Intraoperative pathological assessment requires rapid, high-quality stained sections to support surgical decisions. Currently, however, this area lacks cost-effective and rapid staining methods for intraoperative procedures. This study integrated a low-cost phase contrast microscope (PCM) imaging and Pix2Pix Generative Adversarial Network (GAN) to take the microimages and create a virtual staining framework of intraoperative sample-frozen sections for converting label-free phase contrast images (PCIs) into H&E-stained equivalents for the first time. PCIs and H&E-stained whole-slide images were captured from sample-frozen sections of 15 registered patients with breast cancer and compared with their autofluorescence image results. Pix2Pix and CycleGAN models were trained on these images. It’s found that the Pix2Pix GAN excels in PCI and generates a clear nuclear boundary and accurate tissue structure without staining artifacts. Quantitatively, Pix2Pix model provides a higher structural similarity index (SSIM [Formula: see text] 0.57) and peak signal-to-noise ratio (PSNR [Formula: see text] 21.65) in phase contrast mode than CycleGAN in the comparison study. The virtual staining results of PCM integrated with Pix2Pix are clearly superior to those of autofluorescence with Pix2Pix, as well as with CycleGAN. Additionally, the approach offers fast processing (1.5[Formula: see text]s/mm[Formula: see text] and can generate a standard WSI virtual stained section in 3 min. This low-cost virtual staining approach bypasses the cumbersome H&E staining process in conventional pathology, significantly accelerating diagnostic procedures. It’s very possible and of significance for this intelligent intraoperative technology to be extended to other intraoperative pathological assessment scenarios requiring rapid histological assessment.

Accurate identification of malignant surgical margins remains a major challenge in breast cancer surgery due to the lack of rapid and molecularly specific diagnostic tools capable of operating in complex biological environments. Surface-enhanced Raman spectroscopy (SERS) holds promise for molecular detection; however, its application in tissues is limited by nonspecific macromolecular adsorption, signal instability, and biological screening effects. Here, we present the BioGate SERS Chip, a nanobiotechnological platform designed to enable robust, metabolite-oriented SERS detection by regulating molecular access to plasmonic hotspots. The chip features a hierarchical gold nanoparticle architecture that forms bioselective nanochannels, allowing low–molecular-weight metabolites to access electromagnetic hotspot regions while excluding macromolecules responsible for biofouling and signal suppression. This bioselective hotspot-gating strategy couples plasmonic field confinement with size-selective molecular transport, enabling reliable metabolite sensing directly in complex tissue environments. The BioGate platform exhibited a detection limit of 10⁻¹³ M for molecular probes and enabled reliable identification of malignant signatures in mixed cell populations containing as little as 1

Cancer screening can enable early detection and improve survival but a focus on single cancers limits cost-effectiveness. Here we present OMAFound (carcinOMA Finder foundation), a foundation model capable of simultaneous multi-cancer screening at both organ level and patient level using widely accessible non-contrast computed tomography (CT). The model was developed and tested on 325,197 CT volumes from 151,386 patients across 10 Chinese and international datasets, achieving performance comparable to mammography-based approaches for breast cancer detection and matching existing lung-specific models for lung cancer detection. In a prospective multi-centre cohort of 21,601 patients undergoing low-dose CT screening, OMAFound demonstrated balanced accuracy of 82.2% for breast cancer and 88.0% for lung cancer in females, while attaining 86.1% balanced accuracy for lung cancer detection in males. When assisted by OMAFound, 7 generalist radiologists showed improvement in sensitivity (mean increases of 38.9% for breast cancer, 16.0% for lung cancer and 21.3% at patient level), without compromising specificity. These findings highlight the potential of OMAFound as a multi-cancer screening tool to offer robust preventive medicine strategies with minimal costs. A foundation model validated in a prospective multi-centre cohort of 21,601 patients undergoing low-dose computed tomography screening for breast and lung cancer showed comparable performance to existing screening by radiologists.

Breast cancer surgery urgently requires a rapid and objective intraoperative diagnostic strategy, as conventional frozen-section pathology depends on time-consuming staining and subjective interpretation, limiting timely surgical decision-making. However, existing label-free microscopic imaging approaches exclude conventional pathological information and lack diagnostic power. To overcome these limitations, a multimodal fusion framework based on unstained multi-angle orthogonal polarization micro-imaging (OPMI) and bright-field micro-imaging (BFMI) was proposed for rapid intraoperative breast cancer diagnosis. By superimposing and differencing multi-angle polarized images, optical contrast of anisotropic collagen fibers was markedly enhanced without staining, enabling objective visualization of collagen morphology and spatial alterations associated with malignancy. Local radial alignment and increased density of collagen fibers corresponding to invasive progression were identified. A dual-encoder fusion network integrating global and local feature representation improved lesion structure and boundary delineation. The method achieved 91.83% accuracy and area under the curve (AUC) of 0.9295, outperforming unimodal and bimodal approaches, offering a promising label-free intraoperative decision-support strategy.

Investigating the essential function of CD300LG within the tumor microenvironment in triple-negative breast cancer (TNBC). Transcriptomic and single-cell data from TNBC were systematically collected and integrated. Four machine learning algorithms were employed to identify distinct target genes in TNBC patients. Specifically, CIBERSORT and ssGSEA algorithms were utilized to elucidate immune infiltration patterns, whereas TIDE and TCGA algorithms predicted immune-related outcomes. Moreover, single-cell sequencing data were analyzed to investigate the function of CD300LG-positive cells within the tumor microenvironment. Finally, immunofluorescence staining confirmed the significance of CD300LG in tumor phenotyping. After machine learning screening and independent dataset validation, CD300LG was identified as a unique prognostic biomarker for triple-negative breast cancer. Enrichment analysis revealed that CD300LG expression is strongly linked to immune infiltration and inflammation-related pathways, especially those associated with the cell cycle. The presence of CD8+ T cells and M1-type macrophages was elevated in the CD300LG higher group, whereas the abundance of M2-type macrophage infiltration showed a significant decrease. Immunotherapy prediction models indicated that individuals with low CD300LG expression exhibited better responses to PD-1 therapy. Additionally, single-cell RNA sequencing and immunofluorescence analyses uncovered a robust association between CD300LG and genes involved in tumor invasion. CD300LG plays a pivotal role in the tumor microenvironment of TNBC and represents a promising therapeutic target.

Machine learning models for the diagnosis of breast cancer can facilitate the prediction of cancer risk and subsequent patient management among other clinical tasks. For the models to impact clinical practice, they ought to follow standard workflows, help interpret mammography and ultrasound data, evaluate clinical contextual information, handle incomplete data and be validated in prospective settings. Here we report the development and testing of a multimodal model leveraging mammography and ultrasound modules for the stratification of breast cancer risk based on clinical metadata, mammography and trimodal ultrasound (19,360 images of 5,216 breasts) from 5,025 patients with surgically confirmed pathology across medical centres and scanner manufacturers. Compared with the performance of experienced radiologists, the model performed similarly at classifying tumours as benign or malignant and was superior at pathology-level differential diagnosis. With a prospectively collected dataset of 191 breasts from 187 patients, the overall accuracies of the multimodal model and of preliminary pathologist-level assessments of biopsied breast specimens were similar (90.1% vs 92.7%, respectively). Multimodal models may assist diagnosis in oncology. A multimodal model for the stratification of breast cancer risk based on clinical metadata, mammography and trimodal ultrasound images performed as well as or better than radiologists at tumour classification and at differential diagnosis.

Abstract Background Though deep learning has consistently demonstrated advantages in the automatic interpretation of breast ultrasound images, its black-box nature hinders potential interactions with radiologists, posing obstacles for clinical deployment. Methods We proposed a domain knowledge-based interpretable deep learning system for improving breast cancer risk prediction via paired multimodal ultrasound images. The deep learning system was developed on 4320 multimodal breast ultrasound images of 1440 biopsy-confirmed lesions from 1348 prospectively enrolled patients across two hospitals between August 2019 and December 2022. The lesions were allocated to 70% training cohort, 10% validation cohort, and 20% test cohort based on case recruitment date. Results Here, we show that the interpretable deep learning system can predict breast cancer risk as accurately as experienced radiologists, with an area under the receiver operating characteristic curve of 0.902 (95% confidence interval = 0.882 – 0.921), sensitivity of 75.2%, and specificity of 91.8% on the test cohort. With the aid of the deep learning system, particularly its inherent explainable features, junior radiologists tend to achieve better clinical outcomes, while senior radiologists experience increased confidence levels. Multimodal ultrasound images augmented with domain knowledge-based reasoning cues enable an effective human-machine collaboration at a high level of prediction performance. Conclusions Such a clinically applicable deep learning system may be incorporated into future breast cancer screening and support assisted or second-read workflows.

Background: Triple-negative breast cancer (TNBC) is a distinct subtype of breast cancer, accounting for 12-18% of all breast cancer cases. It exhibits high heterogeneity and aggressiveness, resulting in a poorer prognosis with a high risk of early recurrence and metastasis. Due to the lack of expression of estrogen receptors (ER), progesterone receptors (PR), and human epidermal growth factor receptor 2 (HER2), as well as insensitivity to endocrine therapy, determining a standard treatment for TNBC is challenging. The identification of potential prognostic biomarkers is crucial for developing personalized treatment strategies for patients. Material/Methods: Our study investigated the potential value of HSP90 a in TNBC prognosis. A retrospective analysis was conducted on 127 TNBC patients and 127 Healthy controls from March 1, 2019 to July 31, 2022. Venous blood was collected and tested for HSP90 a, CEA, CA199, and CA125, and we recorded the clinical characteristics of the patients, including age, BMI, alcohol consumption status, surgical history, CEA level, CA199 level, CA125 level, HSP90 a level, tumor size, distant metastases, lymph node metastasis, and TNM stage. Univariate and multivariate methods were used to screen independent risk factors for progression-free survival (PFS) and overall survival (OS). Results: HSP90 a is not only upregulated in TNBC but is also highly correlated with lymph node metastasis and TNM stage. The results of multivariate analysis showed that distant metastasis, TNM stage and HSP90 a level were independent factors associated with PFS. BMI, tumor size, TNM stage, surgical history, and HSP90 a level were independent factors influencing OS. Conclusions: Our research findings demonstrate a significant association between high HSP90 a expression and adverse clinical features, suggesting a poorer prognosis for TNBC patients.

Background:Commonly, the thyroid gland is regarded as an organ with fewer metastatic diseases, and colorectal metastasis to the thyroid (CMT) is rarely reported, especially, with that the clinical sign of thyroid metastasis nidus is the chief complaint. The CMT occurs in advanced colorectal cancer and is associated with poor prognosis and short survival.Case Report:In this case, we reported a patient with the sign of neck mass as the first manifestation of CMT. The patient underwent a partial thyroidectomy in June 2019, immunohistochemical findings of thyroid carcinoma suggested the possibility of adenocarcinoma of gastrointestinal tract. The patient underwent a colonoscopy in July 2019 and a colonic mass was found. Pathological examination diagnosed rectal adenocarcinoma. The patient underwent neoadjuvant chemotherapy, surgical treatment, postoperative adjuvant chemotherapy and targeted therapy. The patient died in June 2022.Conclusion:The metastasis disease would not be ignored at all, when a patient complains at signs of neck mass. Further, the possibility of metastasis cancer should be considered once thyroid nodules occur in patients with colorectal cancer. Even though the biological characteristics and stage of the primary tumor have an important impact on the prognosis, positive standardized treatments can also be helpful.

Intracellular reactive oxygen species (ROS) are closely associated with cancer cell types. Therefore, ROS-based pattern recognition is a promising strategy for precise diagnosis of cancer, but such a possibility has never been reported yet. Herein, we proposed an ROS-responsive fluorescent sensor array based on pH-controlled histidine-templated gold nanoclusters (AuNCs@His) to distinguish cancer cell types and their proliferation states. In this strategy, three types of AuNCs@His with diverse fluorescence profiles were first synthesized by only adjusting the pH value. Upon the addition of various ROS, fluorescence quenching of three types of AuNCs@His occurred with different degrees, thereby forming unique optical "fingerprints", which were well-clustered into several separated groups without overlap by principal component analysis (PCA). The sensing mechanism was attributable to the oxidation of AuNCs@His by ROS, as revealed by X-ray photoemission spectroscopy, Fourier transform infrared spectroscopy, 1H nuclear magnetic resonance spectroscopy, and electrospray ionization mass spectrometry. Based on the ROS-responsive sensing pattern, cancer cell types were successfully differentiated via PCA with 100% accuracy. Additionally, the proposed sensor array exhibited excellent performance in distinguishing the proliferation states of cancer cells, which was supported by the results of the Ki-67 immunohistochemistry assay. Overall, the ROS-responsive fluorescent sensor array can serve as a promising tool for precise diagnosis of cancer, indicating great potential for clinical application.

BackgroundAI-based clinical decision support system (CDSS) has important prospects in overcoming the current informational challenges that cancer diseases faced, promoting the homogeneous development of standardized treatment among different geographical regions, and reforming the medical model. However, there are still a lack of relevant indicators to comprehensively assess its decision-making quality and clinical impact, which greatly limits the development of its clinical research and clinical application. This study aims to develop and application an assessment system that can comprehensively assess the decision-making quality and clinical impacts of physicians and CDSS.MethodsEnrolled adjuvant treatment decision stage early breast cancer cases were randomly assigned to different decision-making physician panels (each panel consisted of three different seniority physicians in different grades hospitals), each physician made an independent “Initial Decision” and then reviewed the CDSS report online and made a “Final Decision”. In addition, the CDSS and guideline expert groups independently review all cases and generate “CDSS Recommendations” and “Guideline Recommendations” respectively. Based on the design framework, a multi-level multi-indicator system including “Decision Concordance”, “Calibrated Concordance”, “ Decision Concordance with High-level Physician”, “Consensus Rate”, “Decision Stability”, “Guideline Conformity”, and “Calibrated Conformity” were constructed.Results531 cases containing 2124 decision points were enrolled; 27 different seniority physicians from 10 different grades hospitals have generated 6372 decision opinions before and after referring to the “CDSS Recommendations” report respectively. Overall, the calibrated decision concordance was significantly higher for CDSS and provincial-senior physicians (80.9%) than other physicians. At the same time, CDSS has a higher “ decision concordance with high-level physician” (76.3%-91.5%) than all physicians. The CDSS had significantly higher guideline conformity than all decision-making physicians and less internal variation, with an overall guideline conformity variance of 17.5% (97.5% vs. 80.0%), a standard deviation variance of 6.6% (1.3% vs. 7.9%), and a mean difference variance of 7.8% (1.5% vs. 9.3%). In addition, provincial-middle seniority physicians had the highest decision stability (54.5%). The overall consensus rate among physicians was 64.2%.ConclusionsThere are significant internal variation in the standardization treatment level of different seniority physicians in different geographical regions in the adjuvant treatment of early breast cancer. CDSS has a higher standardization treatment level than all physicians and has the potential to provide immediate decision support to physicians and have a positive impact on standardizing physicians’ treatment behaviors.

Abstract Breast cancer is the most common global malignancy and the leading cause of cancer deaths. CDC73 (Human cell division cycle 73), a nuclear protein, participates transcription regulation and its functions are controversial in malignancies. CDC73 has been reported to be upregulated in breast cancer. The underlying mechanism, however, has not been fully illuminated. In breast cancer, CDC73 could promote the proliferation of tumor cells, and the expression of CDC73 was related to poor prognosis in patients. Here, we found that CBL, an E3 ubiquitin ligase, could interact with CDC73 and promote MAPK1 ubiquitination and degradation of this protein. In addition, silencing MAPK1 led to a suppression of breast cancer cell growth in vitro and in vivo , and even abolished the promoting effects of CDC73 overexpression. We also found that mTOR pathway played a role in CDC73-mediated breast cancer. mTOR pathway inhibitor reversed cell phenotypes induced by CDC73 overexpression. Our study revealed the underlying mechanism of CDC73 in breast cancer: it promoted MAPK1 ubiquitination and degradation so that affected MAPK1 level and subsequently led to tumor progression, providing a novel therapeutic strategy to combat cancer.

BACKGROUND:Before the discovery of cuproptosis, copper-loaded nanoparticle is a wildly applied strategy for enhancing the tumor-cell-killing effect of chemotherapy. Although copper(ii)-related researches are wide, details of cuproptosis-related bioprocess in pan-cancer are not clear yet now, especially for prognosis and drug sensitivity prediction yet now.METHODS:In this study, VOSviewer is used for the literature review, and R4.2.0 is used for data analysis. Public data are collected from TCGA and GEO, local breast cancer cohort is collected to verify the expression level of CDKN2A.RESULTS:7036 published articles exhibited a time-dependent linear relationship (R=0.9781, p<0.0001), and breast cancer (33.4%) is the most researched topic. Cuproptosis-related-genes (CRGs)-based unsupervised clustering divides pan-cancer subgroups into four groups (CRG subgroup) with differences in prognosis and tumor immunity. 44 tumor-driver-genes (TDGs)-based prediction model of drug sensitivity and prognosis is constructed by artificial intelligence (AI). Based on TDGs and clinical features, a nomogram is (C- index: 0.7, p= 6.958e- 12) constructed to predict the prognosis of breast cancer. Importance analysis identifies CDKN2A has a pivotal role in AI modeling, whose higher expression indicates worse prognosis in breast cancer. Furthermore, inhibition of CDKN2A down-regulates decreases Snail1, Twist1, Zeb1, vimentin and MMP9, while E-cadherin is increased. Besides, inhibition of CDKN2A also decreases the expression of MEGEA4, phosphorylated STAT3, PD-L1, and caspase3, while cleaved-caspase3 is increased. Finally, we find down-regulation of CDKN2A or MAGEA inhibits cell migration and wound healing, respectively.CONCLUSIONS:AI identified CRG subgroups in pan-cancer based on CRGs-related TDGs, and 44-gene-based AI modeling is a novel tool to identify chemotherapy sensitivity in breast cancer, in which CDKN2A/MAGEA4 pathway played the most important role.

BACKGROUND:Altered epigenetic reprogramming and events contribute to breast cancer (Bca) progression and metastasis. How the epigenetic histone demethylases modulate breast cancer progression remains poorly defined. We aimed to elucidate the biological roles of KDM4A in driving Notch1 activation and Bca progression.METHODS:The KDM4A expression in Bca specimens was analyzed using quantitative PCR and immunohistochemical assays. The biological roles of KDM4A were evaluated using wound-healing assays and an in vivo metastasis model. The Chromatin Immunoprecipitation (ChIP)-qPCR assay was used to determine the role of KDM4A in Notch1 regulation.RESULTS:Here, we screened that targeting KDM4A could induce notable cell growth suppression. KDM4A is required for the growth and progression of Bca cells. High KDM4A enhances tumor migration abilities and in vivo lung metastasis. Bioinformatic analysis suggested that KDM4A was highly expressed in tumors and high KDM4A correlates with poor survival outcomes. KDM4A activates Notch1 expressions via directly binding to the promoters and demethylating H3K9me3 modifications. KDM4A inhibition reduces expressions of a list of Notch1 downstream targets, and ectopic expressions of ICN1 could restore the corresponding levels. KDM4A relies on Notch1 signaling to maintain cell growth, migration and self-renewal capacities. Lastly, we divided a panel of cell lines into KDM4Ahigh and KDM4Alow groups. Targeting Notch1 using specific LY3039478 could efficiently suppress cell growth and colony formation abilities of KDM4Ahigh Bca.CONCLUSION:Taken together, KDM4A could drive Bca progression via triggering the activation of Notch1 pathway by decreasing H3K9me3 levels, highlighting a promising therapeutic target for Bca.

研究保乳术联合腺体筋膜瓣转移成型术的临床效果.收集2015年3月至2019年7月收治的71例早期乳腺癌患者病例资料,按照治疗方式差异分成联合组(31例)和传统组(40例).传统组行传统手术治疗,联合组予以保乳术联合腺体筋膜瓣转移成型术.比较两组各项手术指标、乳房美容效果、术后并发症发生情况及术后复发率等差异.联合组手术时间以及术后住院时间均短于传统组,且术中出血量少于传统组(均P<0.05).联合组乳房美容优良率高于传统组(P<0.05).联合组术后并发症总发生率低于传统组(P<0.05).两组术后1年复发率比较差异无统计学意义(P>0.05).保乳术联合腺体筋膜瓣转移成型术临床效果优于传统手术,且术后并发症发生率较低,乳房形态较好,不增加复发率,具有临床推广价值.

目的 探讨静脉治疗护理亚专科团队的建设和实践效果.方法 我院于2019年7月正式成立静脉治疗护理亚专科,通过建立静脉治疗团队、明确静脉治疗亚专科的工作内容与管理机制来实现静脉治疗亚专科的运行.结果 通过2年的静脉治疗护理亚专科团队构建与管理,临床护士的静脉导管维护能力显著改善(P＜0.05)、科研创新能力较前提高;输液工具选择合理率、输液接头使用正确率、导管固定正确率、敷料维护正确率、穿刺点发红发生率较实施前有所改善,差异有统计学意义(均P＜0.05);项目实施后静脉血管通路门诊工作量较前增加,患者和护士满意度得分显著提高(均P＜0.05).结论 静脉治疗护理亚专科的建设与实践不仅提升了专科护理人员综合水平,在为患者提供安全、优质静脉治疗护理服务的同时也满足了患者多元化治疗需求,进一步提高专科护理质量,深化专科内涵,促进静脉治疗专科不断向纵深发展.

目的 探究乳腺癌患者术后上肢三阶段康复方案的应用效果,为减轻患者身心痛苦、提高生活质量提供依据.方法 2020 年 7－12 月,采用便利抽样法选取合肥市某三级甲等医院日间病房收治的 80 例行术前新辅助化疗的乳腺癌患者为研究对象,按照入院时间先后分为对照组和试验组,各40 例,对照组实施常规护理,试验组实施上肢三阶段康复方案,比较两组患者术后 3 个月上肢功能锻炼依从性评分、上肢功能评分及生活质量评分的差异.结果 术后 3 个月,试验组上肢功能锻炼依从性量表和生活质量测定量表各个维度得分均明显高于对照组,上肢功能评定表得分明显低于对照组(均P＜0.001).结论 上肢三阶段康复方案,能增加乳腺癌患者术后功能锻炼依从性,促进其上肢功能的良好恢复,提高术后生活质量.

The contributions of long non-coding RNAs (lncRNAs) and microRNAs (miRNAs) to breast cancer are critical areas of investigation. In this study, we identified a novel lncRNA RP11-283G6.5 which was lowly expressed in breast cancer and whose low expression was correlated with poor overall survival and disease-free survival of breast cancer patients. Functional experiments revealed that ectopic expression of RP11-283G6.5 confined breast cancer cellular growth, migration, and invasion, and promoted cellular apoptosis. Conversely, RP11-283G6.5 silencing facilitated breast cancer cellular growth, migration, and invasion, and repressed cellular apoptosis. Moreover, RP11-283G6.5 was found to confine breast cancer tumour growth and metastasis in vivo. Mechanistically, RP11-283G6.5 competitively bound to ILF3, reduced the binding of ILF3to primary miR-188 (pri-miR-188), abolished the suppressive effect of ILF3 on pri-miR-188 processing, and therefore promoted pri-miR-188 processing, leading to the reduction of pri-miR-188 and the upregulation of mature miR-188-3p. The expression of RP11-283G6.5 was significantly positively correlated with that of miR-188-3p in breast cancer tissues. Through increasing miR-188-3p, RP11-283G6.5 decreased TMED3, a target of miR-188-3p. RP11-283G6.5 further suppressed Wnt/β-catenin signalling via decreasing TMED3. Rescue assays revealed that inhibition of miR-188-3p, overexpression of TMED3 or blocking Wnt/β-catenin signalling all attenuated the roles of RP11-283G6.5 in breast cancer. Collectively, these findings demonstrated that RP11-283G6.5 is a tumour suppressive lncRNA in breast cancer via modulating miR-188-3p/TMED3/Wnt/β-catenin signalling. This study indicated that RP11-283G6.5 might be a promising prognostic biomarker and therapeutic target for breast cancer.

目的 利用剪切波弹性成像技术评估乳腺肿块的各向异性,探讨各向异性因子(AF)及其联合BI-RADS分类对乳腺肿块的诊断价值.方法 收集102例患者共106个肿块在最大径线及垂直切面的弹性定量指标,分析AFs及其联合BI-RADS分类对乳腺良恶性肿块的诊断价值.结果 恶性肿块的AFs明显高于良性肿块(P＜0.05).BI-RADS分类及AFratio诊断的灵敏度、特异度、准确度、Kappa值分别为85.4％、75.9％、80.2％、0.605及72.9％、86.2％、80.2％、0.597.二者联合诊断的灵敏度、特异度、准确度、Kappa值分别为83.3％、89.7％、86.8％、0.733.联合诊断的Kappa值最高,特异度较单一BI-RADS分类高(x2=3.876,P=0.049＜0.05).结论 AF可作为乳腺癌的诊断指标,AFratio联合BI-RADS分类可提高乳腺肿块诊断的特异度及可靠性,有一定的应用价值,值得进一步研究.