Computer-aided diagnosis in breast DCE-MRI--quantification of the heterogeneity of breast lesions

Purpose

In our study we aim at the quantification of the heterogeneity for differential diagnosis of breast lesions in MRI.

Materials and methods

We tested a software tool for quantification of heterogeneity. The software tool provides a three-dimensional analysis of the whole breast lesion. The lesions were divided in regions with similar perfusion characteristics. Voxels were merged to the same region, if the perfusion parameters (wash-in, wash-out, integral, peak enhancement and time to peak) correlated to 99%. We evaluated 68 lesions from 50 patients. 31 lesions proved to be benign (45.6%) and 37 malignant (54.4%). We included small lesions which could only be detected with MRI.

Results

The analysis of heterogeneity showed significant differences (p < 0.005; AUC 0.7). Malignant lesions were more heterogeneous than benign ones. Significant differences were also found for morphologic parameters such as shape (p < 0.001) and margin (p < 0.007). The analysis of the enhancement dynamics did not prove successful in lesion discrimination.

Conclusion

Our study indicates that the region analysis for quantification of heterogeneity may be a helpful additional method to differentiate benign lesions from malignant ones.     

Multimodality computerized diagnosis of breast lesions using mammography and sonography

RATIONALE AND OBJECTIVES: The purpose of this study is to investigate the use of computer-extracted features of lesions imaged by means of two modalities, mammography and breast ultrasound, in the computerized classification of breast lesions. MATERIAL AND METHODS: We performed computerized analysis on a database of 97 patients with a total of 100 lesions (40 malignant, 40 benign solid, and 20 cystic lesions). Mammograms and ultrasound images were available for these breast lesions. There was an average of three mammographic images and two ultrasound images per lesion. Based on seed points indicated by a radiologist, the computer automatically segmented lesions from the parenchymal background and automatically extracted a set of characteristic features for each lesion. For each feature, its value averaged over all images pertaining to a given lesion was input to a Bayesian neural network for classification. We also investigated different approaches to combine image-based features into this by-lesion analysis. In that analysis, mean, maximum, and minimum feature values were considered for all images representing a lesion. We considered performance by using a leave-one-lesion-out approach, based on image features from mammography alone (two to five features), ultrasound alone (three to four features), and a combination of features from both modalities (three to five features total). RESULTS: For the classification task of distinguishing cancer from other abnormalities in a lesion-based analysis by using a single modality, areas under the receiver operating characteristic curves (A(z) values) increased significantly when the computer selected the manner (mean, minimum, or maximum) in which image-based features were combined into lesion-based features. The highest performance was found for lesion-based analysis and automated feature selection from mean, maximum, and minimum values of features from both modalities (resulting in a total of four features being used). That A(z) value for the task of distinguishing cancer was 0.92, showing a statistically significant increase over that achieved with features from either mammography or ultrasound alone. CONCLUSION: Computerized classification of cancer significantly improved when lesion features from both modalities were combined. Classification performance depended on specific methods for combining features from multiple images per lesion. These results are encouraging and warrant further exploration of computerized methods for multimodality imaging.     

Patterns of metastatic breast carcinoma: influence of tumour histological grade

AIM: To assess if the pattern of metastatic spread of carcinoma of the breast varies according to tumour histological grade. MATERIALS AND METHODS: The clinical details, histological features of the primary tumour, and imaging findings at presentation of patients with metastatic breast cancer have been recorded prospectively since 1997. The pattern of metastatic spread, age at metastasis, metastasis-free interval (MFI), and length of survival with metastases were analysed by tumour grade. RESULTS: There was a significant association between histological high-grade tumours and high frequency of intra-pulmonary metastases (p=0.013); liver metastases (p=0.039); para-aortic lymphadenopathy (p=0.022) and metastatic presentation under 50 years of age (p=0.003). A significant correlation was also demonstrated between histological low-grade tumours and increased frequency of pleural disease (p=0.020); increased frequency of bone metastases (p=0.004); prolonged MFI (MFI>5 years; p<0.0001); and increased length of survival (p<0.0001). CONCLUSION: There is a correlation between patterns of metastatic spread and tumour histological grade. This partly explains the negative prognostic value of high tumour grade, as metastases from grade 3 tumours more commonly occur at sites associated with a worse prognosis. This finding may also prove useful in interpreting imaging in patients who have a history of breast cancer and undergo subsequent imaging because of new symptoms.     

Magnetic resonance imaging features of papillary breast lesions

PurposeThis study was aimed to assess the role of magnetic resonance imaging (MRI) in the evaluation of the papillary lesions of the breast and their morphological relationship with the mammary ducts. The potential diagnostic contributory role of ductal oriented protocols to conventional dynamic magnetic resonance examination was also explored.Materials and methodsRetrospective data were collected from 46 patients who had been diagnosed with papillary breast lesions and undergone magnetic resonance examination.The presence of dilated ducts and their morphological relation with the lesion were recorded. Lesions were classified as follows: papilloma, papillomatosis and malignant papillary lesion. Statistical difference between groups was studied for each morphological and dynamic lesion characteristic.ResultsDilated ducts and characteristics of intraductal material can be identified by magnetic resonance imaging. Certain MRI findings such as a mass with crescentic peripheral fluid or focal intraductal mass on T2 weighted images may suggest the presence of an intraductal/papillary lesion. In this respect, non-fatsat T2 weighted images appear particularly useful. There was a significant difference between papilloma and papillomatosis with regard to segmental and heterogeneous contrast enhancement (p < 0.05 for both comparisons). In addition, there was a significant difference between papillomas and carcinomas with regard to homogenous, heterogeneous and segmental contrast enhancement (p < 0.05 for all). On the other hand, papillomatosis and carcinoma did not differ significantly in terms of any of the morphological or dynamical MR criteria compared.ConclusionPapillary lesions can be detected by MRI. Despite some overlaps in MRI findings between carcinoma, papilloma and papillomatosis, MRI may help differentiate these lesions. Major benefit of retroareolar imaging appears to arise from its ability to demonstrate ductal relation and extension of contrast enhanced regions.     

早期乳腺癌保乳手术治疗的临床研究

目的探讨早期乳腺癌保乳手术治疗的适应证及方法。方法分析我院近5年来20例保乳手术治疗的乳腺癌患者临床资料。结果保乳术后患者乳房外形均较好,两侧乳房基本对称,患者3年生存率为95％,复发率为5％。结论早期乳腺癌保乳手术治疗安全、疗效确切,全身综合治疗是保乳手术治疗成功的关键。     

早期乳腺癌X线征象421例分析

目的提高对早期乳腺癌X线征象的认识。方法回顾性分析421例经病理证实的早期乳腺癌的X线特征。结果421例病例中,非浸润性癌87例(20.6%),早期浸润性癌97例(23.0%),早期浸润性特殊型癌52例(12.4%),浸润性非特殊型癌(肿块直径≤1 cm)185例(44.0%)。乳腺X线检查显示89%的非浸润性癌和70%的早期浸润性癌灶可见钙化,而肿块在2组的发现率仅分别为18.0%和43.0%。结论乳腺X线片中微小钙化灶对早期乳腺癌的诊断具有重要价值。如果患者有乳头溢液,建议进一步进行乳腺导管造影检查。     

DIMA enlargement mammography in microcalcifications: a prospective study with ROC analysis

The purpose of this study was to investigate whether the four-fold magnification mammography (direct magnification, DIMA) technique would perform better than conventional 1.5-fold magnification mammography in the differentiation of breast microcalcifications into benign and malignant. Fifty patients with non-palpable microcalcifications detected by mammography were examined immediately prior to surgical biopsy using both a conventional (1.5-fold) and the DIMA (fourfold) magnification mammography techniques. The microcalcifications were classified by five experienced radiologists using morphological criteria. A receiver operating characteristics curve (ROC) analysis of the sensitivity and specificity of both techniques in assessing malignancy was then carried out. The DIMA mammography technique was slightly but non-significantly superior to the conventional method in detecting malignancy (p > 0.05). Coarse granular and pleomorphic calcifications were detected more frequently with the DIMA technique. Coarse calcifications were significantly more frequently associated with histologically benign findings, whereas fine granular calcifications were significantly more likely to be malignant lesions. Assessment of malignancy associated with microcalcifications using morphological criteria is not significantly improved by mammography techniques with higher magnification.     

Imaging-histologic discordance at sonographically guided percutaneous biopsy of breast lesions

OBJECTIVE: To determine the frequency of imaging-histologic discordance at percutaneous breast biopsy and to evaluate differences in clinical and radiologic findings between pathologically upgraded lesions and non-upgraded lesions. MATERIALS AND METHODS: From February 2000 to June 2005, we reviewed 386 cases that had suspicious imaging findings but yielded benign histology at US-core needle biopsy and that underwent subsequent excisional biopsy. In 74 of 386 cases, the benign histology at core needle biopsy could not provide a satisfactory explanation for the radiologically suspicious lesions. The clinical, radiologic and histologic findings were reviewed for those 74 cases that were classified as the upgrade group and the non-upgrade group after excisional biopsy. RESULTS: The upgrade rate was 17.6% (95% confidence interval, 10.6-27.8%, 13 of 74 cases) revealed upgraded pathology at the subsequent excisional biopsy. Besides the size of masses, there were no statistically significant differences in imaging findings between the upgrade and non-upgrade groups at excisional biopsy. CONCLUSION: This upgrade rate of 17.6% suggests that excisional biopsy or re-biopsy is warranted in those cases presenting imaging-histologic discordance at US-guided core biopsy.     

Outcome of breast lesions detected at screening ultrasonography

Objective

To assess the final outcome of breast lesions detected during screening ultrasonography (US) and categorized by BI-RADS final assessment.

Materials and methods

During a 1-year period, 3817 consecutive asymptomatic women with negative findings at both clinical breast examinations and mammography underwent bilateral whole breast US and BI-RADS categories were provided for US-detected breast lesions. The reference standard was a combination of histology and US follow-up (≥12 months), and the final outcomes of 1192 US-detected lesions were analyzed.

Results

Of 904 category 2 lesions, 890 remained stable for 12–60 months. Biopsies of 14 lesions revealed no malignancies (NPV = 100%). Of 247 category 3 lesions, 232 remained stable for 12–60 months. Biopsies of 15 lesions revealed 2 malignancies, which were diagnosed within 6 months of the index examination and were node negative (NPV = 99.2%). Of 41 category 4 lesions, biopsies of 38 lesions revealed 5 malignancies (PPV = 12.2%), and 3 remained stable for 37–51 months. No US-detected lesion was classified as category 5.

Conclusion

The rates of malignancy for US-detected BI-RADS categories 2, 3, and 4 lesions were 0%, 0.8%, and 12.2%, respectively. The final assessment of US BI-RADS categorization showed it to be an appropriate predictor of malignancy for screening US-detected breast lesions.     

A head-to-head comparison between technetium-99m-tetrofosmin and technetium-99m-MIBI scintigraphy to evaluate suspicious breast lesions

The aim of this study was to assess the diagnostic value of technetium-99m-tetrofosmin and technetium-99m-MIBI in a head-to-head comparison. Both radiopharmaceuticals are routinely used for detecting breast cancer. In a prospective, open, diagnostic trial, the two radiopharmaceuticals were administered randomly on different days to the same 101 women suffering from 103 breast tumours. Planar images and single photon emission computer tomography (SPET) were performed. After histological examination of the tumours, sensitivity, specificity and positive and negative predictive value were compared. ^99mTc-tetrofosmin and ^99mTc-MIBI showed low sensitivity in planar images (44% vs 46%, respectively). SPET improved sensitivity (70% vs 69%, respectively). Specificity in planar images was 83% and 87%, and it was even lower using SPET (70% vs 78%, respectively). Positive predictive value in planar images was 76% vs 81%, and it was not changed by SPET. Negative predictive value was low in planar images (54% vs 57%, respectively), but it was improved by using SPET (65% vs 67%, respectively). In conclusion, ^99mTc-tetrofosmin and ^99mTc-MIBI scintigraphy show similar diagnostic value in assessing suspicious breast lesions. Received 19 April and in revised form 1 August 1999     

Computer-aided diagnosis with textural features for breast lesions in sonograms

Rationale and objectives

Computer-aided diagnosis (CAD) systems provided second beneficial support reference and enhance the diagnostic accuracy. This paper was aimed to develop and evaluate a CAD with texture analysis in the classification of breast tumors for ultrasound images.

Materials and methods

The ultrasound (US) dataset evaluated in this study composed of 1020 sonograms of region of interest (ROI) subimages from 255 patients. Two-view sonogram (longitudinal and transverse views) and four different rectangular regions were utilized to analyze each tumor. Six practical textural features from the US images were performed to classify breast tumors as benign or malignant. However, the textural features always perform as a high dimensional vector; high dimensional vector is unfavorable to differentiate breast tumors in practice. The principal component analysis (PCA) was used to reduce the dimension of textural feature vector and then the image retrieval technique was performed to differentiate between benign and malignant tumors. In the experiments, all the cases were sampled with k-fold cross-validation (k = 10) to evaluate the performance with receiver operating characteristic (ROC) curve.

Results

The area (A_Z) under the ROC curve for the proposed CAD system with the specific textural features was 0.925 ± 0.019. The classification ability for breast tumor with textural information is satisfactory.

Conclusions

This system differentiates benign from malignant breast tumors with a good result and is therefore clinically useful to provide a second opinion.     

The breast lesion excision system (BLES) under stereotactic guidance cannot be used as a therapeutic tool in the excision of small areas of microcalcifications in the breast

ObjectiveThe breast lesion excision system (BLES) is a new, automatic percutaneous breast biopsy device that excises single large specimens using radiofrequency cutting. The aim of this study was to determine whether BLES, under stereotactic guidance, can be used as a therapeutic tool in the assessment of small areas of microcalcifications in the breast by providing samples with clear margins.Material and methodsIn this retrospective study, 149 patients with suspicious (BIRADS 4 or 5) small areas of microcalcifications underwent stereotactic-guided BLES. Of these, 34 patients (22.8%) with microcalcifications that had a diameter smaller than the basket size (≤15 mm) underwent both BLES and subsequent surgery. Histopathology findings from BLES and subsequent surgery were compared. Identical, underestimation and total excision findings were assessed.ResultsBLES revealed fourteen (41.1%) high-risk lesions, ten (29.4%) ductal carcinomas in situ, and ten (29.4%) invasive cancers. Identical results between BLES and surgery were seen in 17/34 (50%) lesions. Surgery confirmed total excision of BLES in 15/34 (44.1%) lesions. Underestimation was seen in 2/34 (5.8%) lesions.ConclusionBLES allows accurate diagnosis of small areas of microcalcifications, with few underestimates. BLES is a diagnostic, but cannot be considered to be a therapeutic tool in the case of suspicious microcalcifications because total excision was seen in only 44.1% of these lesions. Studies are needed to address the therapeutic benefit of this procedure in solid lesions.     

Let's go out of the breast: Prevalence of extra-mammary findings and their characterization on breast MRI

Purpose

The aim of this study is to assess the prevalence, the site and the nature of extra-mammary findings on breast magnetic resonance imaging (MRI) and to determine its accuracy in the characterization of the discovered lesions.

Materials and methods

A retrospective review of 308 female patients (mean age 50 ± 20) who underwent breast MRI with 1.5 T device was performed. 125 out of 308 (40.5%) had a positive personal history of breast cancer (pre-operative n = 80; follow-up n = 45), while the remaining 183 without history of breast cancer (high familiar risk for breast cancer n = 80; dense breast n = 103). All incidental findings were characterized by means of additional imaging (US; Bone scintigraphy-MRI; CT-PET-CT).

Results

59 incidental findings were found in 53/308 (17%) examined patients. 9/59 incidental findings (15%) were confirmed to be malignant while the remaining 50/59 (84%) benign. The most common site was the liver (33/59; 55.8%), followed by the lung (6/59; 10.1%), bone (6/59; 10.1%), diaphragm (6/59; 10.1%) spleen (3/59; 5%), kidney (2/59; 3.4%), gall bladder (1/5; 1.5%), ascending aorta (1/59; 1.5%), thyroid (1/59; 1.5%). The incidence of malignant incidental findings resulted to be higher in the group of patients with personal breast cancer (36%) than in the other one (8%). By comparing MRI findings with the additional definitive imaging tools, breast MRI allowed a correct diagnosis in 58/59 cases with a diagnostic accuracy value of 98%.

Conclusion

Incidental extramammary findings on breast MRI are common. Benign lesions represent the most frequent findings, however malignant ones need to be searched especially in patients with personal history of breast cancer because they could influence the clinical patient management. Breast MRI can characterize incidental findings with high accuracy value.     

乳腺癌组织学分级与ER、PR、CerbB-2、Ki-67表达的关系

目的探讨乳腺癌组织学分级与雌激素受体（ER）、孕激素受体（PR）、人表皮生长因子受体-2（CerbB-2）、Ki-67表达的相关性。方法用免疫组织化学法测定169例乳腺癌组织中ER、PR、CerbB-2、Ki-67的表达;Bloom-Richardson系统Nottingham改良方案测定组织学分级。结果169例患者ER、PR、CerbB-2和Ki-67表达的阳性率分别为56.21%、63.91%、54.44%和75.15%;Ⅰ、Ⅱ、Ⅲ级乳腺癌组织中ER、PR的阳性表达率间差异无显著性意义（P〉0.05）;随着组织学分级的递增CerbB-2、Ki-67的阳性表达率随之增加（P〈0.05）;ER的表达与CerbB-2、Ki-67的表达间有相关关系（P〈0.05）。结论乳腺癌组织中ER、PR的表达不随组织学分级的增加而增加;CerbB-2、Ki-67的表达随组织学分级的增加而增加;ER的表达与CerbB-2呈负相关,而与Ki-67呈正相关。     

Multifeature analysis of Gd-enhanced MR images of breast lesions

The objective of this study was to determine whether linear discriminant analysis of different independent features of MR images of breast lesions can increase the sensitivity and specificity of this technique. For MR images of 23 benign and 20 malignant breast lesions, three independent classes of features, including characteristics of Gd-DTPA-uptake curve, boundary, and texture were evaluated. The three classes included five, four and eight features each, respectively. Discriminant analysis was applied both within and across the three classes, to find the best combination of features yielding the highest classification accuracy. The highest specificity and sensitivity of the different classes considered independently were as follows: Gd-up-take curves, 83% and 70%; boundary features, 86% and 70%; and texture, 70% and 75%, respectively. A combination of one feature each from the first two classes and age yielded a specificity of 79% and sensitivity of 90%, whereas highest figures of 93% and 95%, respectively, were obtained when a total of 10 features were combined across different classes. Statistical analysis of different independent classes of features in MR images of breast lesions can improve the classification accuracy of this technique significantly.     

Value of MR imaging in clinical evaluation of breast lesions

Purpose:
To evaluate the value of dynamic MR imaging as an adjunctive tool to triple diagnosis (TD) (physical examination, mammography and fine needle aspiration cytology) in diagnosing breast lesions. Material and Methods:
Ninety-three consecutive patients with 114 palpable or mammographically detected breast lesions were examined with TD and MR imaging. The MR examination was performed dynamically using a dedicated breast coil. Five diagnostic groups were defined on a scale from 1=normal, to 5 = malignant, where groups 1-3 were defined as benign and groups 4-5 as malignant. All lesions were histopathologically examined. The sensitivity and specificity were calculated for mammography alone, TD, MR and the combination of MR and TD. Results:
Histopathology revealed 32 benign and 82 malignant lesions (73 invasive and 9 cancer in situ). The sensitivity/specificity was 84%/59% for mammography, 93%/41% for TD, 94%/47% for MR and 99%/19% for TD + MR. In the 32 mammographically dense breasts, mammography/TD/MR/TD + MR had a sensitivity of 72%/94%/94%/100% and a specificity of 79%/57%/36%/29%, respectively. Conclusion:
MR imaging might be of value as an additive method to TD by increasing the sensitivity, but at the cost of decreasing specificity. Patients with mammographically dense parenchyma might have more benefit of the additive value, as the decrease in specificity in these patients was less pronounced in our study.     

Indeterminate or suspicious breast lesions detected initially with MR imaging: value of MRI-directed breast ultrasound

RATIONALE AND OBJECTIVES: To retrospectively determine the value of magnetic resonance imaging (MRI)-directed breast ultrasonography (US) in the evaluation of indeterminate or suspicious lesions identified on contrast-enhanced, breast MRI. MATERIALS AND METHODS: A total of 395 patients presenting for breast MRI during a 4-year period was retrospectively reviewed. Seventy-one patients were recommended for MRI-directed US for further characterization of indeterminate or suspicious breast lesions detected on MRI. Fifty-five patients (all female; age 31-80 years) had US. Their MRI and US were reviewed and tested for correlations with histologic results or long term follow-up. Logistic regression analyses were used to test for associations between MRI lesion characteristics and US detection rate. RESULTS: US identified 46 of 97 (47%) lesions depicted at MRI from 55 patients (47 [85%] of these patients had histories of breast malignancies). Twelve cancers were found from the 97 lesions (12%). Biopsy was avoidable in 10 lesions (10%). The detection rate with US was slightly higher with "mass" (55% [23/42]) lesions described in MRI than "non-mass" lesions or lymph nodes (42% [23/55]). There was a significant positive association (odd ratio = 1.23: 95% CI = 1.05-1.43, P = .01) between US detection rate and MRI mass lesion size. There was no statistical significance between US detection rate and the presence of malignancies; 42% (5/12) of MRI malignant lesions were not visualized with US. CONCLUSIONS: MRI-directed US reduced the number of biopsies required for indeterminate or suspicious MRI lesions. Nevertheless, the lesions which were biopsied had a low rate of malignancy.     

Accuracy of ultrasound-guided,large-core needle breast biopsy

Ultrasound-guided, large-core needle biopsy (US-LCNB) of suspicious breast lesions is acknowledged as less invasive and less expensive and less time consuming than surgical biopsy, and provides a histologic diagnosis with a comparable high degree. US-LCNB has been proven to help reduce the number of unnecessary surgeries for benign disease. Its limitations, however, are false-negative results and underestimation of disease. Thus, the demand for breast teams is to carefully adhere to the principles of triple assessment and imaging-histologic correlation, and follow-up of lesions with a specific benign histology after biopsy. Also, the acceptance of guidelines and rigorous quality controls help to reliably minimize the delay in the diagnosis of breast cancer in patients with false-negative biopsies. This paper aims to summarize the equipment and methods as well as the benefits and limitations of US-LCNB. Also, guidelines of quality assessment are suggested. Finally, recent developments which may help to overcome the limitations of US-LCNB will be discussed, i.e., directional vacuum-assisted biopsy (VAB), three-dimensional (3D) US-guided biopsy, as well as the use of tissue harmonic imaging (THI) and compound imaging (CI) during biopsy.     

Stereotactic biopsy of noncalcified breast lesions: utility of vacuum-assisted technique compared to multipass automated gun technique

The purpose of this study was to evaluate the stereotactic 14-gauge, vacuum-assisted biopsy technique in diagnosing noncalcified breast lesions. Stereotactic biopsy of 116 noncalcified breast lesions was performed with either 14-gauge, vacuum-assisted technique or multipass, automated large-core technique. The number of core samples and time required for each biopsy were compared. Outcome for each technique was compared based on recommendations after the mammographic-histologic review process. Results from surgical excision and mammographic follow-up were reviewed. The vacuum-assisted technique obtained more tissue cores per lesion than the automated gun technique, showed marginal decrease in number of repeat biopsies for discordant results, and required fewer short-term follow-up mammograms, although the difference did not achieve statistical significance. The vacuum-assisted technique can be used successfully to sample noncalcified breast masses, with marginal improvement over the automated gun technique.     

彩色多普勒超声对良恶性乳腺肿瘤的研究

目的：探讨彩色多普勒超声（ColourDopplerUltrasonography．CDU）在判断乳腺肿物良恶性和腋淋巴结转移的价值。材料和方法：作者分析了58例乳腺实性肿物的彩色多普勒超声和临床怀疑乳腺癌且CDU为阳性的30例病人的腋淋巴结CDU扫查结果。结果：发现34例乳腺癌中26例CDU有阳性表现，而24例良性肿物则均无阳性结果。得出CDU对乳腺癌的敏感性为77％，特异性100％，阳性符合率100％，肿癌直径小于1cm和浸润性小叶癌的CDU阳性率最低。对照术后病理，10例腋淋巴结CDU阳性者均已有淋巴结浸润，另有4例已受累的淋巴结CDU为阴性。CDU对腋淋巴结受累敏感性为71％，特异性100％，阳性符合率100％。结论：CDU对乳腺肿物良恶性以及腋淋巴结转移的判断上有重要的应用价值。