The Clinical Significance of Real-Time Contrast-Enhanced Ultrasonography in the Differential Diagnosis of Breast Tumor

The objective of this study was to investigate the clinical significance of real-time contrast-enhanced ultrasonography in the differential diagnosis of breast tumor. Fifty-seven breast tumor patients with 63 lesions were studied. Among the lesions, 34 are malignant and 29 are benign. A Philips iU-22 ultrasound scanner with L12-5 probe was used. Bolus SonoVue was injected via antecubital vein. Dynamic imaging was stored and analyzed with QLAB software. Parameters including initial time of perfusion (ITP), time to peak (TTP), peak intensity (PI), the enhancement pattern and the wash out pattern (WP) of contrast agent were observed. Results showed that about 85.3% of the malignant lesions showed heterogeneous enhancement and 79.3% of the benign ones showed homogeneous enhancement. The persistence time of the contrast agents was clearly longer inside the malignant lesion than inside the benign ones. Nevertheless, there were no significant differences in the value as ITP and TTP between the malignant and the benign lesions, while the PI value of the malignant lesions was significantly higher than the benign lesions. This study suggested that real-time contrast-enhanced ultrasonography is helpful to the differential diagnosis of breast tumors; however, the WP of the contrast agent inside the lesion also seems to be an important factor.     

Algorithm for a DNA-cytophotometric diagnosis and grading of malignancy

An algorithm for processing data on nuclear DNA content obtained cytophotometrically was developed (1) to obtain an objective discrimination between benign and malignant lesions in conventional cytologic smears secondarily stained according to Feulgen and (2) to obtain an objective degree of tumor malignancy on a continuous scale of malignancy grades. Investigations in 258 malignant tumors (95 malignant lymphomas, 52 uterine cervix carcinomas, 28 prostate carcinomas, 18 breast carcinomas, 45 malignant bone tumors and 19 larynx carcinomas) and in 74 benign lesions in these organs yielded a diagnostic accuracy of no false-positive, no false-negative and 21% suspicious diagnoses. The probability that "suspicious" cases were malignant was 81%. The overall diagnostic accuracy for non-negative cases thus amounted to 100%. Results in 95 patients with different malignant lymphomas and in 16 patients with squamous-cell carcinoma of the larynx demonstrated the prognostic validity of the DNA-grading system.     

Cytochemistry of imprints from breast biopsies

In order to investigate the cytoplasmic features of cells from benign and malignant breast lesions, a series of cytochemical reactions was applied to 117 sets of imprints made from breast biopsies. Alkaline phosphatase was present in all benign lesions except one but absent in 56% of the carcinomas. Lipids detected by the oil red O method were significantly more common in carcinomas but also present in a substantial number of benign lesions, making this difference of minor importance in diagnostic cytology. Carcinomas with a cytoplasmic lipid content stainable with oil red O and with moderate or marked sudanophilia had a significantly higher rate of axillary lymph node metastases, illustrating the prognostic value of the lipid-staining methods. Intracytoplasmic inclusions (ICI) detected by the periodic acid Schiff (PAS) method were found in 62% of the carcinomas and in only one benign lesion, making their presence of great diagnostic value. It is concluded that the application of certain cytochemical methods may improve the diagnostic accuracy of breast cytology.     

Classification of breast lesions based on a dual S-shaped logistic model in dynamic contrast enhanced magnetic resonance imaging

This study proposes a novel dual S-shaped logistic model for automatically quantifying the characteristic kinetic curves of breast lesions and for distinguishing malignant from benign breast tumors on dynamic contrast enhanced (DCE) magnetic resonance (MR) images. D(α,β) is the diagnostic parameter derived from the logistic model. Significant differences were found in D(α,β) between the malignant benign groups. Fisher's Linear Discriminant analysis correctly classified more than 90% of the benign and malignant kinetic breast data using the derived diagnostic parameter (D(α,β)). Receiver operating characteristic curve analysis of the derived diagnostic parameter (D(α,β)) indicated high sensitivity and specificity to differentiate malignancy from benignancy. The dual S-shaped logistic model was effectively used to fit the kinetic curves of breast lesions in DCE-MR. Separation between benign and malignant breast lesions was achieved with sufficient accuracy by using the derived diagnostic parameter D(α,β) as the lesion's feature. The proposed method therefore has the potential for computer-aided diagnosis in breast tumors.     

剪切波弹性成像在乳腺实性病变良恶性鉴别中的诊断价值

目的:探讨剪切波弹性成像在乳腺实性病变良恶性鉴别中的诊断价值。方法:收集2012年3月-2013年6月于我院收治的乳腺实性病变患者54例,共65个病灶,先后给予乳腺二维超声检查与剪切波弹性成像检查,采用弹性模量值与钼靶BI-RADS分级方法诊断,比较两种方法诊断的准确性。结果:良性病灶组弹性最小、最大值以及平均值、标准差与恶性病灶组比较差异具有统计学意义(P0.05);ROC曲线分析显示,在乳腺实性病变良恶性的鉴别中,弹性最大值明显优于平弹性均值;剪切波弹性成像对乳腺实性病变良恶性鉴别的敏感度、特异度、准确度、阳性预测值、阴性预测值高于二维超声技术(P0.05)。结论:剪切波弹性成像在乳腺实性病变良恶性鉴别中具有良好的诊断价值,能够提高诊断的准确性。     

Can Spectral CT Imaging Improve the Differentiation between Malignant and Benign Solitary Pulmonary Nodules?

Purpose

To quantitatively assess the value of dual-energy CT (DECT) in differentiating malignancy and benignity of solitary pulmonary nodules.

Materials and Methods

Sixty-three patients with solitary pulmonary nodules detected by CT plain scan underwent contrast enhanced CT scans in arterial phase (AP) and venous phase (VP) with spectral imaging mode for tumor type differentiation. The Gemstone Spectral Imaging (GSI) viewer was used for image display and data analysis. Region of interest was placed on the relatively homogeneous area of the nodule to measure iodine concentration (IC) on iodine-based material decomposition images and CT numbers on monochromatic image sets to generate spectral HU curve. Normalized IC (NIC), slope of the spectral HU curve (λ_HU) and net CT number enhancement on 70keV images were calculated. The two-sample t-test was used to compare quantitative parameters. Receiver operating characteristic curves were generated to calculate sensitivity and specificity.

Results

There were 63 nodules, with 37 malignant nodules (59%) and 26 benign nodules (41%). NIC, λ_HU and net CT number enhancement on 70keV images for malignant nodules were all greater than those of benign nodules. NIC and λ_HU had intermediate to high performances to differentiate malignant nodules from benign ones with the areas under curve of 0.89 and 0.86 respectively in AP, 0.96 and 0.89 respectively in VP. Using 0.30 as a threshold value for NIC in VP, one could obtain sensitivity of 93.8% and specificity of 85.7% for differentiating malignant from benign solitary pulmonary nodules. These values were statistically higher than the corresponding values of 74.2% and 53.8% obtained with the conventional CT number enhancement.

Conclusions

DECT imaging with GSI mode provides more promising value in quantitative way for distinguishing malignant nodules from benign ones than CT enhancement numbers.     

Analysis of fine needle aspiration cytology of the breast: a review of 1,297 cases and correlation with histologic diagnoses

OBJECTIVE: To evaluate the accuracy of fine needle aspiration (FNA) cytology of the breast and to ascertain its usefulness. STUDY DESIGN: The authors reviewed 1,297 cases of FNA cytology of the breast which were performed at Chonnam National University Hospital from 1999 to 2002. Cytologic diagnoses were compared with histologic diagnoses in 457 cases that underwent both cytologic and histologic examination. RESULTS: Of 1,297 cases, 1,201 (92.6%) were satisfactory and 96 (7.4%) unsatisfactory. Subsequent histologic examination was performed on 291 cases (29.7%) out of 981 "benign" lesions, 28 (73.7%) of 38 "suspicious," 124 (68.1%) of 182 "malignant" and 14 (14.6%) of 96 "unsatisfactory." FNA cytology revealed 77.7% sensitivity, 99.2% specificity, 98.4% positive predictive value and 88.0% negative predictive value. Diagnostic accuracy was 91.1%. Of the 291 benign cases on cytology, 35 cases were malignancy on histology. Of the 124 cases reported as malignant, 2 were benign. Interpretive error was the leading cause of false positive diagnoses. CONCLUSION: Although FNA cytology is a useful diagnostic procedure for the evaluation of breast lesions, it should be combined with other diagnostic modalities, such as physical examination, ultrasonography and mammography.     

Accuracy of [18F]FDG PET/MRI for the Detection of Liver Metastases

Background

The aim of this study was to compare the diagnostic accuracy of [¹⁸F]FDG-PET/MRI with PET/CT for the detection of liver metastases.

Methods

32 patients with solid malignancies underwent [¹⁸F]FDG-PET/CT and subsequent PET/MRI of the liver. Two readers assessed both datasets regarding lesion characterization (benign, indeterminate, malignant), conspicuity and diagnostic confidence. An imaging follow-up (mean interval: 185±92 days) and/-or histopathological specimen served as standards of reference. Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were calculated for both modalities. Accuracy was determined by calculating the area under the receiver operating characteristic (ROC) curve. Values of conspicuity and diagnostic confidence were compared using Wilcoxon-signed-rank test.

Results

The standard of reference revealed 113 liver lesions in 26 patients (malignant: n = 45; benign: n = 68). For PET/MRI a higher accuracy (PET/CT: 82.4%; PET/MRI: 96.1%; p<0.001) as well as sensitivity (67.8% vs. 92.2%, p<0.01) and NPV (82.0% vs. 95.1%, p<0.05) were observed. PET/MRI offered higher lesion conspicuity (PET/CT: 2.0±1.1 [median: 2; range 0–3]; PET/MRI: 2.8±0.5 [median: 3; range 0–3]; p<0.001) and diagnostic confidence (PET/CT: 2.0±0.8 [median: 2; range: 1–3]; PET/MRI 2.6±0.6 [median: 3; range: 1–3]; p<0.001). Furthermore, PET/MRI enabled the detection of additional PET-negative metastases (reader 1: 10; reader 2: 12).

Conclusions

PET/MRI offers higher diagnostic accuracy compared to PET/CT for the detection of liver metastases.     

Fine needle aspiration cytology of the breast. Experience at an outpatient breast clinic

OBJECTIVE: To evaluate the accuracy of fine needle aspiration cytology (FNAC) of the breast at our institution and to perform quality assurance. STUDY DESIGN: Two hundred forty-six cases with pathologic confirmation were selected and reviewed. A pathologist performed most of the aspirations at an outpatient breast clinic. We correlated cytologic and histologic findings and evaluated the influence of the size, location, grade, and pathologic subtypes and fibrosis in breast lesions on diagnostic results. RESULTS: The likelihood ratios for malignant, suspicious, atypical, benign and unsatisfactory cytologic diagnoses were 98.71, 5.48, 1.09, 0.07 and 0.55, respectively. The absolute and complete sensitivities for malignant lesions were 64.5% and 90.3%, respectively. The specificity was 71.9%. False negative and positive rates were 4.3% and 0.7%, respectively. The predictive value for a malignant cytologic diagnosis was 98.4%. The rate of unsatisfactory samples was 9.3%. The rate of concordance between cytologic and histologic diagnosis was lower for large and diffusely growing lesions (benign and malignant), for malignancies with abundant fibrosis and of unusual types and for carcinomas of low grade. All axillary and recurrent chest wall lesions were diagnosed cytologically. Cell block sections were useful in a small number of cases. CONCLUSION: Understanding the performance and limitations of FNAC can enhance its value as a diagnostic technique in the management of breast disease.     

Fine needle aspiration cytology of breast fibroadenoma. A cytohistologic correlation study of 405 cases.

OBJECTIVE: To evaluate the cytohistologic correlation of breast fibroadenoma (FA) in order to assess the value of cytology in the conservative management of this lesion. STUDY DESIGN: A retrospective analysis of all aspirates diagnosed as FA or fibroadenomatous lesion (n = 1,549) for which a histologic follow-up study was available (n = 362). Forty-three aspirates, including 14 nonrepresentative smears, from histologically proven FAs with a different cytologic report were also included in the study. RESULTS: Cytohistologic agreement was present in 287 of the 362 cytodiagnoses. Lack of correlation was observed in 75 cases. Most diagnostic errors accumulated in the older patient group. The sensitivity of the cytologic diagnosis of FA was 86.9% (90.8% excluding nonrepresentative cases), with a positive predictive value of 79.3%. In 43 cases a histologic diagnosis of FA was given after previous erroneous (n = 29) or nonrepresentative cytodiagnoses (n = 14). The specificity of the cytologic diagnosis of FA reached 93.8%, with a negative predictive value of 96.3% (97.5% excluding nonrepresentative cases). Regarding malignancy, five tumors were diagnosed as FA and were malignant. No false positive diagnoses of malignancy were given, but nine aspirates were included in the category "suspicious for carcinoma." CONCLUSION: FA of the breast remains a diagnostic challenge for the cytopathologist. A considerable amount of benign breast lesions can mimic FA on cytology, and such diagnostic categories as "fibroadenomatous lesion" or "consistent with FA" are associated with low diagnostic accuracy. While the cytologic requisites for entering a program of conservative management of FA are established, it seems that strict diagnostic criteria should be applied even at the expense of diminishing sensitivity.     

Importance of guided fine needle aspiration cytology (FNAC) for the diagnostics of thyroid nodules - own experience

The aim of our study was to assess the accuracy and limitations of ultrasound guided fine-needle aspiration cytology (FNAC) of thyroid nodules. METHODS: The usg-FNAC results of 245 patients with thyroid nodules, who afterwards underwent thyroid surgery or who died, and autopsies were carried out, and compared retrospectively with cytologic results. Patients with malignant cytologic conclusion without histological confirmation after surgery or autopsy were excluded from the study (9 persons). The usg-FNAC results were divided as follows: group 1: diagnosis of malignancy (n = 30), group 2: suspicion of malignancy (n = 28), group 3: benign (n = 126), group 4: inconclusive (n = 29). RESULTS: Assuming the cytologic results of group 1 and group 2 were interpreted as being malignant and those of group 3 as being benign, the sensitivity, specificity and accuracy of usg-FNAC were 90 %, 85 % and 86 % respectively. Comparing the cytologic conclusions between a group of patients with follicular lesions and a subgroup of other lesions a statistically significant difference (p < 0,01) between both subgroups using Fisher's test was found. Sensitivity, specificity and diagnostic accuracy in subgroup of follicular lesions were low (71 %, 63 %, 67 %), while in the subgroup of other lesions were high (94 %, 86 %, 88 %). FNAC can specify the nature of focal lesion with high sensitivity, specificity and diagnostic accuracy in the cases of non-follicular lesions. Histological evaluation is required to specification of the nature in cases in which cytology is indicative of follicular proliferation.     

Fine needle aspiration biopsy of vertebral lesions

OBJECTIVE: To evaluate the accuracy of fine needle aspiration biopsy (FNAB) in the diagnosis of vertebral lesions. STUDY DESIGN: Eighty-nine FNAB cases of vertebral lesions from January 1996 to December 2001 were retrieved from the Allegheny General Hospital laboratory information system. The cases were reviewed and correlated with clinical findings, including previous clinical history, primary site of malignancy and final pathologic diagnosis. RESULTS: ENAB diagnoses were malignant in 43 cases, benign in 35, suspicious in 1, unsatisfactory in 7 and false negative in 3. Previous clinical history included malignancy (37 patients), osteomyelitis and systemic disease (11), and nonspecific or no history (41). In 34 cases (38.2%) both aspirates and core biopsies were available, and the diagnoses correlated in 29/34 cases (85%). Surgical or core biopsies in the unsatisfactory/suspicious group showed malignancy in 4 cases (50%). The sensitivity of FNAB of vertebral lesions was 96%, specificity 100%, positive predictive value 100% and negative predictive value 92%, with no false positive cases. CONCLUSION: FNAB of vertebral lesions is an effective, sensitive and specific procedure in the diagnostic workup of a patient with or without a prior history of malignancy. Surgical pathology examination, including core biopsies of unsatisfactory or suspicious lesions, can further improve the diagnostic yield.     

Characterization of Incidental Liver Lesions: Comparison of Multidetector CT versus Gd-EOB-DTPA-Enhanced MR Imaging

As a result of recent developments in imaging modalities and wide spread routine medical checkups and screening, more incidental liver lesions are found frequently on US these days. When incidental liver lesions are found on US, physicians have to make a decision whether to just follow up or to undergo additional imaging studies for lesion characterization. In order to choose the next appropriate imaging modality, the diagnostic accuracy of each imaging study needs to be considered. Therefore, we tried to compare the accuracy of contrast-enhanced multidetector CT (MDCT) and Gd-EOB-DTPA-enhanced MRI for characterization of incidental liver masses. We included 127 incidentally found focal liver lesions (94 benign and 33 malignant) from 80 patients (M∶F = 45∶35) without primary extrahepatic malignancy or chronic liver disease. Two radiologists independently reviewed Gd-EOB-DTPA-enhanced MRI and MDCT. The proportion of confident interpretations for differentiation of benign and malignant lesions and for the specific diagnosis of diseases were compared. The proportion of confident interpretations for the differentiation of benign and malignant lesions was significantly higher with EOB-MRI(94.5%–97.6%) than with MDCT (74.0%–92.9%). In terms of specific diagnosis, sensitivity and accuracy were significantly higher with EOB-MRI than with MDCT for the diagnosis of focal nodular hyperplasia (FNH) and focal eosinophilic infiltration. The diagnoses of the remaining diseases were comparable between EOB-MRI and MDCT. Hence, our results suggested that Gd-EOB-MRI may provide a higher proportion of confident interpretations than MDCT, especially for the diagnosis of incidentally found FNH and focal eosinophilic infiltration.     

Q-Elastography in the Presurgical Diagnosis of Thyroid Nodules with Indeterminate Cytology

Quantitative ultrasound (US) elastography (Q-USE), able to evaluate tissue stiffness has been indicated as a new diagnostic tool to differentiate benign from malignant thyroid lesions. Aim of this prospective study, conducted at the Department of Surgical Sciences, of the “Sapienza” University of Rome, was to evaluate the diagnostic accuracy of Q-USE, compared with US parameters, in thyroid nodules with indeterminate cytology (Thy3).The case study included 140 nodules from 140 consecutive patients. Patient’s thyroid nodules were evaluated by Q-USE, measuring the strain ratio (SR) of stiffness between nodular and surrounding normal thyroid tissue, and conventional US parameters prior fine-needle aspiration cytology. Those with Thy3 diagnosis were included in the study. Forty of the nodules analyzed harbored a malignant lesion. Q-USE demonstrated that malignant nodules have a significant higher stiffness with respect to benign one and an optimun SR cut-off value of 2.05 was individuated following ROC analysis. Univariate analysis showed that hypoechogenicity, irregular margins and SR >2.05 associated with malignancy, with an accuracy of 67.2%, 81,0% and 89.8%, respectively. Data were unaffected by nodule size or thyroiditis. These findings were confirmed in multivariate analysis demonstrating a significant association of the SR and the irregular margins with thyroid nodule’s malignancy. In conclusion, we demonstrated the diagnostic utility of Q-USE in the differential diagnosis of thyroid nodules with indeterminate cytology that, if confirmed, could be of major clinical utility in patients’ presurgical selection.     

3.0 T磁共振扩散加权成像在乳腺良恶性病变鉴别中的价值及较优b值下ADC值与预后因子相关性研究

摘要 目的：研究3.0 T磁共振扩散加权成像在乳腺良恶性病变鉴别中的价值及较优b值下ADC值与预后因子的相关性。方法：选取2017年11月～2019年11月于我院接受诊治的乳腺病变患者50例进行研究,将其按照良恶性差异分成恶性组40例与良性组10例,另取同期于我院体检的健康志愿者50例作为对照组。对所有人员均进行3.0 T磁共振扩散加权成像,比较不同b值下ADC值在不同乳腺组织中的差异,比较不同b值下诊断乳腺良恶性病变的效能,分析较优b值下ADC值和乳腺癌患者各项预后因子的相关性。结果：对照组、良性组、恶性组在不同b值下的ADC值均呈逐渐降低趋势（P＜0.05）;对照组、良性组、恶性组b值为1000 s/mm²下的ADC值均低于b值为600 s/mm²（P＜0.05）。b值为1000 s/mm²时诊断乳腺恶性病变的敏感度、特异度、准确度分别为92.50%、100.00%、94.00%,高于b值为600 s/mm²的70.00%、60.00%、68.00%（P＜0.05）。b值为1000 s/mm²下雌激素受体、孕激素受阳性患者的ADC值低于阴性患者,而人类表皮生长因子受体2阳性患者的ADC值高于阴性患者（P＜0.05）。经Spearman相关性分析可得,b值为1000 s/mm²下ADC值与雌激素受体、孕激素受体阳性表达均呈负相关关系,而与人类表皮生长因子受体2阳性表达呈正相关关系（P＜0.05）。结论：3.0 T磁共振扩散加权成像在乳腺良恶性病变鉴别中的价值较高,且以b值为1000 s/mm²的诊断能效较优。此外,b值下ADC值和乳腺癌部分预后因子表达状态密切相关。     

Differential diagnosis of oncocytic lesions of the breast and thyroid utilizing a semiquantitative approach.

OBJECTIVE: To assess whether a light microscopic, semiquantitative approach could reliably distinguish between benign nonneoplastic, benign neoplastic and malignant oncocytic lesions of the breast and thyroid. STUDY DESIGN: Alcohol-fixed, Papanicolaou-stained fine needle aspiration smears of histologically proven goiter and chronic thyroiditis (18 cases), Hürthle cell adenomas (7 cases), Hürthle cell carcinomas (6 cases), fibrocystic disease (17 cases), papillomas and papillomatosis (7 cases) and apocrine carcinomas (6 cases) were rated by three independent observers using the following 10 cytologic criteria: cellularity, nuclear-cytoplasmic ratio, multinucleation, nuclear size, nuclear shape, anisonucleosis, multinucleolation, nucleolar-nuclear ratio, nucleolar size and nucleolar shape. Each of these 10 cytologic criteria was rated using a 1-3 scale. The scores for all 10 features were summed to give a total score for each case. The total scores were statistically analyzed to determine the validity and reproducibility of the summed criteria. RESULTS: The summed criteria of the total scores were reproducible between the three observers, with standard deviations ranging from 1.36 to 2.88 for thyroid and 1.72 to 2.00 for breast oncocytic lesions. The ability of the total scores to differentiate benign from malignant oncocytic lesions of the breast and thyroid was confirmed by a positive predictive value for malignancy of 67% for thyroid and 72% for breast and a negative predictive value for malignancy of 100% for nonneoplastic oncocytic lesions and > 90% for benign oncocytic neoplasms in both. The Kruskal-Wallis test revealed that the total scores were able to distinguish three diagnostic categories of nonneoplastic, benign neoplastic and malignant oncocytic breast and thyroid lesions, with P < .005. CONCLUSION: Without the expenditures of additional time, costs or materials, this semiquantitative approach compared favorably with contemporary morphometric studies involving the differential diagnosis of oncocytic cell pathology in fine needle aspiration cytology.     

Fine needle aspiration of poorly defined indurated and well-defined breast lesions: a cytopathologic comparative study

OBJECTIVE: To compare the cytologic findings and diagnoses of breast fine needle aspiration (FNA) samples of well-defined lesions (WDL) with those of poorly defined indurated lesions. STUDY DESIGN: We examined 371 consecutive breast FNA specimens obtained without diagnostic image guidance. Fifty-eight lesions were described by the examining pathologists as PDILs, and the remaining 313 lesions were described as WDLs. RESULTS: Compared with WDLs, PDILs were more likely to yield hypocellular specimens deemed unsatisfactory for diagnostic evaluation (37.9% vs. 14.1%). However, a substantial number of atypical, suspicious for malignancy and malignant cases (12.1%, 5.2%, and 13.8%, respectively) were identified with PDILs. In addition, benign diagnoses were more frequently rendered with aspirates of WDLs, compared with PDILs (47.9% vs. 31.0%). In our study, FNAs of PDILs were more often diagnostic in white women < 49 years of age and in lesions measuring > 2 cm. CONCLUSION: Given the relatively high frequency of malignant, suspicious and atypical lesions detected with PDILs, FNA is a suitable first diagnostic approach for PDILs, especially considering the relatively low cost and simplicity of FNA procedures without diagnostic imaging guidance.     

Cytodiagnosis of 112 salivary gland lesions. Correlation with histologic and frozen section diagnosis.

Aspirates of 112 cases of salivary gland lesions with histologic correlation were reviewed. Fifty-five cases (49%) had frozen sections made. The 112 cases included 76 cases of benign lesions (31 cases of pleomorphic adenoma, 19 of Warthin's tumor and 26 of nonneoplastic lesions), 22 of primary salivary gland malignancy and 14 of metastatic malignant lesions. The overall accuracy in diagnosing benign and malignant lesions was 95%. The accuracy in diagnosing the exact category of neoplastic lesions was 70%. The diagnostic sensitivity for malignant lesions was 86% and the specificity, 99%. There was one false positive, in which a pleomorphic adenoma was diagnosed as small cell carcinoma. Five false-negative cases were encountered that were due to underdiagnosis of mucoepidermoid carcinoma and adenoid cystic carcinoma. The smears were reviewed, and the diagnostic pitfalls are discussed. A comparison of the cytodiagnosis and frozen section diagnosis was made. In frozen sections there were two false negatives, and two diagnoses were deferred. The overall diagnostic accuracy was 91%. The accuracy in diagnosing the exact category of neoplastic disease was 77%. The diagnostic sensitivity for malignant disease was 70% and specificity, 100%. Frozen section, however, did supplement the fine needle aspiration diagnosis in 13 cases.     

乳腺影像报告数据系统对乳腺疾病的诊断价值

目的:探讨BI-RADS(breast imaging reporting and data system)乳腺影像报告数据系统对乳腺疾病的诊断价值。方法:采用彩色多普勒超声对75例乳腺占位性病变患者(均为女性)进行双侧乳腺探查,并采用BI-RADS乳腺影像报告数据系统进行评价。选取患者BI-RADS 3类、4类、5类的94枚病灶进行研究,以组织病理学诊断结果为金标准,来探讨其关系。结果:在94个病灶中BI-RADS诊断为3类32枚,病理学结果证明良性30枚(93.75%),恶性2枚(6.25%);BI-RADS诊断为4类的36枚,病理学结果证明良性24枚(66.67%),恶性12枚(33.33%);BI-RADS诊断为5类的26枚,病理学结果证明良性2枚(7.69%),恶性24枚(92.31%)。结论:BI-RADS乳腺影像报告数据系统可用来对各级病灶进行描述、评估,有利于提高乳腺病变影像诊断的特异性,为临床医生提供了明确的病变信息,为其进行下一步治疗措施起着重要的指导价值。     

Diagnosis of Breast Masses from Dynamic Contrast-Enhanced and Diffusion-Weighted MR: A Machine Learning Approach

Purpose

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is increasingly used for breast cancer diagnosis as supplementary to conventional imaging techniques. Combining of diffusion-weighted imaging (DWI) of morphology and kinetic features from DCE-MRI to improve the discrimination power of malignant from benign breast masses is rarely reported.

Materials and Methods

The study comprised of 234 female patients with 85 benign and 149 malignant lesions. Four distinct groups of features, coupling with pathological tests, were estimated to comprehensively characterize the pictorial properties of each lesion, which was obtained by a semi-automated segmentation method. Classical machine learning scheme including feature subset selection and various classification schemes were employed to build prognostic model, which served as a foundation for evaluating the combined effects of the multi-sided features for predicting of the types of lesions. Various measurements including cross validation and receiver operating characteristics were used to quantify the diagnostic performances of each feature as well as their combination.

Results

Seven features were all found to be statistically different between the malignant and the benign groups and their combination has achieved the highest classification accuracy. The seven features include one pathological variable of age, one morphological variable of slope, three texture features of entropy, inverse difference and information correlation, one kinetic feature of SER and one DWI feature of apparent diffusion coefficient (ADC). Together with the selected diagnostic features, various classical classification schemes were used to test their discrimination power through cross validation scheme. The averaged measurements of sensitivity, specificity, AUC and accuracy are 0.85, 0.89, 90.9% and 0.93, respectively.

Conclusion

Multi-sided variables which characterize the morphological, kinetic, pathological properties and DWI measurement of ADC can dramatically improve the discriminatory power of breast lesions.