Using Speed of Sound Imaging to Characterize Breast Density

A population of 165 women with negative mammographic screens also received an ultrasound tomography (UST) examination at the Karmanos Cancer Institute in Detroit, MI. Standard statistical techniques were employed to measure the associations between the various mammographic- and UST-related density measures and various participant characteristics such as age, weight and height. The mammographic percent density (MPD) was found to have similar strength associations with UST mean sound speed (Spearman coefficient, r_s = 0.722, p < 0.001) and UST median sound speed (r_s = 0.737, p < 0.001). Both were stronger than the associations between MPD with two separate measures of UST percent density, a k-means (r_s = 0.568, p < 0.001) or a threshold (r_s = 0.715, p < 0.001) measure. Segmentation of the UST sound speed images into dense and non-dense volumes showed weak to moderate associations with the mammographically equivalent measures. Relationships were found to be inversely and weakly associated between age and the UST mean sound speed (r_s = ?0.239, p = 0.002), UST median sound speed (r_s = ?0.226, p = 0.004) and MPD (r_s = ?0.204, p = 0.008). Relationships were found to be inversely and moderately associated between body mass index (BMI) and the UST mean sound speed (r_s = ?0.429, p < 0.001), UST median sound speed (r_s = ?0.447, p < 0.001) and MPD (r_s = ?0.489, p < 0.001). The results confirm and strengthen findings presented in previous work indicating that UST sound speed imaging yields viable markers of breast density in a manner consistent with mammography, the current clinical standard. These results lay the groundwork for further studies to assess the role of sound speed imaging in risk prediction.     

Estimated Mortality of Breast Cancer Patients Based on Stage at Diagnosis and National Screening Guideline Categorization

PurposeTo compare overall and stage I to IV mortalities of patients diagnosed with breast cancer, calculated from stage at diagnosis using the Surveillance, Epidemiology, and End Results (SEER) database stage mortality estimates, which are based on national screening guideline categorization.MethodsFrom the stage at diagnosis of new breast cancer patients between 2010 and 2014, percentages of invasive cancers, stage 0 + I of total cancers, and stage I of invasive cancers, were calculated. Five-year estimated overall and invasive mortalities were calculated based on stage at diagnosis and SEER survival data. Program categories defined included an Annual Program, based on the ACR (annual screening age 40 and above), a Biennial Program, based on the US Preventative Services Taskforce (biennial screening ages 50 to 74 years), and a Hybrid Program, based on the American Cancer Society (annual screening ages 45 to 54 years, then biennially at ages 55 and above), including respective interval cancers.ResultsIn all, 445 breast cancers met the study inclusion criteria. Comparing program categories, the Annual Program had the lowest percentage of invasive cancers (75.3%), highest percentages of stage 0 + I of total cancers (75.3%) and stage I of invasive cancers (67.1%), and the lowest 5-year estimated overall (10.1%) and stage I to IV (12.0%) mortalities. Estimated overall and stage I to IV mortalities for the Annual Program was 37.3% and 30.6% less, respectively, than the Biennial Program, and 31.8% and 26.8% less, respectively, than the Hybrid Program.ConclusionsBased on stage at diagnosis, the greatest mortality reduction is achieved with mammography utilization starting at the age of 40.     

乳腺浸润性导管癌的X光影像与分子分型及其组织学分级的相关性研究

目的 探索不同分子亚型乳腺浸润性导管癌病灶X光影像的差异,以及与组织学分级的相关性。方法 收集因乳腺肿物住院,且均进行术前乳腺X光摄片,并最终通过病理检查确诊为乳腺浸润性导管癌的患者完整诊疗资料。回顾性分析术前X光影像及术后病理学资料,参照2019版中国临床肿瘤学会乳腺癌诊疗指南(CSCO BC指南)推荐方法进行分子分型,比较不同分子亚型组间的X光影像差异并探讨病灶X光征像与组织学分级之间的联系。结果 五组分子亚型的X光影像在腺体分型、肿块类型及肿块边缘特征等方面存在统计学差异(P＜0.05)。其中三阴型的腺体分型多为散在纤维腺体型(50.0%,18/36);对比其余分子亚型,三阴型多表现为单纯肿块(83.3%,30/36)且肿块边缘光整(52.8%,19/36);不同病理组织学分级病灶的X光影像在腺体分型、肿块类型及肿块边缘特征等方面存在统计学差异(P＜0.05)。对比中高分化组,低分化组的腺体分型多为散在纤维腺体型(37.3%,19/51),X线影像多表现为单纯肿块型(52.9%,27/51)且肿块边缘多光整(31.4%,16/51)。多因素Logistic回归分析显示,肿块边缘特征是X光影像辨别低分化组浸润性导管癌病灶的独立危险因素(P＜0.05);乳腺浸润性导管癌的分子分型与组织学分级具有相关性(V=0.482,P＜0.001),其中三阴型低分化组占比(63.9%,23/69)及Luminal A的中高分化组占比(97.2%,72/74)均显著高于其余亚组(P＜0.001)。结论 乳腺浸润性导管癌的X线影像可以一定程度上辨别分子分型及病理组织学分级,能够为乳腺癌患者个性化治疗方案的选择及治疗效果的提前预测提供影像学依据。     

Standardised or individualised X-ray tube angle for mediolateral oblique projection in digital mammography?

IntroductionWe aimed to investigate whether there were any differences in positioning criteria related to the presentation of the pectoralis major muscle (pectoral muscle) for women of different heights using a standardized 60° X-ray tube angle for mammograms in mediolateral oblique (MLO) projection.MethodsData from MLO mammograms of right breasts of 45,193 women screened in BreastScreen Norway 2016–2019 were used. The positioning criteria were related to the pectoral muscle length (measure A and measure B), width and shape and considered adequate or inadequate depending on the degree of fulfilling the criteria. Data associated with the pectoral muscle were extracted from Volpara, an automated software for breast density assessment. Information on height was obtained from a self-reported questionnaire received by the women together with the invitation to attend the screening program. Women were divided into three groups based on the height percentiles (P) in the Norwegian growth curves: < 25th percentile (<P25th: ≤163 cm), 25th-75th percentile (P25–P75th: 164–170 cm), and >75th percentile (>P75th: >170 cm). Logistic regression was used to analyse the odds of adequate pectoral muscle length A and B, and shape, adjusting each model for screening technique and equipment model. Results were presented with odds ratios (OR) and 95% confidence intervals (CI).ResultsMean age of the screened women was 61.5 (SD = 4.8) years. The adequate measure for the pectoral muscle length A was obtained for 25.9% (11,724/45,193), length B for 76.3% (34,489/45,193), width for 75.0% (33,894/45,193) and shape for 97.6% (44,118/45,193) of the mammograms. Adjusted odds of an adequate pectoral muscle length A were lower for women of <P25th (OR = 0.90, 95% CI: 0.86–0.95) compared to women of P25-75th. Odds of an adequate pectoral muscle length B were lower for women of <P25th (OR = 0.88, 95% CI 0.84–0.93) and higher for women of >P75th (OR = 1.08, 95% CI 1.02–1.14) compared to women of P25-75. Odds of an adequate pectoral muscle shape were higher for women of <P25th (OR = 1.14, 95%CI 1.08–1.19) and lower for women of >P75th (OR = 0.92, 95% CI 0.87–0.97) compared to women of P25-75th.ConclusionThe 60° X-ray tube angle might suit most of the female population offered mammographic screening in Norway, but women of a relatively low height (163 cm or lower) might benefit from an X-ray tube angle less than 60-degrees.Implications for practiceUsing 60° X-ray tube angle for the MLO mammograms in BreastScreen Norway fit the majority of the participating women. More research is needed to change the protocol associated with the tube angle for women shorter than 163 cm.     

Assessing Transgender Patient Care and Gender Inclusivity of Breast Imaging Facilities Across the United States

PurposeTo evaluate transgender patient care, gender inclusivity, and transgender health-related policies at breast imaging facilities across the United States.MethodsA survey on breast imaging facilities’ policies and practices regarding transgender care was distributed to the membership of the Society of Breast Imaging, consisting of approximately 2,500 breast radiologists across the United States. The survey was conducted by e-mail in January 2018.ResultsThere were 144 survey respondents. Responses showed that 78.5% of facilities have gender-neutral patient bathrooms, 9.0% have a separate waiting area for transgender patients, and 76.4% do not have dominant pink hues in their facilities, although 54.2% have displays with female gender content. Also, 58.0% of intake forms do not ask patients to provide their gender identity, although 25.9% automatically populate with female phrases. Within the electronic health record, 32.9% lack a distinct place to record patients’ preferred names and 54.9% lack a distinct place to record patients’ gender pronouns. The majority (73.4%) do not have explicit policies related to the care of transgender patients. Only 14.7% of facilities offer lesbian, gay, bisexual, and transgender training.ConclusionOur national survey demonstrates that many breast imaging facilities do not have structures in place to consistently use patients’ preferred names and pronouns, nor provide inclusive environments for transgender patients. All breast imaging facilities should recognize the ways in which their practices may intensify discrimination, exclusivity, and stigma for transgender patients and should seek to improve their transgender health competencies and foster more inclusive environments.     

Development of a phantom to test fully automated breast density software – A work in progress

ObjectivesMammographic density (MD) is an independent risk factor for breast cancer and may have a future role for stratified screening. Automated software can estimate MD but the relationship between breast thickness reduction and MD is not fully understood. Our aim is to develop a deformable breast phantom to assess automated density software and the impact of breast thickness reduction on MD.MethodsSeveral different configurations of poly vinyl alcohol (PVAL) phantoms were created. Three methods were used to estimate their density. Raw image data of mammographic images were processed using Volpara to estimate volumetric breast density (VBD%); Hounsfield units (HU) were measured on CT images; and physical density (g/cm³) was calculated using a formula involving mass and volume. Phantom volume versus contact area and phantom volume versus phantom thickness was compared to values of real breasts.ResultsVolpara recognized all deformable phantoms as female breasts. However, reducing the phantom thickness caused a change in phantom density and the phantoms were not able to tolerate same level of compression and thickness reduction experienced by female breasts during mammography.ConclusionOur results are promising as all phantoms resulted in valid data for automated breast density measurement. Further work should be conducted on PVAL and other materials to produce deformable phantoms that mimic female breast structure and density with the ability of being compressed to the same level as female breasts.Advances in knowledgeWe are the first group to have produced deformable phantoms that are recognized as breasts by Volpara software.     

Manual or auto-mode: Does this affect radiation dose in digital mammography without compromising image quality?

IntroductionIn current practice, auto-filter exposure mode is used for most screening mammography examinations. However, with better understanding of the side effects of radiation, it is important to examine exposure parameters and practises to minimise radiation dose to patients. The purpose of this phantom study is to investigate the impact that different exposure modes (manual, auto-time and auto-filter) have on radiation dose, while maintaining images of diagnostic quality.MethodsThis study was conducted in three stages. In the first stage, 125 images were taken using a Gammex 156 phantom with polymethyl methacrylate blocks to reflect varying thicknesses (4.5, 5.0, 5.5, 6.0 and 6.5 cm). In the second stage, three mammographers independently assessed image quality and assigned scores based on the number of distinct fibers, masses and speck groups visible. Images with acceptable quality were further investigated in the third stage by comparing their average glandular dose (AGD) using the Kruskal–Wallis H test and Dunn's post-hoc pairwise analysis.ResultsSignificant differences in AGD were shown between the auto-filter mode and manual mode techniques for 6.0 cm, and between auto-time mode and manual mode techniques for 6.5 cm (p < 0.05).ConclusionFor 4.5, 5.0 and 5.5 cm phantoms, as AGD was not significantly different among the different modes, the auto-filter may remain the most practical option. However, significant reductions in AGD were obtained for the 6.0 and 6.5 cm phantoms when manual mode techniques were used.Implications for practiceManual mode techniques can potentially provide dose-saving opportunity in 6.0 and 6.5 cm breast thickness though future work on human breast should be done to confirm this. Results from this study will support future research based on patient data.     

全视野数字化乳腺摄影及数字乳腺断层摄影诊断致密型乳腺内病变

目的 探讨全视野数字化乳腺摄影（FFDM）及数字乳腺断层摄影（DBT）对致密型乳腺内病变的诊断价值。方法 收集经病理证实的176例乳腺病变患者,均为致密型乳腺,分别采用FFDM、DBT、FFDM联合DBT进行诊断。以病理结果作为金标准,评价FFDM、FFDM联合DBT对致密型乳腺内病变的诊断价值;比较FFDM与DBT对致密型乳腺内结构扭曲、不同病理类型乳腺癌的检出率及对病灶边缘特征的显示情况。结果 176例中,良性病变83例,恶性病变93例。单纯采用FFDM及以FFDM联合DBT鉴别致密型乳腺内病变良恶性的敏感度分别为77.42%（72/93）、90.32%（84/93）,特异度分别为55.42%（46/83）、71.08%（59/83）,ROC曲线下面积（AUC）分别为0.757、0.885。FFDM、DBT对致密型乳腺内恶性肿块型病变的诊断敏感度分别为81.43%（57/70）、92.86%（65/70）,特异度分别为56.52%（26/46）、76.09%（35/46）,AUC分别为0.787、0.895。恶性肿块中,FFDM、DBT对毛刺征的检出率分别为31.43%（22/70）、54.29%（38/70）,差异有统计学意义（χ²=7.467,P=0.006）。DBT检出结构扭曲22例,FFDM仅检出其中5例,差异有统计学意义（χ²=27.704,P<0.001）。FFDM及DBT对原位癌的检出率均为84.62%（11/13）;对浸润癌的检出率分别为77.22%（61/79）、92.41%（73/79）,差异有统计学意义（χ²=7.075,P=0.008）。结论 对于致密型乳腺,DBT可提高病变检出率,FFDM联合DBT具有较高诊断效能。     

比较超声与乳腺钼靶摄影评估不同类型乳腺微小肿块

目的 比较超声（US）和钼靶X线乳腺摄影（MTM）评估不同类型乳腺微小肿块的价值。方法 选取562例经手术或穿刺确诊,并具有完整US、MTM以及病理结果的乳腺微小肿块患者,根据MTM图像将乳腺分为4种类型,比较US和MTM对不同类型乳腺微小肿块的检出率、敏感度、特异度及准确率,并分析US与MTM的诊断一致性。结果 在美国放射学院（ACR）a类乳腺,MTM对微小肿块的检出率以及诊断恶性微小肿块的敏感度、准确率均高于US（P均<0.05）,US和MTM的诊断一致性一般（Kappa=0.47）;在ACR b类和ACR c类乳腺,MTM和US对微小肿块的检出率以及诊断恶性微小肿块的敏感度、特异度以及准确率差异无统计学意义（P均>0.05）,两者的诊断一致性较好（Kappa=0.78、0.76）;在ACR d类乳腺,US对微小肿块的检出率以及诊断恶性微小肿块的敏感度、准确率均高于MTM（P<0.05）,两者的诊断一致性较差（Kappa=0.35）。结论 对于ACR a类乳腺发生的微小肿块,MTM优于US;对于ACR b类和ACR c类乳腺微小肿块,US和MTM无明显差异;对于ACR d类乳腺微小肿块,US优于MTM。