on the lower end of the range compared to Iran, Sudan, and Uganda (Abdallah et al., 2021, Odongo et al., 2024; Riabi et al., 2010). This may be related to regional differences in body composition, breast density, or population demographics. For example, elevated CBT values for Sudan (Abdallah et al., 2021) and Iran (Riabi et al., 2010) can be explained by either local somatotypes, lifestyle factors, or the different age structures of the samples. Furthermore, the variations in the number of subjects and geographical range could also affect the values reported.

The decreasing MGD values reflect the proper utilization of optimized imaging protocols and modern technology in Saudi Arabia in accordance with the ALARA principle that aims to minimize radiation exposure while maintaining image quality. Our findings suggest that Saudi mammographic screening programs are already in compliance with some global standards for radiation protection and dose optimization. These findings will guide future studies and improve clinical practices in mammography for better patient outcomes in the field of breast cancer screening.

5 Limitations and future works

Further optimization of mammography techniques for the purpose of lowering the doses of radiation while still being of diagnostic quality is required. The population sample size could also be expanded in future studies to include various kinds of breast density, age, and risk categories. One limitation of this study is the absence of breast density data, which was not available in the PACS and therefore not included in the analysis. Future investigations should incorporate breast density assessment to better understand its influence on MGD. Additionally, there could be the long-term follow-up of patients exposed to cumulative doses in repeated mammography examinations regarding efficacy and safety. Future studies should also investigate the impact of different target/filter combinations and AEC modes on MGD. These technical settings can significantly influence dose optimization and image quality. Future studies will endeavor to combine diagnostic performance and image quality grading (e.g., the European Quality Criteria or the reduction of contrast-detail phantoms) with dose evaluation. Finally, comparing it with other imaging methodologies − such as contrast enhancement in mammography or MRI related to dose consideration, diagnostic performance, and patient safety may prove beneficial. These alternatives offer the possibility of lower radiation exposure and the potential for improved awareness of sensitivity for dense breasts.

6 Conclusion

This study emphasizes the importance of balancing image quality and patient safety by assessing MGD and ESD in routine mammography examinations. From this data, it has been observed that the MGD values in this study are far below the average international benchmarks, reflecting good dose management and thus confirming the sample location’s conformity to the diagnostic reference levels. The discrepancies in dose between CC and MLO projections are largely because of differences in breast thickness and the presence of the pectoralis muscle in MLO images. This indicates the importance of regular quality control and precise dosimetric assessment to minimize radiation risks while maintaining diagnostic accuracy. This paper illustrates that, even using worldwide standards and optimization of exposure settings, there is a great potential for reducing patient radiation exposure.

Funding

The author gratefully acknowledges the funding of the Deanship of Graduate Studies and Scientific Research, Jazan University, Saudi Arabia, through Project Number: GSSRD-24.

Conflicts of interest

The author declares no conflicts of interest in relation to this article.

Data availability statement

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Author contribution statement

NA was responsible for the study’s conception, design, and implementation, as well as manuscript preparation, revision, findings interpretation, final approval, and overall project management. RA, SH, SA, WA, and AS played a crucial role in data collection, and analysis. All authors reviewed, contributed to, and approved the final manuscript.

Ethics approval

All ethical considerations were adhered to, including obtaining approval from the Local Committee for Research Ethics at Jazan University (Reference number: REC-46/06/1244).

Informed consent

Because of the retrospective nature of the data collection, the individual patient consent requirement was waived for this study.

References

All Tables