Diagnosis Of BI-RADS Research Articles

The synergy between AI and radiologist in advancing digital mammography: comparative study between stand-alone radiologist and concurrent use of artificial intelligence in BIRADS 4 and 5 female patients

BackgroundRecent significant advancements in speed and machine learning have profoundly changed artificial intelligence (AI). In order to evaluate the value of AI in the detection and diagnosis of BIRADS 4 and 5 breast lesions visible on digital mammography pictures, we compared it to a radiologist. The gold standard was tissue core biopsy and pathology. A total of 130 individuals with 134 BIRADS 4 or 5 mammography lesions were included in the study, and all relevant digital mammography pictures were exported to an AI software system.ObjectivesThe goal of this investigation was to determine how well artificial intelligence performs in digital mammography when compared to a radiologist in identifying and diagnosing BIRADS 4 and 5 breast lesions.MethodsA total of 134 BIRADS 4 and 5 breast lesions in 130 female patients were discovered using digital mammography on both the craniocaudal and mediolateral oblique planes. All mammograms were transferred to an AI software system for analysis, and the results were compared in accordance with the histopathological results, which served as standard of reference in all lesions.ResultsArtificial intelligence was found to be more accurate (90.30%) than radiologist (82.84%) and shows higher positive predictive value (94.5%) than radiologist (82.8%) regarding suspecting malignancy in digital mammography with BIRADS 4 and 5 lesions, while the radiologist achieved higher sensitivity (100%) than AI (93.7%) in detecting malignancy in BIRADS 4 and 5 lesions.ConclusionsRadiologist was found to be more sensitive than AI in detecting malignancy in BIRADS 4 and 5 lesions but AI had a higher positive predictive value. However, specificity as well as negative predictive value could not be assessed for the radiologist, hence could not be compared with AI values because the inclusion criteria of the study did not include BIRADS 1, 2 and 3 so benign-looking lesions by digital mammography were not involved to measure specificity and negative predictive values. All in all, based on the available data, AI was found to be more accurate than radiologist regarding suspecting malignancy in digital mammography. AI can run hand in hand with human experience to give best health-care service in screening and/or diagnosing patients with breast cancer.

High-temporal resolution DCE-MRI improves assessment of intra- and peri-breast lesions categorized as BI-RADS 4

BackgroundBI-RADS 4 breast lesions are suspicious for malignancy with a range from 2 to 95%, indicating that numerous benign lesions are unnecessarily biopsied. Thus, we aimed to investigate whether high-temporal-resolution dynamic contrast-enhanced MRI (H_DCE-MRI) would be superior to conventional low-temporal-resolution DCE-MRI (L_DCE-MRI) in the diagnosis of BI-RADS 4 breast lesions.MethodsThis single-center study was approved by the IRB. From April 2015 to June 2017, patients with breast lesions were prospectively included and randomly assigned to undergo either H_DCE-MRI, including 27 phases, or L_DCE-MRI, including 7 phases. Patients with BI-RADS 4 lesions were diagnosed by the senior radiologist in this study. Using a two-compartment extended Tofts model and a three-dimensional volume of interest, several pharmacokinetic parameters reflecting hemodynamics, including Ktrans, Kep, Ve, and Vp, were obtained from the intralesional, perilesional and background parenchymal enhancement areas, which were labeled the Lesion, Peri and BPE areas, respectively. Models were developed based on hemodynamic parameters, and the performance of these models in discriminating between benign and malignant lesions was evaluated by receiver operating characteristic (ROC) curve analysis.ResultsA total of 140 patients were included in the study and underwent H_DCE-MRI (n = 62) or L_DCE-MRI (n = 78) scans; 56 of these 140 patients had BI-RADS 4 lesions. Some pharmacokinetic parameters from H_DCE-MRI (Lesion_Ktrans, Kep, and Vp; Peri_Ktrans, Kep, and Vp) and from L_DCE-MRI (Lesion_Kep, Peri_Vp, BPE_Ktrans and BPE_Vp) were significantly different between benign and malignant breast lesions (P < 0.01). ROC analysis showed that Lesion_Ktrans (AUC = 0.866), Lesion_Kep (AUC = 0.929), Lesion_Vp (AUC = 0.872), Peri_Ktrans (AUC = 0.733), Peri_Kep (AUC = 0.810), and Peri_Vp (AUC = 0.857) in the H_DCE-MRI group had good discrimination performance. Parameters from the BPE area showed no differentiating ability in the H_DCE-MRI group. Lesion_Kep (AUC = 0.767), Peri_Vp (AUC = 0.726), and BPE_Ktrans and BPE_Vp (AUC = 0.687 and 0.707) could differentiate between benign and malignant breast lesions in the L_DCE-MRI group. The models were compared with the senior radiologist’s assessment for the identification of BI-RADS 4 breast lesions. The AUC, sensitivity and specificity of Lesion_Kep (0.963, 100.0%, and 88.9%, respectively) in the H_DCE-MRI group were significantly higher than those of the same parameter in the L_DCE-MRI group (0.663, 69.6% and 75.0%, respectively) for the assessment of BI-RADS 4 breast lesions. The DeLong test was conducted, and there was a significant difference only between Lesion_Kep in the H_DCE-MRI group and the senior radiologist (P = 0.04).ConclusionsPharmacokinetic parameters (Ktrans, Kep and Vp) from the intralesional and perilesional regions on high-temporal-resolution DCE-MRI, especially the intralesional Kep parameter, can improve the assessment of benign and malignant BI-RADS 4 breast lesions to avoid unnecessary biopsy.

Reducing the Range of Cancer Risk on BI-RADS 4 Subcategories via Mathematical Modelling

Breast Imaging Reporting and Data System, also known as BI-RADS is a universal system used by radiologists and doctors. It constructs a comprehensive language for the diagnosis of breast cancer. BI-RADS 4 category has a wide range of cancer risk since it is divided into 3 categories. Mathematical models play an important role in the diagnosis and treatment of cancer. In this study, data of 42 BI-RADS 4 patients taken from the Center for Breast Health, Near East University Hospital is utilized. Regarding the analysis, a mathematical model is constructed by dividing the population into 4 compartments. Sensitivity analysis is applied to the parameters with the desired outcome of a reduced range of cancer risk. Numerical simulations of the parameters are demonstrated. The results of the model have revealed that an increase in the lactation rate and early menopause have a negative correlation with the chance of being diagnosed with BI-RADS 4 whereas a positive correlation increase in age, the palpable mass, and family history is distinctive. Furthermore, the negative effects of smoking and late menopause on BI-RADS 4C diagnosis are vehemently outlined. Consequently, the model showed that the percentages of parameters play an important role in the diagnosis of BI-RADS 4 subcategories. All things considered, with the assistance of the most effective parameters, the range of cancer risks in BI-RADS 4 subcategories will decrease.

Diagnostic Accuracy of Sonomammography in Diagnosis of BIRADS 4 Suspicious Breast Lesions Keeping Histopathology as Gold Standard

Objective: To assess the diagnostic accuracy of sonomammography in diagnosis of BIRADS ≥4 suspicious breast lesions taking histopathology as gold standard. Design of the Study: It was a descriptive cross-sectional study. Study Settings: This study was carried out at Department Of Diagnostic Radiology And Diagnostic Imaging Combined Military Hospital, Lahore from January 2017 to June 2017. Material and Methods: A total of 190 patients of age 30-70 years, presenting with palpable breast lesions with BIRAD ≥4 undergoing biopsy under local anesthesia were included. Patients with already diagnosed with breast cancer and recurrent breast lesion were excluded. Then all patients were undergone USG by using a high-resolution unit (PLM703AT) with a linear array probe centered at 7, 5 MHz, along with mammography using mamomet 3000 NOVA SIEMENS, General Electric Medical Systems, Milwaukee. Patients were labeled as positive or negative. Sonomammography findings were correlated with histopathological findings. Results of the Study: In 117 sonomammography positive patients, 107 (True Positive) had malignant breast lesions and 08 (False Positive) had benign on histopathology findings. Among, 75 sonomammography negative patients, 06 (False Negative) had malignant breast lesions on histopathology whereas 69 (True Negative) had benign lesions on histopathology. Overall sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy of sonomammography in diagnosis of BIRADS ≥4 suspicious breast lesions taking histopathology as gold standard was 94.69%, 89.61%, 93.04%, 92.0% and 92.63% respectively. Conclusion: This study concluded that diagnostic accuracy of sonomammography in diagnosis of BIRADS ≥4 suspicious breast lesions is quite high. Keywords: Breast lesions, malignant, sonomammography, sensitivity.

Comparison of 99mTc-MIBI scintigraphy, ultrasound, and mammography for the diagnosis of BI-RADS 4 category lesions

BackgroundWe sought to determine the diagnostic efficacy of Breast-specific gamma imaging (BSGI) in Chinese women with BI-RADS 4 category lesions and to compare this efficacy to that of ultrasound/mammography.MethodsWe retrospectively analyzed data from 177 women that had undergone BSGI of BI-RADS 4 category lesions originally detected via ultrasound and/or mammography.ResultsOf these 177 cases, 117 (66.1%) were malignant lesions and 60 (33.9%) were benign. The sensitivity, specificity, positive predictive values, and negative predictive values of BSGI were 94.9% (111/117), 78.3% (47/60), 89.5% (111/124), and 88.7% (47/53), respectively. The specificity and positive predictive values for mammography were 48.3% (29/60) and 77.5% (107/138), while for ultrasound they were 53.3% (32/60) and 79.6% (109/137), respectively. The sensitivity and specificity of BSGI for the detection of lesions ≤1 cm in size were 90.9% (10/11) and 88.0% (22/25), respectively, while for breast lesions >1 cm in size these values were 94.3% (100/106) and 71.4% (25/35), respectively. In addition, BSGI sensitivity and specificity values for dense breast tissue were 94.0% (79/84) and 78.0% (39/50), respectively, whereas for non-dense breast tissue these vales were 97.0% (32/33) and 80.0% (8/10), respectively. The sensitivity of BSGI for invasive ductal carcinomas (IDC) and ductal carcinomas in situ (DCIS) was 98.9% (95/96) and 75.0% (9/12), respectively. The tumor to normal tissue ratio of BSGI for malignant lesions was significantly higher than for benign lesions (2.18 ± 1.17 vs 1.66 ± 0.40, t = 7.56, P<0.05).ConclusionsThese results indicate that BSGI is highly sensitive for the detection of such lesions, achieving good positive/negative predictive values. This suggests that for IDC in particular, BSGI is superior to ultrasound and mammography for the diagnosis of BI-RADS 4 category lesions, although this was less apparent for the diagnosis of DCIS lesions. BSGI exhibited excellent performance in dense breast tissue and for the detection of lesions ≤1 cm in size.

Analysis of 3D power Doppler ultrasound parameters in the diagnosis and efficacy of BI-RADS IV breast masses

Objective To investigate diagnostic efficiency of vascular parameters in the diagnosis of BI-RADS Ⅳ breast masses using three-dimensional power Doppler ultrasound (3D-PDUS). Methods After adjusting the ultrasonic instrument under property conditions, 3D-PDUS were applied to scan BI-RADS Ⅳ breast masses before surgery with QLab software, the following parameters of breast masses were obtained: total tumor volume, vascularization index (VI), flow index (FI), and vascularization flow index (VFI). According to the pathological results, the diagnostic value of vascular parameters from 3D-PDUS were assessed with receiver operating characteristic (ROC) curve, and the diagnostic efficiency were analyzed by best cut-off limits of various vascular parameters. Results The volume and vascular parameters of benign group were remarkably lower than those of malignant group (all P VFI (0.866) >FI (0.778). AUC of FI was significantly smaller than that of VI (P=0.016) and of VFI (P=0.039), while no statistical difference between that of VI and of VFI (P=0.294). The diagnostic accuracies of VI (cut-point: 1.2, sensitivity: 84.1%, specificity: 83.3%, accuracy: 83.8%) and of VFI (cut-point: 0.5, sensitivity: 82.5%, specificity: 85.2%, accuracy: 82.9%) were higher than that of FI (cut-point: 35.3, sensitivity: 74.6%, specificity: 70.4%, accuracy: 72.6%). AUC of VI (P=0.001) and of VFI (P=0.033) in small-volume group (<2 cm3) were both larger than those of large-volume group (≥2 cm3), no significant difference in AUC of FI between these two groups (P=0.09). Conclusions Vascular parameters obtained using 3D-PDUS have affirmative value in the diagnosis of BI-RADS Ⅳ breast masses, especially in masses less than 2 cm3 in volume. Key words: Ultrasonography, three-dimensional; Breast disease; Hemodynamics

Time-resolved gadolinium-enhanced MR imaging of the breast: Impact on accuracy of diagnosis of breast carcinoma.

33 Background: The purpose of the study is to differentiate benign and malignant breast lesions based on very early enhancement of the lesion on time-resolved gadolinium-enhanced MR imaging. Methods: 100 women with breast lesions were examined at 1.5 T (Siemens, Erlangen, Germany), by fast T1-weighted GRE images every 5 s for 1 min after injection of Gd-DOTA (Dotarem, Guerbet, France), followed by 5 spacially-resolved MR image series every 30 s. Images were subtracted; maximum-intensity-projection images were obtained; and images were randomized and reviewed by two blinded readers, who reviewed only the spacially-resolved MR images and gave BIRADS diagnosis. After one month, they evaluated only the time-resolved MR images and noted the time of first appearance of enhancement. Results: The patients had 249 enhancing lesions: 66 malignant on histopathology and 183 benign on basis of histopathology or 1-year imaging follow-up. On time-resolved MR imaging, 15 lesions enhanced at 0-5 s (all malignant), 23 at 5-10 s (21 malignant, 2 benign), 34 at 10-15 s (28 malignant, 6 benign), and 117 at 15-20 s (2 malignant, 115 benign). All malignant lesions enhanced before 20 s. Taking 15s as the cut-off point, early enhancement had sensitivity of 96.9%, specificity of 95.6%, positive predictive value of 88.8%, negative predictive value of 98.8%, and accuracy of 95.9% for diagnosis of carcinoma. Enhancement before 15 s corrected BIRADS diagnosis of the spacially-resolved MR imaging in 14 lesions (5.6%). Conclusions: Lesion enhancement in the first 15s on time-resolved MR imaging is a useful sign for the diagnosis of breast carcinoma.

Automatic BI-RADS Diagnosis of Breast Lesions by CAD (Computer-Aid Diagnosis)

Automatic BI-RADS Diagnosis of Breast Lesions by CAD (Computer-Aid Diagnosis)