Segmentation Method of Breast Masses on Ultrasonographic Images Using Level Set Method Based on Statistical Model

It is important to segment mass region accurately in a computer-aided diagnosis (CADx) scheme for evaluating the likelihood of malignancy of the mass on ultrasonographic breast image. The purpose of this study was to develop a novel level set method for segmentation of breast mass on ultrasonographic image. Our database consisted of 151 ultrasonographic images with 70 malignant and 81 benign breast masses. In a novel level set method, an energy function was defined with region-based, edge-based, and regularizing terms. The region-based term analyzed global information, whereas the edge-based term analyzed local information. The regularizing term also controlled the length of the boundary curve. The region of breast mass was segmented so that the energy based on those terms was minimized. With our proposed method, true positive (TP) ratio, false positive (FP) ratio, jaccard similarity (JS), and Dice similarity coefficient (DSC) were 92.2%, 9.1%, 84.2%, and 91.3%, respectively. These results tended to be substantially higher than those with two conventional segmentation methods. Our proposed method based on the novel level set method was shown to segment mass region accurately on ultrasonographic breast image.

Evaluation of Breast Masses Using Mammography and Sonography as First Line Investigations

Purpose: To study the specificity of mammography and ultrasonography separately and in combination for detection of breast masses (ultrasonography-mammography correlation);To study the investigations to evaluate various breast masses;To describe suitable indications, advantages and limitations of each technique compared with other available modalities;To study the mimics of breast masses;To have histopathology follow-up and retrospective evaluation with imaging findings to improve diagnostic skills in series of 166 patients complaining of breast mass. Material: The prospective clinical study was carried out in the department of Radiodiagnosis for a period of 2 year extending from December 2010 to December 2012 infemale patients complaining of breast mass. Well informed written consent was obtained from them. Histopathology follow up was obtained from either biopsy or post operative tissue. USG machine: Philips HD 11 XE USG of the breasts and axillary region done in supine position in presence of female attendant;Mammography machine: Allengers machine with Agfa special mammography cassettes. Cranio caudal and Medio-Lateral Oblique views are taken in the presence of female attendant. MRI: PHILIPS 1.5 T machine;CT: SIEMENS duel slice CT machine. Results: Ultrasonography and mammography was done in most of the cases were sufficient to diagnose the lesion in most of the cases especially in benign breast masses. MRI and CT scan was used in special cases to know the extent of the lesions, in mimics of breast masses, bony extensions, primary muscular and bony lesions. Total 166 patients complaining of breast mass in one or both breasts were examined and evaluated with USG and mammography. The lesions were confirmed on histopathology (FNAC/biopsy). Out of 30 diagnosed malignancies two lesions were missed on mammography and four lesions were missed on ultrasonography. One of them was missed on both. For malignancies specificity of mammography is 93.3% and that of ultrasonography is 86.67%. Combining both the modalities specificity is near 97%. Out of total 92 abnormal breasts 12 were missed on USG and 20 were missed on mammography. Combining both the modalities only 2 lesions were missed and were diagnosed on histopathology alone. Overall specificity for USG in breast masses is 86.9% and for mammography it is 78.6%. Combining both the modalities the specificity is 97.6%. The “p” value is obtained which is highly significant for combination of ultrasonography and mammography in comparison with any individual modality (p = 0.0059 & p = 0.0001 respectively). Conclusion: Our study confirms the higher combined sensitivity rate for ultrasonography and mammography for detection of breast masses including malignancies. USG is useful in cystic lesions, ectasias, infections, pregnancy-lactation, and dense breast evaluation and for image guidance, whereas mammography is useful in detecting microcalcifications, spiculated masses for early detection of malignancies and for stereotactic biopsies. To suggest single modality, ultrasonography is better in younger population and BIRAD 1, 2 & 3 lesions. Whereas, mammography is better in older population and BIRAD 4 & 5 lesions. However, sono-mammographic correlation is best in both.

Complex Cystic Breast Masses: An Ultrasound Imaging Review

Cystic lesions are very commonly encountered entities in the breast. Among these, Complex Cystic Breast Masses (CCBMs), which contain both anechoic and echogenic components, can result in a variety of imaging appearances. These include cystic lesions with thick walls and/or internal septa, intracystic nodular lesions, and solid-cystic masses of varying compositions. Ultrasound is the mainstay for evaluating cystic lesions, and thus recognizing the imaging features appropriately and suggesting suitable interventional procedures are included in their management. In this pictorial essay, we describe the wide range of ultrasound appearances of CCBMs with a number of clinically encountered examples from our institution. This article would enhance the understanding of readers in possible differentials to be included in their clinical practice and to suggest appropriate further intervention, when deemed necessary.

The Role of RNA Epitranscriptomics and the RNA Fat Mass and Obesity-Associated Demethylase in Triple Negative Breast Cancer

Breast cancer is one of the most commonly diagnosed cancers and one of the most significant sources of cancer mortality. Triple negative breast cancer (TNBC) is a particularly aggressive subtype that has proven difficult to treat with standard chemotherapies. Obesity has also been shown to exacerbate breast cancer, and diagnoses of these two diseases frequently overlap. Both conditions are regulated in part by the fat mass and obesity-associated (FTO) demethylase, an RNA demethylase which may drive breast cancers through epigenetic alterations to gene expression. Methods of inhibiting FTO have been researched in vitro and in vivo as an alternative or adjunct to chemotherapies in multiple cancers, including breast cancer. Translating knowledge of the role of FTO in breast cancer and the development of novel agents may allow for improvements in the treatment of this refractory cancer. This review therefore aims to provide an overview of existing and developing chemical inhibitors of FTO that could be innovatively studied for the treatment of TNBC and associated comorbidity.

Body Mass Index and Breast Cancer Risk: A Retrospective Multi-Institutional Analysis in Saudi Arabia

Background: The purpose of this study was to evaluate the association between the body mass index (BMI) of breast cancer patients and non-cancer females of the Eastern Province of Saudi Arabia. Methods: The weight, height and age was obtained from the patient records of 706 newly diagnosed breast cancer patients and of 20,872 non-cancer female patients who consulted the two largest hospitals in the Eastern Province of Saudi Arabia between 2006 and 2012. Factorial analysis of variance (ANOVA) was used to assess the association between the BMI, age and breast cancer status. Results: The mean BMI of the non-cancer females was 29.4 and the percentage of obese patients of the different age groups ranged from 23.9% to 66.5%. The BMI increased significantly with age. The ANOVA revealed that breast cancer patients older than 50 years had a significantly lower BMI compared to their non-cancer counterparts (p = 0.01). Conclusion: Our data confirm the high BMI of the Saudi Arabian female population. The reason for our finding of a lower BMI of postmenopausal breast cancer patients compared to their non-cancer counterparts is unclear. Future studies are warranted to assess the impact of possible confounding factors on the association between obesity and breast cancer risk. An interesting factor to investigate in future studies would particularly be the use of the anti-diabetic and cancer-protective drug metformin considering that diabetes mellitus is endemic in Saudi Arabia with a prevalence of 30%.

The Effect of Body Mass Index on Survival in Breast Cancer

Background: Obesity is a well-known risk factor for breast cancer recurrence and poor prognosis. The objective of this study was to evaluate the effect of body mass index (BMI) on survival in breast cancer patients. Methods: We performed a retrospective analysis of 50 breast cancer patients treated in our hospital from January 2012 to December 2013. Patients were divided according to body mass index when diagnosed into: normal weight BMI 25 Kg/m2, over weight BMI ≥ 25 Kg/m2 to 2, obesity BMI ≥ 30 Kg/m2. In this study the effect of body mass index on progression free survival (PFS) and overall survival (OS) was evaluated. Results: The disease free survival (DFS) and overall survival (OS) decreased in overweight and obese patients. Both overweight and obesity were predictors for increased risks of breast cancer relapse and mortality with a median disease free survival for overweight 29 mons and obese patients 11 mons and a median overall survival for overweight patients 49 mons and obese patients 39 mons. Conclusion: Obesity and overweight are associated with poorer disease free survival and overall survival in patients with breast cancer.

Analysis of Machine Learning Techniques Applied to the Classification of Masses and Microcalcification Clusters in Breast Cancer Computer-Aided Detection

Breast cancer is one of the most common and deadliest types of cancer among women and early detection is of major importance to decrease mortality rates. Microcalcification clusters and masses are two major indicators of malignancy in the early stages of this disease, when mammography is typically used as the screening technology. Computer-Aided Diagnosis (CAD) systems can support the radiologists’ work, by performing a double-reading process, which provides a second opinion that the physician can take into account in the detection process. This paper presents a CAD model based on computer vision procedures for locating suspicious regions that are later analyzed by artificial neural networks, support vector machines and linear discriminant analysis, to classify them into benign or malignant, based on a set of features that are extracted from lesions to characterize their visual content. A genetic algorithm is used to find the subset of features that provide the greatest discriminant power. Our results show that the SVM presented the highest overall accuracy and specificity for classifying microcalcification clusters, while the NN outperformed the rest for mass-classification in the same parameters. Overall accuracy, sensitivity and specificity were measured.

Breast non-mass-like lesions on contrast-enhanced ultrasonography: Feature analysis, breast image reporting and data system classification assessment

BACKGROUND Breast non-mass-like lesions(NMLs)account for 9.2%of all breast lesions.The specificity of the ultrasound diagnosis of NMLs is low,and it cannot be objectively classified according to the 5th Edition of the Breast Imaging Reporting and Data System(BI-RADS).Contrast-enhanced ultrasound(CEUS)can help to differentiate and classify breast lesions but there are few studies on NMLs alone.AIM To analyze the features of benign and malignant breast NMLs in grayscale ultrasonography(US),color Doppler flow imaging(CDFI)and CEUS,and to explore the efficacy of the combined diagnosis of NMLs and the effect of CEUS on the BI-RADS classification of NMLs.METHODS A total of 51 breast NMLs verified by pathology were analyzed in our hospital from January 2017 to April 2019.All lesions were examined by US,CDFI and CEUS,and their features from those examinations were analyzed.With pathology as the gold standard,binary logic regression was used to analyze the independent risk factors for malignant breast NMLs,and a regression equation was established to calculate the efficiency of combined diagnosis.Based on the regression equation,the combined diagnostic efficiency of US combined with CEUS(US+CEUS)was determined.The initial BI-RADS-US classification of NMLs was adjusted according to the independent risk factors identified by CEUS,and the diagnostic efficiency of CEUS combined with BI-RADS(CEUS+BI-RADS)was calculated based on the results.ROC curves were drawn to compare the diagnostic values of the three methods,including US,US+CEUS,and CEUS+BI-RADS,for benign and malignant NMLs.RESULTS Microcalcification,enhancement time,enhancement intensity,lesion scope,and peripheral blood vessels were significantly different between benign and malignant NMLs.Among these features,microcalcification,higher enhancement,and lesion scope were identified as independent risk factors for malignant breast NMLs.When US,US+CEUS,and CEUS+BI-RADS were used to identify the benign and malignant breast NMLs,their sensitivity rates were 82.6%,91.3%,and 87.0%,respectively;their specificity rates were 71.4%,89.2%,and 92.9%,respectively;their positive predictive values were 70.4%,87.5%,and 90.9%,respectively;their negative predictive values were 83.3%,92.6%,and 89.7%,respectively;their accuracy rates were 76.5%,90.2%,and 90.2%,respectively;and their corresponding areas under ROC curves were 0.752,0.877 and 0.903,respectively.Z tests showed that the area under the ROC curve of US was statistically smaller than that of US+CEUS and CEUS+BI-RADS,and there was no statistical difference between US+CEUS and CEUS+BI-RADS.CONCLUSION US combined with CEUS can improve diagnostic efficiency for NMLs.The adjustment of the BI-RADS classification according to the features of contrastenhanced US of NMLs enables the diagnostic results to be simple and intuitive,facilitates the management of NMLs,and effectively reduces the incidence of unnecessary biopsy.

Novel Approach for Quantitative Measurement of Matrix Metalloprotease-1 (MMP1) in Human Breast Cancer Cells Using Mass Spectrometry

Identification and quantification of low abundance growth factors and regulators in complex biological samples still present a challenging task in analytical biochemistry. Immunoassays are often used for such purpose but immunoassays face limitation of both availability and qualities of antibody reagents that are necessary for development of immune assays. With genomics data base available, mass spectrometry (MS) can analyze protein tryptic peptides directly for quantitative determination of proteins. In this study, we report a method for detection of matrix metalloproteinase 1 (MMP1), an important extracellular matrix modulator, in human breast cancer cells by quadrupole time-of-flight (Q-TOF) MS. Absolute quantification of MMP1 was conducted using the selected reaction monitoring (SRM) on a triple quadrupole (Triple-Quad) MS via transitions selected from MMP1 tryptic peptides using non isotope labeled MMP1 protein as a titration standard. In comparison with immune based assay, this MS method showed picogram level sensitivity for quantitative determination of MMP1 intotal cell lysates. Our results demonstrated the feasibility of absolute quantification of low abundance proteins using label-free protein standard by mass spectrometry. Therefore, this method provides not only advantages of high sensitivity but also cost saving in comparison with the commonly used mass spectrometry that currently employs isotype labeled proteins for quantitative analysis.

GZMB在乳腺癌和乳腺良性肿物患者中的表达及与T淋巴细胞水平的关系研究

目的:探讨颗粒酶B(granzyme B,GZMB)在乳腺癌和乳腺良性肿物患者中的表达及与T淋巴细胞水平的关系。方法:选取2022年1月至2023年8月南阳市第一人民医院收治的乳腺癌患者63例为观察组,根据肿瘤分期分为T_(1)N_(0)M_(0)组、T_(2)N_(0)M_(0)组、T_(2)N_(1)M_(0)组、T_(3)N_(0)M_(0)组及T_(3)N_(1)M_(0)组;选择同期治疗的乳腺良性肿物患者9例为对照组;采用流式细胞仪测定各组GZMB及T淋巴细胞水平;对乳腺癌患者GZMB与T淋巴细胞参数完成相关性分析。结果:观察组乳腺癌患者GZMB水平高于对照组(P<0.05);CD_(3)^(+)、CD_(4)^(+)及CD_(4)^(+)/CD_(8)^(+)水平低于对照组(P<0.05);乳腺癌不同分期下CD_(8)^(+)水平无统计差异(P>0.05);乳腺癌患者中T_(3)N_(1)M_(0)组GZMB水平高于其余4组(P<0.05),CD_(3)^(+)、CD_(4)^(+)及CD_(4)^(+)/CD_(8)^(+)水平低于其余4组(P<0.05);乳腺癌患者分期越高,GZMB水平越高,CD_(3)^(+)、CD_(4)^(+)及CD_(4)^(+)/CD_(8)^(+)水平越低;Pearson相关性结果表明,乳腺癌患者GZMB水平与CD_(3)^(+)、CD_(4)^(+)及CD_(4)^(+)/CD_(8)^(+)水平呈负相关(P<0.05)。结论:GZMB在乳腺癌患者中呈高表达,其表达水平在不同分期中存在差异,且与T淋巴细胞水平存在相关性,可指导临床诊疗。

分子分型在肿块型及非肿块型乳腺癌中的差异及其与X线征像关系研究

目的 探讨肿块型及非肿块型乳腺癌的临床及分子分型差异、X线征象与分子分型的关系。方法 纳入372例乳腺X线表现为肿块型或非肿块型并经病理证实为乳腺癌的女性患者的临床、病理及影像资料,其中肿块型共210例,非肿块型共162例,回顾分析两组临床及分子分型差异及不同分子分型乳腺癌X线征象差异。结果非肿块型病灶发生于相对低龄者,非浸润性癌占比高;肿块型患者相对高龄,以浸润性癌占比更高。Luminal B型多见于肿块型,Luminal HER-2型多见于非肿块型。不同分子分型的肿块型乳腺癌的影像特征如肿块是否伴随不对称及非肿块型乳腺癌的影像特征如结构扭曲、钙化等征象发生率、钙化形态等差异有统计学意义(P均<0.05)。结论 肿块型及非肿块型乳腺癌临床、病理特征及分子分型不同,部分X线征象与肿瘤分子分型有关,对治疗方案选择、预后判断等有一定价值。

对比增强能谱乳腺摄影用于非肿块样强化乳腺癌的诊断价值

目的探讨对比增强能谱乳腺摄影(CESM)影像特征对非肿块样强化(NME)乳腺导管癌不同组织学类型与腋窝淋巴结(ALN)转移的诊断价值。方法回顾性纳入完成CESM检查的NME乳腺癌女性病人63例,经病理确诊为导管原位癌(DCIS)或浸润性导管癌(IDC)。根据组织学类型分为DCIS组(32例)和IDC组(31例);根据有无微浸润将32例DCIS病人分为单纯DCIS组(18例)和DCIS伴微浸润(DCIS-MI)组(14例);根据ALN转移情况分为转移阳性组(19例)和转移阴性组(44例)。采用t检验、Mann-Whitney U检验、χ^(2)检验比较各分组的CESM影像特征的差异。将筛选出来有统计学意义的CESM影像特征行多因素Logistic回归分析,确定IDC、DCIS-MI及ALN转移的独立预测因素。绘制受试者操作特征(ROC)曲线,计算并比较ROC曲线下面积(AUC)、敏感度、特异度。结果DCIS与IDC间在内部强化方式、头足位减影影像相对强化比值(%RSCC)及时间信号强化曲线(TIC)类型存在统计学差异(均P<0.05);单纯DCIS与DCIS-MI的最大截面面积、%RSCC及TIC类型间差异有统计学意义(均P<0.05);ALN转移阳性组与阴性组间内部强化方式、形态分布特征、%RSCC及TIC类型差异有统计学意义(均P≤0.05)。%RSCC>3.42%和集簇/簇环强化是IDC的独立预测因素(均P<0.05),其AUC值分别为0.732、0.704;%RSCC>2.47%和最大截面面积大于4.37 cm2是DCIS-MI的独立预测因素(均P<0.05),其AUC值分别为0.802、0.764;%RSCC>2.71%和集簇/簇环强化是ALN转移的独立预测因素(均P<0.05),其AUC值分别为0.717、0.786。结论在CESM上测得的%RSCC、最大截面面积、TIC、形态分布特征及内部强化特征对于非肿块型乳腺导管癌组织学类型及ALN转移具有一定的诊断价值,可为临床医生决策及病人预后提供参考依据。

高频超声引导下粗针穿刺活检及肿瘤标志物检测在乳腺肿块良恶性鉴别诊断中的价值研究

目的 探讨高频超声引导下粗针穿刺活检及肿瘤标志物检测在乳腺肿块良恶性鉴别诊断中的价值。方法 选择140例乳腺肿块患者作为研究对象,经手术术后病理检查确诊乳腺癌72例、乳腺良性病变68例,所有患者术前均接受高频超声检查、高频超声引导下粗针穿刺活检及肿瘤标志物[癌胚抗原(CEA)、糖类抗原199(CA199)、糖类抗原125(CA125)]检测。比较不同检查方法 (高频超声、高频超声引导下粗针穿刺活检)对乳腺癌的诊断效能,乳腺癌和乳腺良性病变患者的肿瘤标志物(CEA、CA199、CA125)水平,不同高频超声引导下粗针穿刺活检结果患者的超声表现特征及肿瘤标志物水平。结果 高频超声引导下粗针穿刺活检将4例乳腺癌误诊为乳腺良性病变,将1例乳腺纤维瘤和2例纤维样囊肿误诊为乳腺癌。高频超声引导下粗针穿刺活检诊断乳腺癌的灵敏度、特异度及准确率分别为94.44%、95.59%、95.00%,均明显高于高频超声的68.06%、79.41%、73.57%(P<0.05)。乳腺癌患者CEA、CA199、CA125水平分别为(4.55±1.22)ng/ml、(31.86±9.56)U/ml、(26.12±6.18)U/ml,均明显高于乳腺良性病变患者的(1.68±0.52)ng/ml、(12.10±4.33)U/ml、(11.54±4.83)U/ml(P<0.05)。高频超声引导下粗针穿刺活检良性与恶性患者的肿块最大直径、形态、边界、内部回声、淋巴结肿大、钙化、血流信号等超声特征比较差异有统计学意义(P<0.05)。高频超声引导下粗针穿刺活检良性患者的CEA、CA199、CA125水平分别为(1.74±0.55)ng/ml、(12.43±4.40)U/ml、(11.62±4.88)U/ml,均低于恶性患者的(4.53±1.26)ng/ml、(31.88±9.68)U/ml、(26.25±6.20)U/ml(P<0.05)。结论 高频超声引导下粗针穿刺活检能够较好的鉴别诊断乳腺癌与乳腺良性病变,其诊断效能明显优于单纯高频超声,穿刺活检结果与肿瘤标志物、超声表现特征存在明显关系,能够为乳腺肿块治疗提供指导依据。

Simultaneous Paragonimus infection involving the breast and lung:A case report

BACKGROUND Paragonimiasis is a food-borne parasitic infection caused by lung flukes of the genus Paragonimus. Although the most common site of infection is the pleuropulmonary area, the parasite can also reach other parts of the body on its journey from the intestines to the lungs, ending up in locations such as the brain,abdomen, skin, and subcutaneous tissues. Ectopic paragonimiasis is difficult to diagnose due to the rarity of this disease.CASE SUMMARY Here, we report a rare case of simultaneous breast and pulmonary paragonimiasis in a woman presenting painless breast mass and lung nodule with a history of eating raw trout. To confirm the diagnosis, serologic testing and tissue confirmation of the breast mass were performed. The patient was treated with surgical resection of the mass and praziquantel medication.CONCLUSION Ectopic paragonimiasis is difficult to diagnose due to the rarity of this disease.Thus, thorough history-taking and clinical suspicion of parasitic infection are important.

MRI预测乳腺肿块良恶性的研究进展

随着医疗水平的进步,乳腺疾病的检出率越来越高,特别是乳腺肿块。早期对肿块良恶性进行精准预测,不仅可以提高患者生存率,而且可以降低治疗成本。随着动态对比增强磁共振成像、扩散加权成像和磁共振光谱等各种技术的出现,早期预测乳腺肿块良恶性成为可能。该文主要对磁共振成像预测乳腺肿块良恶性的研究进展进行综述。

基于超声BI-RADS分类对不同病理类型乳房肿块诊断的结果分析

目的分析基于超声乳腺影像报告和数据系统(Breast Imaging Reporting and Data System,BI-RADS)分类对不同病理类型乳腺肿块诊断的结果。方法方便选取2022年3月—2023年8月在枣庄市台儿庄区人民医院和济南市平阴县中医医院进行检查的74例女性患者为研究对象。以病理检查为金标准,经病理证实乳腺肿块共80个,采用kappa检验分析超声检查与病理结果的一致性。结果病理结果显示,良性肿块61个,恶性肿块19个(kappa值为0.710,P<0.01)。超声BI-RADS分类诊断的灵敏度、特异度、准确度分别为96.72%、84.21%、93.75%。超声BI-RADS对良性肿瘤的诊断符合率均≥85.71%,而对恶性肿瘤诊断中,符合率最低时为75.00%。结论虽然不同病理类型乳腺肿块的超声诊断符合率较高,但在实际操作过程中仍然存在漏诊误诊等情况,故需要临床医师了解超声检查的不足,对超声诊断质量做好把控。

乳腺肿块良恶性诊断中乳腺超声的价值分析

目的:探讨乳腺肿块良恶性诊断中乳腺超声的价值。方法:选取2023年1月—10月宝鸡市中心医院治疗的95例乳腺肿块患者作为研究对象,均实施乳腺超声及乳腺X射线摄影,以活检或手术病理结果作为金标准,对比不同检查方式诊断结果,比较诊断效能,比较不同检查方式对不同类型乳腺疾病检出情况,分析超声良恶性肿块血流指标。结果:病理结果显示,95例乳腺肿块患者中阳性21例,阴性74例;乳腺超声诊断乳腺癌灵敏度90.48%、特异度93.24%、准确率92.63%、阳性预测值79.17%均高于乳腺X射线摄影的57.14%、81.08%、76.84%、48.15%(P<0.05);乳腺超声检查乳腺增生、乳腺囊肿、乳腺纤维腺瘤、乳腺导管内乳头状瘤、乳腺浸润性导管癌、乳腺导管原位癌、髓样癌等检出率与乳腺X射线摄影比较差异不显著(P>0.05)。恶性肿块的最大血流速度、最小血流速度、阻力指数均高于良性肿块,血流信号分级多在2~3级(P<0.05)。结论:乳腺肿块良恶性诊断中乳腺超声的价值显著,能提高对良恶性肿瘤的鉴别诊断效果,提升诊断的灵敏度、特异度、准确率。

基于DCE-MRI影像组学模型对乳腺NME病变诊断价值的研究

目的 探讨结合机器学习早期动态增强磁共振成像(DCE-MRI)的影像组学模型在鉴别良恶性乳腺非肿块强化(NME)病变中的价值。方法 选取行乳腺DCE-MRI检查并获得病理结果的NME病变患者242例,分为训练集163例、测试集55例,外部验证集24例。基于早期DCE-MRI序列的特征选择,采用支持向量机(SVM)建立组学预测模型;由2位放射科医师独立评估MRI特征,建立传统诊断模型,预测病灶的良恶性;运用测试集和外部验证集进行测试和外部验证。采用受试者工作特征(ROC)曲线评价组学模型与放射医师的诊断效能。结果 影像组学模型鉴别乳腺NME病变良恶性达到了与高年资放射医师[曲线下面积(AUC)=0.82, 95%CI 0.66, 0.89]相当的诊断水平[(AUC=0.82, 95%CI 0.67, 0.90);P=0.30],均优于低年资放射医师的评估结果(Z=2.63,P=0.01;Z=2.41,P=0.02),同时利用外部验证集进一步验证该模型的预测效能。结论 基于早期DCE-MRI组学模型可以有效地鉴别NME病变的良恶性,与高年资放射医师诊断水平相当,并优于低年资医师诊断水平,可以辅助低年资医师做出更佳诊断。

浸润性乳腺癌患者非肿块强化和肿块强化的动态增强MRI特征分析

目的探讨分析动态增强MRI特征为非肿块强化和肿块强化的浸润性乳腺癌患者影像特征。方法选取浸润性乳腺癌患者55例,行乳腺动态增强MRI检查显示为非肿块强化和肿块强化患者,分别作为观察组和对照组均经手术或病理活检证实,评估两组表观扩散系数(ADC)值、早期强化率、肿块边界清晰、增强平台类型,病灶大小、雌激素受体(ER)、人表皮生长因子受体-2(HER-2)阳性、T分期、N分期结果。结果观察组表观扩散系数(ADC)值为(1.29±0.20)×10^(-3) mm^(2)/s,高于对照组(0.89±0.18)×10^(-3) mm^(2)/s,差异具有统计学意义(P<0.05),观察组肿块边界不清晰发生率为80.52%,高于对照组20.59%,差异具有统计学意义(P<0.05),观察组病灶体积为(25.50±4.21)cm^(3),高于对照组(16.65±3.92)cm^(3),差异具有统计学意义(P<0.05),观察组HER-2阳性率为76.19%,高于对照组26.47%,两组T分期、N分期比较,差异具有统计学意义(P<0.05)。结论对非肿块强化和肿块强化浸润性乳腺癌患者,动态增强MRI通过多参数、多方位成像进行鉴别,非肿块强化病灶ADC值、边界不清晰发生率及ER、HER-2阳性率表现为增高。

自动乳腺全容积成像预测乳腺结节良恶性列线图模型的建立与验证

目的:建立并验证自动乳腺全容积成像(automated breast volume scanner,ABVS)预判乳腺结节良恶性的列线图模型,评价其临床应用价值。方法:回顾性收集2021年11月01日至2022年12月31日经ABVS检查诊断为BI-RADS 3~5类乳腺结节的患者临床资料,共有398例患者,532个乳腺结节纳入本研究。随机按照7∶3的比例分为训练集(n=372)和验证集(n=160)。通过LASSO回归筛选出潜在的变量,纳入Logistic回归筛选出独立危险因素,基于独立危险因素构建联合预测模型,并绘制可视化列线图。使用受试者工作特征(receiver operating characteristic,ROC)曲线评价列线图诊断效能。使用DeLong检验来比较独立变量与列线图的诊断性能。使用Hosmer-Lemeshow检验评价列线图的校准度,采用决策曲线分析(decision curve analysis,DCA)验证列线图的临床有效性。结果:通过LASSO回归、单因素和多因素Logistic回归筛选出ABVS诊断乳腺癌的独立危险因素为年龄、微钙化、边界、微分叶、皱缩征。基于上述变量绘制的列线图预测模型在训练集中曲线下面积、灵敏度及特异度分别为0.967、87.5%和96.0%;在验证集中分别为0.991、93.5%和96.1%。DeLong检验显示列线图在所有模型中具有最佳的诊断效能(P<0.05)。Hosmer-Lemeshow检验显示该模型具有良好的校准度,决策曲线分析(DCA)显示其具有良好的临床有效性。结论:通过建立ABVS预判乳腺结节良恶性的列线图模型,对于鉴别乳腺良恶性结节具有较高的临床应用价值,可作为临床医师预判乳腺癌的辅助工具。