ABSTRACT
Objective:To evaluate the performance of machine learning (ML) based on automated breast volume scanner (ABVS) radiomics in distinguishing benign and malignant BI-RADS 4 lesions.Methods:Between May to December 2020, patients with BI-RADS 4 lesions from the Affiliated Hospital of Southwest Medical University (Center 1) and Guangdong Provincial Hospital of Traditional Chinese Medicine (Center 2) were prospectively collected and divided into training cohort (Center 1) and external validation cohort (Center 2). The radiomics features of BI-RADS 4 lesions were extracted from the axial, sagittal and coronal ABVS images by MaZda software. In the training cohort, 7 feature selection methods and thirteen ML algorithms were combined in pairs to construct different ML models, and the 6 models with the best performance were verified in the external validation cohort to determine the final ML model. Finally, the diagnostic performance and confidence (5-point scale) of radiologists (R1, R2 and R3, with 3, 6 and 10 years of experience, respectively) with or without the model were evaluated.Results:①A total of 251 BI-RADS 4 lesions were enrolled, including 178 lesions (91 benign, 87 malignant) in the training cohort and 73 lesions (44 benign, 29 malignant) in the external validation cases. ②In the training cohort, the 6 ML models (DNN-RFE, AdaBoost-RFE, LR-RFE, LDA-RFE, Bagging-RFE and SVM-RFE) had the best diagnostic performance, and their AUCs were 0.972, 0.969, 0.968, 0.968, 0.967 and 0.962, respectively. ③In the external validation cohort, the DNN-RFE still had the best performance with the AUC, accuracy, sensitivity, specificity, PPV and NPV were 0.886, 0.836, 0.934, 0.776, 86.8% and 82.5%, respectively. ④Before model assistance, R1 had the worst diagnostic performance with the accuracy, sensitivity, specificity, PPV and NPV were 0.680, 0.750, 0.640, 57% and 81%, respectively. After model assistance, the diagnostic performance of R1 was significantly improved ( P=0.039), and its diagnostic sensitivity, specificity, accuracy, PPV and NPV improved to 0.730, 0.810, 0.930, 68% and 94%; while the improvement of R2 and R3 were not significantly ( P=0.811, 0.752). Meanwhile, the overall diagnostic confidence of the 3 radiologists increased from 2.8 to 3.3 ( P<0.001). Conclusions:The performance of ML based on ABVS radiomics in distinguishing between benign and malignant BI-RADS 4 lesions is good, which may improve the diagnostic performance of inexperienced radiologists and enhance diagnostic confidence.
ABSTRACT
Objective:To investigate the clinical value of conventional ultrasound combined with automated breast volume scanner (ABVS) in predicting axillary lymph node metastasis (ALNM) of patients with invasive ductal carcinoma.Methods:A retrospective analysis was performed in 96 patients in the General Hospital of Eastern Theater Command from January 2014 to December 2020. All patients were examined by conventional ultrasound and ABVS before treatment. The patients were divided into the ALNM group and non-axillary lymph node metastasis (N-ALNM) group according to the postoperative pathological results. The differences of ultrasound parameters between the two groups were compared, and Logistic regression was used to analyze the independent predictive factors of ALNM. The ROC curve was plotted to evaluate the diagnostic efficacy for ALNM.Results:Compared with the N-ALNM group, the ALNM group had the characteristics of larger long diameters, unclear boundary, uneven internal echo, Adler blood flow grade Ⅱ-Ⅲ and retraction phenomenon (all P<0.05). Logistic regression analysis showed that the long diameter, uneven internal echo, and retraction phenomenon were independent predictors of ALNM ( OR=1.051, 4.055, 3.493, all P<0.05). The area under curve of ALNM was 0.752(0.653-0.834), the sensitivity and specificity were 54.7% and 83.7%, respectively. Conclusions:The long diameter, uneven internal echo, and retraction phenomenon measured by conventional ultrasound and ABVS are independent predictors of ALNM. The combination of the three can provide imaging references for the evaluation of ALNM of breast cancer.
ABSTRACT
Objective To compare the efficiencies of handheld ultrasound ,automated breast volume scanner ( ABVS) and breast specific gamma imaging (BSGI) in the diagnosis of breast cancer . Methods A retrospective review was performed in 200 women ( 210 breast lesions) underwent handheld ultrasound , ABVS and BSGI before surgery . The results were verified with histological examination . Results There was no obvious difference among the three methods in the sensitivity for the diagnosis of breast cancer( P >0 .05) . There was no difference of specificity between handheld ultrasound and ABVS ,BSGI( P = 0 .393 , 0 .139) . Compared with BSGI ,ABVS was an imaging modality with highest specificity for the diagnosis of breast cancer( P = 0 .021) ,and there was no difference between handheld ultrasound and ABVS ,BSGI ( P =0 .07 ,0 .29) . The areas under the ROC curve of handheld ultrasound ,ABVS and BSGI were 0 .855 ,0 .894 and 0 .818 ,respectively . The difference was obvious between ABVS and BSGI ( P = 0 .02) . Conclusions The diagnostic efficacy of ABVS in diagnosis of breast malignant lesions is similar to that of handheld ultrasound . BSGI has certain clinical value in the diagnosis of breast cancer ,and it is an effective supplement for breast cancer ultrasound examination .
ABSTRACT
Objective: To investigate the diagnostic value of automatic breast volume system (ABVS) for breast carcinoma in clinical practice so as to use these reliable data to provide imaging reference for further therapy. Methods: 300 patients with suspected breast cancer were selected and the postoperative pathological examinations were chosen as the golden standard to implement screening. At the same time, 2 radiologists were selected to carry out the grading diagnoses of breast imaging reporting and data system (BI-RADS) so as to observe the consistency between diagnoses of ABVS and pathological results. Results: 210 patients were diagnosed as breast tumor by the post-operative pathological examination, and these patients included 89 malignant tumor (29.67%, 89/300). In these malignant cases, there were 42 cases were infiltrative ductal carcinoma, and its rate was 47.19% (42/89). Besides, there were 121 cases were benign tumor (40.33%), and the rate of patients with fibrosum adenoma (78 cases) was the highest in the 121 cases (64.46%, 78/121). And the diagnostic consistency of them was preferably (K=0.724). The sensitivity and specificity of ABVS in diagnosing breast carcinoma were 92.86% and 85.56%, respectively. Its positive predictive value and negative predictive value were 93.75% and 83.70%, respectively. And the Youden index was 0.78. Conclusion: ABVS has relatively high sensitivity and specificity, and the screening test has good authenticity and reproducibility. Therefore, it can provide a reliable basis for clinical diagnosis and treatment.
ABSTRACT
Objective To compare patients'satisfaction and comfort associated with automated breast volume scanner(ABVS),handheld ultrasound(HHUS)and mammography(MMG).Methods Five hundred and twenty-eight women who performed ABVS examination and previously had undergone MMG and HHUS were enrolled.BMI,bra size,with or without lesion,lesion size,pain score of MMG,ABVS and HHUS were documented and analyzed.Results The pain score of MMG was significantly higher than that of ABVS[4.0(2.0,6.0)vs 2.0(1.0,4.0)](P <0.05).92.4% of the cases were preferred for the ABVS examination,which was higher than that for the MMG(45.5%),and was similar to HHUS(94.7%).Bra size,BMI,with or without lesion and lesion size could affect the pain degree of ABVS.Conclusions Routine use of ABVS will be well tolerated by patients than MMG.
ABSTRACT
Objective To evaluate the effectiveness of automated breast volume scanner(ABVS) and magnetic resonance imaging(MRI) in the detection and diagnosis of breast cancer and to assess the value of these modalities as well as the joint use of the two.Methods In this study,a total of 50 breast lesions in 37 patients proved by surgery and pathology were included.Before operation,all patients were underwent both ABVS and dynamic contrast-enhanced (DCE-MRI) examinations,and classified into groups according to BI-RADS classification.Then the effectiveness of the two examinations were contrasted,the image features on the two techniques were analyzed,and their differential diagnosis in benign and malignant breast lesions were compared.Results The sensitivity,specificity,accuracy,positive predictive value (PPV) and negative predictive value(NPV) of ABVS and MRI for the diagnosis of breast cancer were 91.67％ and 95.83％,88.46％ and 80.77％,90.00％ and 88.00％,88.00％ and 82.14％,92.00％ and 95.45％,respectively,and there was no statistical difference between ABVS and MRI(P ＞0.05);The sensitivity and specificity of the combination of the two were 99.65％ and 97.78％,respectively.The difference of the ABVS coronal features and MRI dynamic enhancement curve types between malignant breast lesions and the benign were statistically significant (P ＜ 0.05).Conclusions Both ABVS and MRI are effective to diagnose breast cancer well,while the combination of the two can improve the diagnosis more accurately.
ABSTRACT
Objective To evaluate the diagnostic features, characteristics, value, and clinical significance of the automated breast volume scanner (ABVS) in breast lesions. Methods A total of 288 patients with breast lesions diagnosed at the Breast Surgery Department of Peking Union Medical College Hospital between 2011 and 2015 were selected. Diagnostic and image data of preoperative ABVS examinations, hand-held breast ultrasound (HHUS), and surgery or biopsy pathology were collected. Pathology and imaging report results were recorded, accounting for the retraction phenomenon; receiver operating characteristic(ROC)curve analysis was used to calculate the diagnostic performance of the single and combined diagnostic methods. Results (1) A total of 311 breast lesions were found in 288 patients using the ABVS;histopathological diagnosis showed that there were 141(45.3%)malignant lesions and 170 (54.7%)benign lesions.(2)The detection rates of the retraction phenomenon using the ABVS in malignant and benign lesions were, respectively, 31.2% (44/141) and 1.8% (3/170); the difference was statistically significant (χ2=52.075,P=0.000). The detection rates of the retraction phenomenon using the ABVS in invasive ductal carcinoma (IDC), ductal carcinoma in situ (DCIS), and other types of carcinomas were, respectively, 38.5% (40/104), 10.5% (2/19), and 11.1% (2/18). There were significant differences between IDC and DCIS and between IDC and other types of carcinomas (χ2=5.575, P=0.018; χ2=5.085, P=0.024, respectively). (3) The sensitivity, specificity, and accuracy rates of single ABVS were 89.4%, 80.6%, and 90.1%,respectively,and those of single HHUS were 91.5%,74.1%,and 91.3%,respectively,for malignant lesion diagnosis. For diagnosis with combined ABVS with HHUS, the sensitivity, specificity, and accuracy rates were 93.6%,72.9%,and 93.2%,respectively.Sensitivity and specificity rates,and the advantage ratio of the retraction phenomenon were, 31.2%, 98.2%, and 25.251, respectively. Conclusions Use of the ABVS for coronary sections with the retraction phenomenon has important clinical value in identifying malignant breast lesions,especially in identifying IDC,but ABVS cannot completely replace HHUS.ABVS combined with HHUS can improve the diagnostic capacity, and is helpful for early diagnosis of malignant breast lesions.
ABSTRACT
Objectives To investigate the diagnostic performance of the combination of ultrasound elastography and automated breast volume scanner (ABVS) in differentiation of benign and malignant breast imaging reporting and data system (BI-RADS) 4 breast lesions. Methods Data from 137 breast cancer patients (147 tumors) confirmed pathologically were analyzed. Each tumor was examined by ABVS and ultrasound elastography. All tumors were diagnosed as BI-RADS 4 by ABVS. With final pathology results as the gold standard, the predictive value in differentiating BI-RADS 4 breast lesions between ultrasound elastography and the combination of ultrasound elastography and ABVS were compared. Results There were 54 benign nodules and 93 malignant nodules in this study. The diagnostic sensitivity of ultrasound elastography and the combination of ultrasound elastography and ABVS were 94.6% and 98.9%,the specificity were 57.4% and 57.4%, the accuracy were 81.0% and 83.7%, the area under the curve were 0.858 and 0.965, respectively. The diagnostic performance of ultrasound elastography combined with ABVS was better than that of ultrasound elastography. Conclusions Ultrasound elastography have certain value in differential diagnosis of BI-RADS 4 breast lesions, especially when combining with ABVS, which will improve its diagnostic accuracy. Ultrasound elastography combined with ABVS can improve the detection rate of malignant lesions in BI-RADS 4 breast lesions and reduce the rate of preoperative biopsy, and it has a good application prospect.
ABSTRACT
Objective To compare the application of ultrasound and positron emission tomography/computed tomography (PET/CT) in diagnosis and lymph node staging of breast cancer. Methods Using automated breast volume scanner (ABVS) and PET/CT examination of 40 lesions in 38 patients, and for comparison with the pathological findings, the diagnostic accuracy of PET/CT and ABVS in breast cancer was comparatively analyzed. Results The sensitivity of ABVS in the diagnosis of breast cancer was 95.65%, the specificity was 70.59%, and the accuracy rate was 85.00%; the sensitivity of PET/CT in the diagnosis of breast cancer was 86.96%, the specificity was 82.35%, and the accuracy rate was 85.00%;the diagnostic sensitivity of their combined application was 86.96%, the specificity was 94.12%, and the accuracy rate was 90.00%. In 10 cases of breast cancer patients with lymph node metastasis, the sensitivity、specificity and accuracy rate of PET/CT in the diagnosis of metastatic lymph nodes were 90.0%, 92.8%, and 92.1% respectively; the sensitivity, specificity and accuracy rate of ultrasound in the diagnosis of metastatic lymph nodes were 70.0%, 96.4% and 89.5%, respectively. Conclusions The ABVS and PET/CT are both effective approaches in the diagnosis of breast cancer. Their combined application may significantly improve the specificity and accuracy of diagnosis. For patients with lymph node metastases, the sensitivity of PET/CT for lymph node metastases was high and conducive to clinical staging, which is an important implication in guiding the choice of clinical treatment and prognosis of patients.
ABSTRACT
Objective To explore the image performance of mucinous breast carcinoma( MBC) on automated breast volume scanner ( ABVS ) and its pathologic features ,for the purpose of improving the recognition of this disease on ABVS ,as well as diagnostic ability . Methods Retrospectively analyzed the sonogram and pathological data of 17 patients with 18 MBC lesions that included 9 PMBC ( pure mucinous breast carcinoma) and 9 MMBC( mixed mucinous breast carcinoma) ,all patients were examined by ABVS before operation and confirmed by pathology . Results Luminal A was the main molecular subtype of MBC ,as well as rarely lymph node metastases;The image of the MBC on ABVS mainly showed a round like or irregular mass ,isoechoic or slightly lower with posterior acoustic enhancement ,and clearly edge with microlobulated or angular ;While on the coronal plane ,it might presented unsmoothed contour ,retraction phenomenon and expansive growth pattern like a spherical nodular;The image characteristics of the PMBC and MMBC lesions on ABVS had no statistical difference( P > 0 .05) ,but the PMBC often performed isoechoic mass with clearly marge ,in contrast , the MMBC usually performed hypoechoic mass with indistinct and angular edge ,and the malignant image features on the coronal plane . Conclusions The three vertical sections of ABVS are beneficial to observe the morphological features of MBC completely ,which can help to identify it from the benign lesions ,reduce misdiagnosis ,and provide some guidance for MBC surgical treatment in the early stage .
ABSTRACT
Objective To investigate the correlation between retraction phenomenon on the coronal planes obtained by an automated breast volume scanner ABVS and biological behaviors of breast cancer Methods Two hundred and eighty-eight female patients with 293 pathologically proven breast cancers were included in this study All of them underwent ABVS examinations before surgery Retraction phenomenon on the coronal planes was evaluated retrospectively Correlation between retraction phenomenon and hormone receptor HR human epidermal growth factor receptor 2 HER-2 Ki-67 molecular subtypes histologic type histologic grade lymph node status lesion size and patient age were evaluated by univariate logistic regression analysis Above significant variables were then enrolled in multivariate logistic regression analysis to select independent risk factors for retraction phenomenon Results By univariate logistic regression analysis Luminal A subtype odds radio OR =8 395 P <0 001 HR positive OR=4 876 P <0 001 HER-2 negative OR = 4 407 P < 0 001 lower expression of Ki-67 OR = 3 716 P <0 001 lower grade OR=3 574 P <0 001 lymph node invasion OR=2 053 P =0 004 and smaller lesion OR=1 762 P =0 033 were risk factors for retraction phenomenon while histologic type P =0 3 1 8 and patient age P = 0 612 were not By multivariate logistic regression analysis Luminal A subtype OR=9 877 P <0 001 lower grade OR=3 573 P <0 001 and lymph node invasion OR=2 730 P = 0 004 were independent risk factors for retraction phenomenon among which Luminal A subtype was the strongest one Conclusions Retraction phenomenon on the coronal planes obtained by ABVS is closely related with Luminal A subtype and may be an imaging feature of less aggressive behaviors of breast cancer.
ABSTRACT
Objective To invesigate the reproducibility of automated breast volume scanner ( ABVS) in detection of breast lesions . Methods Thirty‐five patients with breast lesions diagnosed by conventional hand‐hold ultrasound were involved in the study . Every patient underwent ABVS twice by two different doctors . The number ,location and size of findings were documented and analyzed . Intra‐class correlation coefficients were used to evaluate the reproducibility . Results The number of findings was 67 and 68 respectively ,and the cronbach′s ɑwas 0 .997 . The ɑ of distance from nipple ,distance from skin ,and clock face location were 0 .343 ,0 .883 ,and 0 .949 . The ɑ of length superior‐inferior ,medial‐lateral ,and anterior‐posterior were 0 .938 ,0 .936 ,and 0 .948 . Conclusions The reproducibility of ABVS in detection of breast lesions is perfectly high . ABVS may be a good modality in monitoring benign lesions and evaluating neoadjuvant chemotherapy in the future .
ABSTRACT
Objective To evaluate the value of automated breast volume scanner (ABVS) and conventional ultrasound in differentiation of benign and malignant breast imaging reporting and data system (BI-RADS) 4 breast lesions. Methods Totally 239 breast lesions from 217 patients, with diagnosing of BI-RADS 4 by conventional ultrasound and automatically breast volume imaging, were analyzed retrospectively, using postoperative pathology as golden standard. The sensitivity, specificity, accuracy and area under the curve of ABVS and conventional ultrasound were calculated separately. Results There were 154 benign breast lesions, 83 malignant lesions and 2 borderline lesions. The statistical analysis results of ABVS and conventional ultrasound were 96.10% and 91.80% in sensitivity, 84.30% and 80.20% in specificity,89.30% and 84.10% in accuracy, and 0.952 and 0.833 in area under the curve. Therefore, ABVS was superior to the conventional ultrasound. Conclusion Compared with conventional ultrasound, ABVS could improve the diagnostic efficacy for BI-RADS 4 breast lesions in the aspects of sensitivity, specificity, accuracy, which was useful in detection of small and atypical breast cancer and could be used as a noninvasive and reliable complement for conventional ultrasound.
ABSTRACT
Objective To compare the application value of three different planes obtained by automated breast volume scanner for the observation of margin characteristics of breast masses.Methods Two hundred and eight women patients with 237 masses (120 benign and 1 1 7 malignant,confirmed by pathology)were included in this study.The detection rates of mass margin characteristics in three different planes were compared.The diagnostic performance of mass margin characteristics in differentiating benign and malignant breast masses in three different planes was compared by area under the ROC curve (AUC). Results The detection rate of angular margins for malignant masses in sagittal planes(41 .9%)was higher than that in coronal planes(22.2%)and axial planes(26.5%)with statistical difference (P =0.001 ,0.013);the detection rate of spiculated margins for malignant masses in coronal planes(42.7%)was higher than that in sagittal planes (1 1 .1 %)and axial planes (1 7.1 %)with statistical difference (both P < 0.001 ).No significant difference was found between any two of the three planes in the detection rates of margin characteristics for benign masses.The AUC of angular margins in sagittal planes (0.693)was higher than that in coronal planes (0.607)and axial planes (0.624);the AUC of spiculated margins in coronal planes (0.697)was higher than that in sagittal planes (0.55 1 )and axial planes (0.573 );the AUC of indistinct margins in axial planes(0.606)was higher than that in coronal planes (0.552)and sagittal planes (0.552);the AUC of microlobulated margins in axial planes (0.825)was higher than that in coronal planes (0.778) and sagittal planes (0.81 7).Conclusions Three different planes of ABVS have their own advantages for the observation of margin characteristics of breast masses.
ABSTRACT
Objective To analyze sonographic characteristics of the automated breast volume scanner (ABVS) in non-mass-like breast carcinoma.Methods 49 patients with non-mass-like breast carcinoma were confirmed by pathology after operation.All of the patients underwent ABVS,the sonographic characteristics of internal echo and microcalcifications in the lesions,the retraction phenomenon in coronal plane and surrounding change were analyzed.All the lesions were assessed by breast imaging reporting and data system (BI-RADS).Results In 49 patients with non-mass-like breast carcinoma,27(55.10％) cases were ductal carcinoma in situ,21 (42.86％) cases were infitrating ductal carcinoma,and 1 (2.04％) case were infitrating lobular carcinoma.The findings of ABVS:43 (87.76 ％) lesions appeared as segmental hypoechoic areas,41 (83.67％) lesions appeared as scattered or clustered microcalcifications,6(12.24％) lesions appeared as the retraction phenomenon in coronal plane,and 1 (2.04％) lesion appeared as pachyderma and subcutaneous dropsy.Microcalcifications in ductal carcinoma in situ appeared more than that in infitrating ductal carcinoma (85.19 ％ vs 80.95 ％,P =0.715) ;the retraction phenomenon in coronal plane of infitrating ductal carcinoma were higher than that of ductal carcinoma in situ(19.05％ vs 7.41％,P =0.383),but there were no significant difference in those between ductal carcinoma in situ with infitrating ductal carcinoma(P ＞0.05).All the cases were assessed according to BI-RADS,including 4a-5 category 44 cases,0-2 category 5 cases.The coincidence rate in diagnosis of ABVS were 89.80％.Conclusions The ABVS features of non-mass-like breast carcinoma are characteristic.Segmental hypoechoic areas,microcalcifications and the retraction phenomenon in coronal plane are important for the diagnosis of non-mass-like breast carcinoma,ABVS can improve the detect ability and decrease the missed diagnosis of non-mass-like breast carcinoma.
ABSTRACT
Objective To assess the values of automated breast volume scanner (ABVS) for differentiating of benign and malignant breast masses.Methods A total 174 breast masses in 148 patients were subjected both to conventional handheld B-mode ultrasound (HHUS) and ABVS examinations.The masses were defined as five categories of benign,probably benign,equivocal,probably malignant,and malignant with each method.The results of ABVS and HHUS were compared with pathology.By using the definitive diagnosis and the five levels of suspicion categories,receiver operating characteristic (ROC) curves were drawn to evaluate their diagnostic results.In addition,the diagnostic accuracy for breast masses of futures including retraction phenomenon and hyperechoic rim in coronal plane of ABVS was evaluated.Results The area under the ROC curve of ABVS (0.927) was larger than that of HHUS (0.903) (Z =2.256,P =0.024).The specificity and the positive predictive values both reached to 100％ and false positive rate was 0 with retraction phenomenon,and the specificity and the negative predictive value were 88.89％ and 94.51％ respectively with hyperechoic rim in coronal plane of ABVS.Conclusions ABVS plays an important role in the clinical practice.The retraction phenomenon and hyperechoic rim of breast masses in coronal plane of this new modality have high specialty for differentiating malignant from benign breast masses.