Measures of angular spread and entropy for the detection of architectural distortion in prior mammograms

Purpose

Architectural distortion is an important sign of early breast cancer. We present methods for computer-aided detection of architectural distortion in mammograms acquired prior to the diagnosis of breast cancer in the interval between scheduled screening sessions.

Methods

Potential sites of architectural distortion were detected using node maps obtained through the application of a bank of Gabor filters and linear phase portrait modeling. A total of 4,224 regions of interest (ROIs) were automatically obtained from 106 prior mammograms of 56 interval-cancer cases, including 301 true-positive ROIs, and from 52 mammograms of 13 normal cases. Each ROI was represented by three types of entropy measures of angular histograms composed with the Gabor magnitude response, angle, coherence, orientation strength, and the angular spread of power in the Fourier spectrum, including Shannon’s entropy, Tsallis entropy for nonextensive systems, and Rényi entropy for extensive systems.

Results

Using the entropy measures with stepwise logistic regression and the leave-one-patient-out method for feature selection and cross-validation, an artificial neural network resulted in an area under the receiver operating characteristic curve of 0.75. Free-response receiver operating characteristics indicated a sensitivity of 0.80 at 5.2 false positives (FPs) per patient.

Conclusion

The proposed methods can detect architectural distortion in prior mammograms taken 15 months (on the average) before clinical diagnosis of breast cancer, with a high sensitivity and a moderate number of FPs per patient. The results are promising and may be improved with additional features to characterize subtle abnormalities and larger databases including prior mammograms.     

Measures of divergence of oriented patterns for the detection of architectural distortion in prior mammograms

Purpose We propose a method for the detection of architectural distortion in prior mammograms of interval-cancer cases based on the expected orientation of breast tissue patterns in mammograms. Methods The expected orientation of the breast tissue at each pixel was derived by using automatically detected landmarks including the breast boundary, the nipple, and the pectoral muscle (in mediolateral-oblique views). We hypothesize that the presence of architectural distortion changes the normal expected orientation of breast tissue patterns in a mammographic image. The angular deviation of the oriented structures in a given mammogram as compared to the expected orientation was analyzed to detect potential sites of architectural distortion using a measure of divergence of oriented patterns. Each potential site of architectural distortion was then characterized using measures of spicularity and angular dispersion specifically designed to represent spiculating patterns. The novel features for the characterization of spiculating patterns include an index of divergence of spicules computed from the intensity image and Gabor magnitude response using the Gabor angle response; radially weighted difference and angle-weighted difference (AWD) measures of the intensity, Gabor magnitude, and Gabor angle response; and AWD in the entropy of spicules computed from the intensity, Gabor magnitude, and Gabor angle response. Results Using the newly proposed features with a database of 106 prior mammograms of 56 interval-cancer cases and 52 mammograms of 13 normal cases, through feature selection and pattern classification with an artificial neural network, an area under the receiver operating characteristic curve of 0.75 was obtained. Free-response receiver operating characteristic analysis indicated a sensitivity of 0.80 at 5.3 false positives (FPs) per patient. Combining the features proposed in the present paper with others described in our previous works led to significant improvement with a sensitivity of 0.80 at 3.7 FPs per patient. Conclusion The proposed methods can detect architectural distortion in prior mammograms taken 15 months (on the average) before clinical diagnosis of breast cancer, but the FP rate needs to be reduced.     

Reduction of false positives in the detection of architectural distortion in mammograms by using a geometrically constrained phase portrait model

Objective One of the commonly missed signs of breast cancer is architectural distortion. We have developed techniques for the detection of architectural distortion in mammograms, based on the analysis of oriented texture through the application of Gabor filters and a linear phase portrait model. In this paper, we propose constraining the shape of the general phase portrait model as a means to reduce the false-positive rate in the detection of architectural distortion. Material and methods The methods were tested with one set of 19 cases of architectural distortion and 41 normal mammograms, and with another set of 37 cases of architectural distortion. Results Sensitivity rates of 84% with 4.5 false positives per image and 81% with 10 false positives per image were obtained for the two sets of images. Conclusion The adoption of a constrained phase portrait model with a symmetric matrix and the incorporation of its condition number in the analysis resulted in a reduction in the false-positive rate in the detection of architectural distortion. The proposed techniques, dedicated for the detection and localization of architectural distortion, should lead to efficient detection of early signs of breast cancer.     

基于分形特征序列的乳腺X线图像分类方法

目的表征乳腺图像中肿块部分纹理特征,通过纹理分析实现乳腺图像中肿块部分与正常腺体部分的分类。方法应用分形特征值表征乳腺图像纹理特征,利用多级分形特征提取法将乳腺图像分解成一系列细节图像,提取出多个分形特征值;利用分类精度、ROC曲线及曲线下面积(AUC)进行特征选择构建分形特征序列,最后应用支持向量机(SVM)方法进行分类。结果对60幅图像的可疑病变区域进行分形特征序列提取分析,SVM交叉验证分类精度达84.50%。结论基于分形维数的乳腺图像分类方法不仅能对肿块与正常腺体进行图像分类,还可有效表征乳腺图像的纹理信息,有助于提高乳腺肿块诊断的准确率。     

Needs assessment for next generation computer-aided mammography reference image databases and evaluation studies

Introduction  

Breast cancer is globally a major threat for women’s health. Screening and adequate follow-up can significantly reduce the mortality from breast cancer. Human second reading of screening mammograms can increase breast cancer detection rates, whereas this has not been proven for current computer-aided detection systems as “second reader”. Critical factors include the detection accuracy of the systems and the screening experience and training of the radiologist with the system. When assessing the performance of systems and system components, the choice of evaluation methods is particularly critical. Core assets herein are reference image databases and statistical methods.     

Causal attributions to epidemiological risk factors and their associations to later psychological adjustment among Japanese breast cancer patients

Goals  The objective of this study was to evaluate the causal attributions for breast cancer and their association with the subsequent psychological adjustment. Materials and methods  Sixty-three Japanese patients newly diagnosed with breast cancer were asked by an interviewer about risk factors and explanations for the etiology of breast cancer using a four-point scale. The General Health Questionnaire (GHQ-28) and the Center for Epidemiologic Studies Depression Scale (CES-D) were administered about 2 months later to measure psychological adjustment to the illness. The relationship between each causal attribution and measure of psychological adjustment was assessed with the Spearman rank correlation after adjusting for potential confounders (age, marital status, cancer stage at diagnosis, time between the day of diagnosis and follow-up, and physical symptoms at the follow-up). Main results  Attributions to several explanations, including “stress” and “personal characteristics,” were positively associated with the GHQ-28 measures. Of the attributions to risk factors, “body size” was significantly inversely (r = −0.29) associated and “never having children” was insignificantly and highly inversely (r = −0.77) associated with the GHQ-28 measures, indicating better adjustment to the illness. On the other hand, the attribution to “tobacco” was significantly and positively associated with the GHQ-28 measures (r = 0.34), indicating that the attributions antagonized adjustment to the illness. Conclusion  The current study indicates that forming causal attributions influences the adjustment to the illness in Japanese breast cancer patients and attributions to certain risk factors for breast cancer may contribute to better adjustment.     

基于改进局部三元模式的乳腺癌预测模型

目的 探讨基于改进局部三元模式（LTP）算法提取的乳腺新型纹理特征及其与常规特征融合预测乳腺癌的价值。方法 对钼靶图像进行乳腺分割,采用基于改进LTP算法提取双侧乳腺的新型纹理特征和常规特征;合并左右侧乳腺纹理特征;采用主成分分析法对提取的高维纹理特征降维;以K最近邻（KNN）和LADTree算法分别对纹理特征及融合纹理特征进行分类。结果 基于改进LTP算法提取的新型纹理特征预测乳腺癌的ROC曲线下面积（AUC）为0.732 4±0.042 8,敏感度为72.04%（67/93）,特异度为74.51%（76/102）,准确率为73.33%（143/195）;融合常规特征后AUC为0.865 5±0.014 8,敏感度为84.95%（79/93）,特异度为88.23%（90/102）,准确率为86.67%（169/195）。结论 基于LTP算法提取的新型纹理特征预测乳腺癌的精度较高,与常规特征融合后可进一步提高预测效能。     

A pilot study of architectural distortion detection in mammograms based on characteristics of line shadows

Objective  We present herein a novel algorithm for architectural distortion detection that utilizes the point convergence index with the likelihood of lines (e.g., spiculations) relating to architectural distortion. Materials and methods  Validation was performed using 25 computed radiography (CR) mammograms, each of which has an architectural distortion with radiating spiculations. The proposed method comprises five steps. First, the lines were extracted on mammograms, such as spiculations of architectural distortion as well as lines in the mammary gland. Second, the likelihood of spiculation for each extracted line was calculated. In the third step, point convergence index weighted by this likelihood was evaluated at each pixel to enhance distortion only. Fourth, local maxima of the index were extracted as candidates for the distortion, then classified based on nine features in the last step. Results  Point convergence index without the proposed likelihood generated 84.48/image false-positives (FPs) on average. Conversely, the proposed index succeeded in decreasing this number to 12.48/image on average when sensitivity was 100%. After the classification step, number of FPs was reduced to 0.80/image with 80.0% sensitivity. Conclusion  Combination of the likelihood of lines with point convergence index is effective in extracting architectural distortion with radiating spiculations.     

A qualitative assessment of upper quarter dysfunction reported by physical therapists treated for breast cancer or treating breast cancer sequelae

Purpose  

As survival rates for breast cancer improve, long-term effects of treatment are receiving increasing attention, including upper quarter impairments and functional limitations. The purpose of this study was to assess, through qualitative means, the long-term effects of breast cancer treatment on upper quarter function as reported by those with expertise in upper quarter dysfunction. Participants were physical therapists who were either breast cancer survivors (“survivors”) or those who specialized in treatment of individuals post-breast cancer (“treaters”).     

平扫CT特征联合纹理分析鉴别经治乳腺癌患者肺部单发转移癌与原发腺癌北大核心CSCD

目的评估基于平扫CT特征及纹理分析联合模型鉴别经治乳腺癌患者单发肺内病灶为乳腺癌孤立性肺转移(SPMBC)或原发性肺腺癌(PLA)的价值。方法回顾性分析111例经治乳腺癌伴术后病理证实的SPMBC(SPMBC组,n=54)或PLA(PLA组,n=57)患者,对比其肺内病灶CT表现。采用MaZda软件提取并筛选CT所示病灶最佳纹理参数,以集成算法为分类器,针对组间差异有统计学意义的CT表现、最佳纹理参数或联合二者分别构建CT特征模型、纹理特征模型及联合模型,以鉴别SPMBC与PLA;以5折交叉验证法及受试者工作特征(ROC)曲线评估各模型的诊断效能。结果CT特征模型、纹理特征模型及联合模型鉴别经治乳腺癌患者伴SPMBC与PLA的平均曲线下面积分别为0.64±0.08、0.82±0.07及0.85±0.05。CT特征模型的平均分类准确率(0.60±0.09)低于纹理特征模型及联合模型(0.78±0.09、0.82±0.08,t=-3.14、-4.06,P均<0.05),纹理特征模型平均分类准确率与联合模型差异无统计学意义(t=-0.66,P>0.05)。结论平扫CT特征联合纹理分析有助于鉴别诊断经治乳腺癌患者SPMBC与PLA。     

Neuromuscular electrical stimulation for a patient with metastatic lung cancer—a case report

A 47-year-old female patient suffering from advanced lung cancer with metastatic bone and brain disease participated in a passive exercise program, consisting of neuromuscular electrical stimulation (NMES) five times a week, carried out for 4 weeks. After the training period, the results of the 6-min walk (420 m before and 603 m after the training period) have improved by 44%, which demonstrates the increase of physical performance (mobility, endurance capacity). The results of the “Timed up and go” indicate an improvement of mobility and functional health of skeletal muscles. Furthermore, the quality of life (QOL)-scales (assessed by using the SF-36 health survey) “Physical functioning”, “Role—physical”, “Mental health”, “Role—emotional”, “Vitality”, “Bodily pain”, and “General health” showed improvements after the intervention period. Feasibility, safety, and beneficial effects of the NMES program were proven for the patient in this case study. These findings indicate that NMES, initiated and executed with appropriate care, may serve as a useful supportive means of palliative treatment in some patients with advanced cancer and metastatic disease, especially in cases of metastatic involvement of the brain and of the skeletal system with the risk of seizures and pathological fractures where volitional training is not allowed.     

Understanding the cultural health belief model influencing health behaviors and health-related quality of life between Latina and Asian-American breast cancer survivors

Purpose  The purpose of this study was to (1) describe health behaviors and health-related quality of life (HRQOL) of Latina and Asian-American breast cancer survivors (BCS), (2) estimate possible culturally driven predictors of health behaviors and HRQOL, and (3) compare pathways for predicting health behaviors and HRQOL between the two groups. Materials and methods  Secondary data were used to investigate health behaviors and HRQOL among 183 Latina and 206 Asian Americans diagnosed with breast cancer. The study methodology was guided by the health belief model and the contextual model of HRQOL. Structural equation modeling was used to test cultural predictors on health behaviors of BCS. Results  Asian Americans reported higher emotional and physical well-being scores than Latina-Americans. Confirmatory factor analyses demonstrated the adequacy of the two-factor model (“powerful others” and “sociocultural factors”) in the cultural health belief construct for Latina and Asian-American BCS. In the structural model, Latinas and Asian Americans showed different pathways in the predicted relationships among the variables. For Latina-Americans, doctor–patient relationship was positively related to exercise, and in turn, influenced physical and emotional well-being. For Asian Americans, treatment decisions and the “sociocultural factor” were significantly related to stress management. Conclusion  This study adds to the existing literature in that no study has focused on cultural health beliefs and health behaviors between Latina and Asian-American BCS. Evidence that Latinas and Asian Americans varied in the patterns of cultural factors influencing health behaviors and HRQOL might lead to the development of culturally sensitive breast cancer interventions for promoting positive health behavior and ultimately increasing HRQOL.     

钼靶联合hMAM、BRCA-1表达水平检测对乳腺癌的诊断价值

目的探讨钼靶联合乳腺癌人乳腺珠蛋白(hMAM)基因、乳腺癌易感基因-1(BRCA-1)表达水平对乳腺癌的诊断价值。方法选取2016年1月至2019年8月在该院就诊的乳腺癌患者150例纳入肿瘤组,另选取同期良性乳腺肿瘤患者100例纳入良性组。通过受试者工作特征(ROC)曲线分析单独及联合应用钼靶及hMAM、BRCA-1检测诊断乳腺癌的灵敏度、特异度、ROC曲线下面积(AUC)等。结果肿瘤组钼靶检测阳性率(68.7%)明显高于良性组(27.0%),差异有统计学意义(χ2=41.730,P<0.05)。肿瘤组hMAM表达水平明显高于良性组,BRCA-1表达水平明显低于良性组,差异有统计学意义(t=49.723、30.632,P<0.05)。钼靶及hMAM、BRCA-1联合检测诊断乳腺癌的灵敏度(78.7%)明显高于单独检测的灵敏度(68.7%、57.3%、60.7%),差异有统计学意义(P<0.05),还保持了较高的特异度(79.0%),联合检测的AUC(0.864)明显高于单独检测(0.792、0.753、0.642),差异有统计学意义(P<0.05)。结论钼靶检测及hMAM、BRCA-1表达水平检测可提高乳腺癌诊断的灵敏度,并保持较高的特异度,对乳腺癌诊断有重要的临床价值。     

Breast tissue characterization using FARMA modeling of ultrasonic RF echo

A number of empirical and analytical studies demonstrated that the ultrasound RF echo reflected from tissue exhibits 1/f characteristics. In this paper, we propose to model 1/f characteristics of the ultrasonic RF echo by a novel parsimonious model, namely the fractional differencing auto regressive moving average (FARMA) process, and evaluated diagnostic value of model parameters for breast cancer malignancy differentiation. FARMA model captures the fractal and long term correlated nature of the backscattered speckle texture and facilitates robust efficient estimation of fractal parameters. In our study, in addition to the computer generated FARMA model parameters, we included patient age and radiologist's prebiopsy level of suspicion (LOS) as potential indicators of malignant and benign masses. We evaluated the performance of the proposed set of features using various classifiers and training methods using 120 in vivo breast images. Our study shows that the area under the receiver operating characteristics (ROC) curve of FARMA model parameters alone is superior to the area under the ROC curve of the radiologist's prebiopsy LOS. The area under the ROC curve of the three sets of features yields a value of 0.87, with a confidence interval of [0.85, 0.89], at a significance level of 0.05. Our results suggest that the proposed method of ultrasound RF echo model leads to parameters that can differentiate breast tumors with a relatively high precision. This set of RF echo features can be incorporated into a comprehensive computer-aided diagnostic system to aid physicians in breast cancer diagnosis.     

Three-dimensional echocardiographic virtual endoscopy for the diagnosis of congenital heart disease in children

Virtual endoscopy (VE) is a new post-processing method that uses volumetric data sets to simulate the tracks of a “conventional” flexible endoscope. However, almost all studies of this method have involved virtual visualizations of the cardiovascular structures applied to computed tomography (CT) and magnetic resonance (MR) datasets. This paper introduces a novel visualization method called the “three-dimensional echocardiographic intracardiac endoscopic simulation system (3DE IESS)”, which uses 3D echocardiographic images in a virtual reality (VR) environment to diagnose congenital heart disease. The aim of this study was to analyze the feasibility of VE in the evaluation of congenital heart disease in children and its accuracy compared with 2DE. Three experienced pediatric cardiologists blinded to the patients’ diagnoses separately reviewed 40 two-dimensional echocardiographic (2DE) datasets and 40 corresponding VE datasets and judged whether abnormal intracardiac anatomy was present in terms of a five-point scale (1 = definitely absent; 2 = probably absent; 3 = cannot be determined; 4 = probably present; and 5 = definitely present). Compared with clinical diagnosis, the diagnostic accuracy of VE was 98.7% for ASD, 92.4% for VSD, 92.6% for TOF, and 94% for DORV, respectively. Diagnostic accuracy of VE was significantly higher than that of 2DE for TOF and DORV except for ASD and VSD. The receiver operating characteristic (ROC) curve for VE was closer to the optimal performance point than was the ROC curve for 2DE. The area under the ROC curve was 0.96 for VE and 0.93 for 2DE. Kappa values (range, 0.73–0.79) for VE and 2DE indicated substantial agreement. 3D echocardiographic VE can enhance our understanding of intracardiac structures and facilitate the evaluation of congenital heart disease.     

能谱CT成像碘(水)图纹理分析预测结直肠癌微卫星不稳定状态

目的 探讨基于能谱CT成像碘（水）图像的纹理分析在术前预测结直肠癌微卫星不稳定（MSI）状态方面的价值。方法 回顾性分析23例MSI结直肠癌（MSI组）及46例微卫星稳定（MSS）结直肠癌（MSS组）患者的资料。所有患者均经术后病理检查证实，且术前均接受腹部能谱CT成像。采用Viewer分析软件获取动脉期及静脉期碘（水）图像，并将其导入Omni-Kinetics软件进行ROI勾画及特征提取。提取的纹理参数包括最小值、最大值、平均值、中位值、标准差、偏度、峰度、均匀性、能量值、熵。比较2组间各纹理参数的差异。并采用Logistic回归将纹理参数进行联合，通过ROC曲线分析不同纹理参数预测及多种参数联合预测的效能。结果 MSI组动脉期及静脉期最小值、最大值、平均值、中位值、均匀性均明显低于MSS组（P均<0.05），2组间标准差、偏度、峰度、能量值差异均无统计学意义（P均>0.05）；MSI组静脉期熵明显高于MSS组（t=1.81，P=0.04），2组间动脉期熵差异无统计学意义（t=0.22，P=0.80）。ROC曲线分析显示，以动脉期及静脉期最小值、最大值、平均值、中位值、均匀性和静脉期熵单一参数在术前预测结直肠癌MSI状态的AUC为0.64~0.82。多参数联合的Logistic回归模型为-2.598-0.124×动脉期最小值-0.039×动脉期最小值-0.774×动脉期中位值+1×动脉期平均值-1.892×动脉期均匀性+0.14×静脉期最小值+0.2×静脉期最大值+0.343×静脉期中位值-0.61×静脉期平均值+13.711×静脉期均匀性-2.598×静脉期熵，联合预测的AUC为0.83。结论 基于能谱CT成像碘（水）图像纹理分析，可在术前无创预测结直肠癌MSI状态，且将多种纹理参数联合后预测效能更优。     

Classification of mammographic breast density and its correlation with BI-RADS in elder women using machine learning approach

IntroductionMammographic breast density (MBD) is a known risk factor for breast cancer and older women have higher incidence rates of breast cancer occurrence. The Breast Imaging Reporting and Data System (BI-RADS) is a commonly used MBD classification tool for mammogram reporting. However, they have limitations since there are reading inconsistencies between different radiologists with the visual assessment of breast density.MethodsDigitised film-screen mammographic images were extracted from the Digital Database for Screening Mammography (DDSM). A machine learning project was developed using commercially available software with several predictive models applied to classify different amount of MBD on mammograms into different density groups. The effectiveness of different predictive models used in classifying the mammograms were tested by receiver operator characteristics (ROC) curve with comparison to the gold standard of BI-RADS classification.ResultsThree predictive models, Decision Tree (Tree), Support Vector Model (SVM) and k-Nearest Neighbour (kNN) showed high AUC values of 0.801, 0.805 and 0.810 respectively. High AUC values for the three predictive models indicates that the accuracy of the model is approaching that of the BI-RADS method.DiscussionOur machine learning project showed to have capabilities to be potentially used in the clinical settings to help categorise mammograms into extremely dense breasts (BI-RADS Group A) from entirely fatty breasts (BI-RADS Group D).ConclusionFindings from the present study suggest that the machine learning method is potentially useful to quantify the amount of MBD in mammograms.     

CAD(计算机辅助探察)对提高乳腺癌诊断敏感性的意义

目的探讨CAD技术对提高乳腺癌诊断敏感性的意义.方法回顾性分析49位患者的常规乳腺钼靶照片(共162张),分别由低年、中年临床医师、CAD独立阅片及低年医师 CAD共同读片,将结果与最终诊断结果相比较.结果恶性钙化19例:低年医师漏诊7例;中年医师漏诊1例;CAD漏诊1例;低年医师 CAD漏诊2例.恶性肿块14例:低年医师漏诊6例;中年医师漏诊2例;CAD漏诊6例;低年医师 CAD漏诊3例.结论 CAD可以提高乳腺癌诊断的敏感性.     

Financial and family burden associated with cancer treatment in Ontario, Canada

Goals of work To determine the financial and family resources burden associated with the treatment of cancer. A questionnaire was developed to determine the direct monthly “out-of-pocket costs” (OOPC), the indirect costs, and the associated perceived family burden.Materials and methods A self-administered questionnaire using a quota sample from five cancer clinics in Ontario, Canada was given to 282 cancer patients (74 breast, 70 colorectal, 68 lung, and 70 prostate). Monthly OOPC were obtained for: drugs, home care, homemaking, complementary and alternative medicines, vitamins and supplements, family care, travel, parking, accommodations, devices, and others. The questionnaire asked if OOPC for treatment were a burden, and if others took time from work to provide caregiving.Main results The mean monthly OOPC was $213, with an additional $372 related to imputed travel costs. For those patients who responded that the burden was “significant” (16.5%), their OOPC was $452. In the case of patients responding that their burden was “unmanageable” (3.9%), their OOPC was $544. The survey showed that 35.6% of patients required others to take time from work and this was higher in the under-65 category. The mean number of days lost from work in the previous 30 days for these caregivers was 7 days.Conclusions These results suggest the financial burden is problematic for 20% of this sample. The caregivers’ lost time from work influence this burden, and for 36% of this sample, it amounts to one third of their working days in any given month. Policies and programs to address these gaps are needed.