Review of the current state of digital image analysis in breast pathology

Advances in digital image analysis have the potential to transform the practice of breast pathology. In the near future, a move to a digital workflow offers improvements in efficiency. Coupled with artificial intelligence (AI), digital pathology can assist pathologist interpretation, automate time‐consuming tasks, and discover novel morphologic patterns. Opportunities for digital enhancements abound in breast pathology, from increasing reproducibility in grading and biomarker interpretation, to discovering features that correlate with patient outcome and treatment. Our objective is to review the most recent developments in digital pathology with clear impact to breast pathology practice. Although breast pathologists currently undertake limited adoption of digital methods, the field is rapidly evolving. Care is needed to validate emerging technologies for effective patient care.     

Evaluation of physician performance using a concurrent-read artificial intelligence system to support breast ultrasound interpretation

PurposeThe purpose of this study was to compare the diagnostic performance and the interpretation time of breast ultrasound examination between reading without and with the artificial intelligence (AI) system as a concurrent reading aid.Material and methodsA fully crossed multi-reader and multi-case (MRMC) reader study was conducted. Sixteen participating physicians were recruited and retrospectively interpreted 172 breast ultrasound cases in two reading scenarios, once without and once with the AI system (BU-CAD™, TaiHao Medical Inc.) assistance for concurrent reading. Interpretations of any given case set with and without the AI system were separated by at least 5 weeks. These reading results were compared to the reference standard and the area under the LROC curve (AUCLROC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated for performance evaluations. The interpretation time was also compared between the unaided and aided scenarios.ResultsWith the help of the AI system, the readers had higher diagnostic performance with an increase in the average AUCLROC from 0.7582 to 0.8294 with statistically significant. The sensitivity, specificity, PPV, and NPV were also improved from 95.77%, 24.07%, 44.18%, and 93.50%–98.17%, 30.67%, 46.91%, and 96.10%, respectively. Of these, the improvement in specificity reached statistical significance. The average interpretation time was significantly reduced by approximately 40% when the readers were assisted by the AI system.ConclusionThe concurrent-read AI system improves the diagnostic performance in detecting and diagnosing breast lesions on breast ultrasound images. In addition, the interpretation time is effectively reduced for the interpreting physicians.     

Teaching cross-cultural design thinking for healthcare

ObjectivesArtificial intelligence (AI) is poised to transform breast cancer care. However, most scientists, engineers, and clinicians are not prepared to contribute to the AI revolution in healthcare. In this paper, we describe our experiences teaching a new undergraduate course for American students that aims to prepare the next generation for cross-cultural designthinking, which we believe is crucial for AI to achieve its full potential in breast cancer care.Materials and methodsThe key course activities are planning, conducting, and interpreting interviews of healthcare professionals from both Portugal and the United States. Since the course is offered as a short-term faculty-led study abroad program in Portugal, students are able to explore the impact of culture on healthcare delivery and the design of healthcare technologies.ResultsThe learning assessments demonstrated student growth in several areas pertinent for future development of AI for breast cancer care. With respect to understanding breast cancer care, prior to taking this course, most students had underestimated the impact of cancer and its treatment on women’s quality of life and most were unaware of the importance of multidisciplinary care teams. Regarding AI in medicine, students became more mindful of data privacy issues and the need to consider the effect of AI on healthcare professionals.ConclusionThis course illustrates the potential benefits for AI in medicine of introducing future scientists, engineers, and clinicians to cross cultural design-thinking early in their educational experiences.     

基层医院乳腺癌筛查模式的研究

目的探讨基层医院乳腺癌筛查的优化模式。方法对1998年1月至2006年12月新会地区的47976名妇女进行乳腺保健教育实施的乳腺临床检查加乳腺B超检查的乳腺癌筛查模式进行分析和总结。结果47976名妇女中筛查出乳腺异常情况者12694人,其中1126人接受手术活检或空心针穿刺活检,确诊为乳腺良性病变1069例,占94.1%,确诊为乳腺癌57例,占5.1%,其中非浸润性癌7例（均为T0期）,包括有小叶原位癌3例,导管内癌4例;浸润性癌50例,其中T1期是16例。本组乳腺癌筛查的乳腺癌检出率为0.12%。结论临床乳腺检查加乳腺B超检查的乳腺癌筛查模式能提高基层医院的乳腺癌早期诊断水平,有一定的临床推广价值。     

Transgenic introduction of androgen receptor into estrogen-receptor-, progesterone-receptor-, and androgen-receptor-negative breast cancer cells renders them responsive to hormonal manipulation

BACKGROUND: Estrogen-receptor (ER)-, progesterone-receptor (PR)-, and androgen-receptor (AR)-negative breast cancer cells are unaffected by treatment with dehydroepiandrosterone-sulfate (DHEAS) and an aromatase inhibitor (AI). We hypothesized that cell growth would be inhibited with DHEAS/AI treatment after successful transfection of an AR expression vector. METHODS: ER/PR/AR-negative breast cancer cells were transfected with an AR expression vector and treated with DHEAS/AI for 2 days. Growth inhibition of these cells was compared with that of transfected cells treated with only AI or with nontransfected cells treated with DHEAS/AI. Mann-Whitney U test was used to determine statistical significance. RESULTS: Cell death rates of 53.5% (P = .001) and 40.1% (P = .006) were seen in transfected cells treated with DHEAS/AI compared with controls for days 1 and 2, respectively. Nontransfected cells were unaffected by treatment. COMMENTS: ER/PR/AR-negative cells transfected with AR were killed by DHEAS/AI treatment, providing evidence that AR is responsible for this effect. This provides the first AR-targeted hormonal therapy for ER breast cancer.     

人工智能在肝癌领域应用研究进展

近年来人工智能（AI）技术发展迅速，其子领域深度学习已广泛应用于医学图像处理和临床肿瘤学分析。很多研究结果已经表明将人工智能与医学相结合能够为医生带来较好的辅助作用，可以很好地帮助医生更快更准确地诊断疾病。医学影像方面，AI能够分类出CT、MRI影像片中不同器官，并通过影像片判断肝脏肿瘤的良恶性、预测肝癌微血管侵犯（MVI）；数字病理方面，AI可以判断肝细胞癌的分化等级、诊断并分类MVI以及预测病人的术后生存率等。肝癌治疗方面，AI可以协助医生制定手术策略、为医生提供最适合病人的治疗方案，帮助医生做出治疗决策。     

Breast cancer surgery with augmented reality

Introduction: Innovations in 3D spatial technology and augmented reality imaging driven by digital high-tech industrial science have accelerated experimental advances in breast cancer imaging and the development of medical procedures aimed to reduce invasiveness. Presentation of case: A 57-year-old post-menopausal woman presented with screen-detected left-sided breast cancer. After undergoing all staging and pre-operative studies the patient was proposed for conservative breast surgery with tumor localization. During surgery, an experimental digital and non-invasive intra-operative localization method with augmented reality was compared with the standard pre-operative localization with carbon tattooing (institutional protocol). The breast surgeon wearing an augmented reality headset (Hololens) was able to visualize the tumor location projection inside the patient’s left breast in the usual supine position. Discussion: This work describes, to our knowledge, the first experimental test with a digital non-invasive method for intra-operative breast cancer localization using augmented reality to guide breast conservative surgery. In this case, a successful overlap of the previous standard pre-operative marks with carbon tattooing and tumor visualization inside the patient’s breast with augmented reality was obtained. Conclusion: Breast cancer conservative guided surgery with augmented reality can pave the way for a digital non-invasive method for intra-operative tumor localization.     

The Genomic Heritage of Lymph Node Metastases: Implications for Clinical Management of Patients with Breast Cancer

Background Metastatic breast cancer is an aggressive disease associated with recurrence and decreased survival. To improve outcomes and develop more effective treatment strategies for patients with breast cancer, it is important to understand the molecular mechanisms underlying metastasis. Methods We used allelic imbalance (AI) to determine the molecular heritage of primary breast tumors and corresponding metastases to the axillary lymph nodes. Paraffin-embedded samples from primary breast tumors and matched metastases (n = 146) were collected from 26 patients with node-positive breast cancer involving multiple axillary nodes. Hierarchical clustering was used to assess overall differences in the patterns of AI, and phylogenetic analysis inferred the molecular heritage of axillary lymph node metastases. Results Overall frequencies of AI were significantly higher (P < 0.01) in primary breast tumors (23%) than in lymph node metastases (15%), and there was a high degree of discordance in patterns of AI between primary breast carcinomas and the metastases. Metastatic tumors in the axillary nodes showed different patterns of chromosomal changes, suggesting that multiple molecular mechanisms may govern the process of metastasis in individual patients. Some metastases progressed with few genomic alterations, while others harbored many chromosomal alterations present in the primary tumor. Conclusions The extent of genomic heterogeneity in axillary lymph node metastases differs markedly among individual patients. Genomic diversity may be associated with response to adjuvant therapy, recurrence, and survival, and thus may be important in improving clinical management of breast cancer patients.     

PET/CT复合功能成像系统诊断乳腺癌

目的评价18F-FDG PET复合CT灌注成像诊断原发性乳腺癌的价值。方法对33例疑似乳腺癌病人行CT灌注成像和PET成像检查,并进行真实性和可靠性评价。结果27例乳房肿块经病理证实为乳腺癌。PET/CT诊断乳腺癌的敏感度、特异度和精确度分别为92.6%、100%、93.9%。结论PET/CT诊断乳腺癌有着较高的敏感度、特异度和可靠性,作为一种非侵袭性检查方法,可提供乳腺癌在活体内的代谢和血流灌注信息。     

Aromatase inhibitor-induced inflammatory myopathies: A case series

ObjectiveAromatase inhibitors (AIs) are widely used in the adjuvant therapy setting in patients with estrogen receptor–positive breast cancer. Rheumatic side effects of AIs have been reported, including bone loss, arthralgias, myalgias, and tenosynovitis. There is emerging evidence that AIs are linked to new-onset autoimmune diseases. Here, we describe three novel cases of inflammatory myopathies that occurred during AI therapy.MethodsIn total, three patients with inflammatory myopathy in the setting of AI therapy were treated in the Section of Rheumatology, MD Anderson Cancer Center. All the patients were retrospectively chart reviewed. We obtained verbal consent from the patients for use of their cases for teaching and publication purposes and only de-identified data have been used.ResultsNone of our patients from these three cases had a history of autoimmune disease and all had undergone recent cancer screenings, indicating no re-occurrence of breast cancer and no evidence of new cancer. The completion or discontinuation of AI therapy was associated with the resolution of the myopathies in all three cases.ConclusionsThis case series suggests a link between AIs and the new onset of inflammatory myopathy. If a patient develops an inflammatory myopathy while on an AI, the AI therapy should be considered the possible trigger. If the myopathy cannot be controlled with immunosuppression, then discontinuation of the AI should be considered.     

Mammographic breast density and the risk of breast cancer: A systematic review and meta-analysis

ObjectivesMammographic density is a well-defined risk factor for breast cancer and having extremely dense breast tissue is associated with a one-to six-fold increased risk of breast cancer. However, it is questioned whether this increased risk estimate is applicable to current breast density classification methods. Therefore, the aim of this study was to further investigate and clarify the association between mammographic density and breast cancer risk based on current literature.MethodsMedline, Embase and Web of Science were systematically searched for articles published since 2013, that used BI-RADS lexicon 5th edition and incorporated data on digital mammography. Crude and maximally confounder-adjusted data were pooled in odds ratios (ORs) using random-effects models. Heterogeneity regarding breast cancer risks were investigated using I² statistic, stratified and sensitivity analyses.ResultsNine observational studies were included. Having extremely dense breast tissue (BI-RADS density D) resulted in a 2.11-fold (95% CI 1.84–2.42) increased breast cancer risk compared to having scattered dense breast tissue (BI-RADS density B). Sensitivity analysis showed that when only using data that had adjusted for age and BMI, the breast cancer risk was 1.83-fold (95% CI 1.52–2.21) increased. Both results were statistically significant and homogenous.ConclusionsMammographic breast density BI-RADS D is associated with an approximately two-fold increased risk of breast cancer compared to having BI-RADS density B in general population women. This is a novel and lower risk estimate compared to previously reported and might be explained due to the use of digital mammography and BI-RADS lexicon 5th edition.     

Imaging techniques in breast cancer

Imaging is the mainstay in breast cancer diagnosis, with mammography being the primary screening tool used to detect small/asymptomatic cancers. Symptomatic patients undergo triple assessment which includes a clinical assessment, imaging and, if indicated, an image-guided biopsy for histological diagnosis. We discuss current and emerging breast imaging techniques used in breast cancer care. These include mammography, digital breast tomosynthesis, ultrasound and magnetic resonance. We also discuss image-guided biopsy and miscellaneous image-guided interventions that influence the management of breast cancer care.     

Effectively nursing patients receiving aromatase inhibitor therapy

Inhibiting estrogen production is a common means of preventing breast cancer recurrence. The aromatase inhibitors (AIs) are becoming the preferred treatment over tamoxifen as adjuvant therapy for postmenopausal women with hormone-sensitive early breast cancer. Like all adjuvant therapies, AIs have adverse events (AEs) associated with their use, many of which resemble symptoms common to menopause. Because of the greater efficacy of AIs in preventing breast cancer recurrence over tamoxifen, these AEs may be considered tolerable by many patients and often can be effectively managed and/or prevented. Educating patients about anticipated AEs may help them understand, accept, and cope with these AEs. This article reviews the AEs associated with different adjuvant AI treatments and highlights some strategies to manage them effectively. It also highlights the importance of patient education regarding AI therapy and involvement in treatment decisions, which may lead to better long-term adherence and ultimately to better outcomes.     

不同年龄化疗导致闭经乳腺癌患者内分泌治疗方案的选择

目的研究不同年龄乳腺癌患者发生化疗诱导闭经（CIA）和卵巢功能恢复（OFR）的情况,探讨不同年龄CIA乳腺癌患者内分泌药物的选择。 方法回顾性分析2012年1月至2015年12月于中山大学附属第一医院初次诊断并接受化疗、绝经前、有完整月经来潮情况和性激素检测随访数据的患者共123例。按年龄分为≤40岁组（35例）、>40~45岁组（34例）和>45岁组（54例）,采用一般线性模型来分析各个年龄组的OFR率。 结果≤40岁组患者中26例（74.29%）出现CIA,29例（82.86%）患者在2年内OFR;>40~45岁组及>45岁组全部出现CIA（100.00%）,2年内分别有17例（50.00%）及3例（5.56%）患者OFR,差异均有统计学意义（χ²=24.42、54.97,均P<0.01）。在密切监测卵巢功能的情况下,对19例≥48岁、出现CIA、激素受体阳性高危乳腺癌患者使用芳香化酶抑制剂（AI）。使用AI时患者的中位年龄为51.08岁（48.00~54.25岁）,从闭经到使用AI中位时间为9个月（2~21个月）,中位随访32个月（8~68个月）,无一例出现OFR。 结论对于年龄≤45岁的CIA乳腺癌患者,由于OFR率高,内分泌药物应避免直接选用AI。对于≥51岁的CIA乳腺癌患者观察至少9个月,无OFR可开始使用AI,在使用AI时应定期监测卵巢功能。     

Breast Cancers Found with Digital Breast Tomosynthesis: A Comparison of Pathology and Histologic Grade

To compare the pathology and histologic grading of breast cancers detected with digital breast tomosynthesis to those found with conventional digital mammography. The institutional review board approved this study. A database search for all breast cancers diagnosed from June 2012 through December 2013 was performed. Imaging records for these cancers were reviewed and patients who had screening mammography with tomosynthesis as their initial examination were selected. Five dedicated breast imaging radiologists reviewed each of these screening mammograms to determine whether the cancer was visible on conventional digital mammography or whether tomosynthesis was needed to identify the cancer. A cancer was considered mammographically occult if all five radiologists agreed that the cancer could not be seen on conventional digital mammography. The size, pathology and histologic grading for all diagnosed breast cancers were then reviewed. The Mann–Whitney U and Fisher exact tests were utilized to determine any association between imaging findings and cancer size, pathologic type and histologic grade. Sixty‐five cancers in 63 patients were identified. Ten of these cancers were considered occult on conventional digital mammography and detected with the addition of tomosynthesis. These mammographically occult cancers were significantly associated with Nottingham grade 1 histologic pathology (p = 0.02), were smaller (median size: 6 mm versus 10 mm, p = 0.07) and none demonstrated axillary nodal metastases. Breast cancers identified through the addition of tomosynthesis are associated with Nottingham grade 1 histologic pathology and prognostically more favorable than cancers identified with conventional digital mammography alone.     

The ethical,legal and social implications of using artificial intelligence systems in breast cancer care

Breast cancer care is a leading area for development of artificial intelligence (AI), with applications including screening and diagnosis, risk calculation, prognostication and clinical decision-support, management planning, and precision medicine. We review the ethical, legal and social implications of these developments. We consider the values encoded in algorithms, the need to evaluate outcomes, and issues of bias and transferability, data ownership, confidentiality and consent, and legal, moral and professional responsibility. We consider potential effects for patients, including on trust in healthcare, and provide some social science explanations for the apparent rush to implement AI solutions. We conclude by anticipating future directions for AI in breast cancer care. Stakeholders in healthcare AI should acknowledge that their enterprise is an ethical, legal and social challenge, not just a technical challenge. Taking these challenges seriously will require broad engagement, imposition of conditions on implementation, and pre-emptive systems of oversight to ensure that development does not run ahead of evaluation and deliberation. Once artificial intelligence becomes institutionalised, it may be difficult to reverse: a proactive role for government, regulators and professional groups will help ensure introduction in robust research contexts, and the development of a sound evidence base regarding real-world effectiveness. Detailed public discussion is required to consider what kind of AI is acceptable rather than simply accepting what is offered, thus optimising outcomes for health systems, professionals, society and those receiving care.     

Pathways to breast cancer screening artificial intelligence algorithm validation

As more artificial intelligence (AI)-enhanced mammography screening tools enter the clinical market, greater focus will be placed on external validation in diverse patient populations. In this viewpoint, we outline lessons learned from prior efforts in this field, the need to validate algorithms on newer screening technologies and diverse patient populations, and conclude by discussing the need for a framework for continuous monitoring and recalibration of these AI tools. Sufficient validation and continuous monitoring of emerging AI tools for breast cancer screening will require greater stakeholder engagement and the creation of shared policies and guidelines.     

Timing of Critical Genetic Changes in Human Breast Disease

Background Breast cancer development has been characterized as a nonobligatory sequence of histological changes from normal epithelium through invasive malignancy. Although genetic alterations are thought to accumulate stochastically during tumorigenesis, little is known about the timing of critical mutations. This study examined allelic imbalance (AI) in tissue samples representing a continuum of breast cancer development to examine the evolution of genomic instability. Methods Laser-microdissected DNA samples were collected from histologically normal breast specimens (n = 25), atypical ductal hyperplasia (ADH, n = 16), ductal carcinoma-in-situ (DCIS, n = 37), and stage I to III invasive carcinomas (n = 72). Fifty-two microsatellite markers representing 26 chromosomal regions commonly deleted in breast cancer were used to assess patterns of AI. Results AI frequencies were <5% in histologically normal and ADH specimens, 20% in DCIS lesions, and approximately 25% in invasive tumors. Mann-Whitney tests showed (1) that levels of AI in ADH samples did not differ significantly from those in histologically normal tissues and (2) that AI frequencies in DCIS lesions were not significantly different from those in invasive carcinomas. ADH and DCIS samples, however, differed significantly (P < .0001). Conclusions DCIS lesions contain levels of genomic instability that are characteristic of advanced invasive tumors, and this suggests that the biology of a developing carcinoma may already be predetermined by the in situ stage. Observations that levels of AI in ADH lesions are similar to those in disease-free tissues provide a genomic rationale for why prevention strategies at the ADH level are successful and why cases with ADH involving surgical margins do not require further resection.     

Aromatase inhibitors use and risk for cardiovascular disease in breast cancer patients: A population-based cohort study

BackgroundPrior studies regarding use of Aromatase inhibitors (AIs) and risk for cardiovascular disease (CVD) have shown conflicting results. This retrospective cohort study aimed to investigate whether AIs use affects risk for CVD events in postmenopausal breast cancer survivors.MethodsUsing a retrospective cohort study design, four CVD outcomes; heart failure or cardiomyopathy, arrhythmia, acute ischemic heart disease and ischemic stroke or Transient Ischemic Attack were compared with uni- and multivariate Cox regression analyses according to exposure to endocrine therapy (use of AI, tamoxifen or AI/tamoxifen sequentially) or no endocrine therapy.ResultsIn total 15815 postmenopausal women, surgically treated to early breast cancer during 2006–2012, were included. No significantly increased risk for CVD events was observed in patients with AI use in the whole cohort. However, two subgroup analyses showed increased risk for CVD events in the AI/tamoxifen sequential group; heart failure in patients older than 75 years (Hazard Ratio (HR) 2.44; 95% Confidence Interval (CI): 1.32–4.54) and arrhythmia in patients without prior CVD (HR 1.45; 95% CI: 1.01–2.10). An increased risk for arrhythmia and acute ischemic heart disease in patients with at least four years of AI treatment compared with no or short-time exposure was observed (HR 2.12; 95% CI: 1.40–3.25 for arrhythmia; HR 2.03; 95% CI: 1.15–3.58 for ischemic heart disease).ConclusionOur results indicate an increased risk for ischemic heart disease and arrhythmia in patients treated for more than four years with AIs. This should be considered in the risk-benefit assessment concerning endocrine therapy.     

核磁共振在乳腺癌诊断中的应用

目的：探讨核磁共振（MRI）在乳腺癌诊断中的临床应用价值。方法：对45例乳腺肿物患者进行乳腺MRI检查,分析影像学表现,并对其中17例患者进行病变的增强扫描,对增强扫描曲线作出评价。结果：45例患者经术后病理证实：乳腺癌24例,乳腺良性病变21例,MRI对乳腺癌的术产诊断与病理符合率为96％（23／24）,假阳性率为2％（1／45）。结论：乳腺MRI作为一种新的检查方法,对乳腺癌诊断有较大的临床应用价值。