Influence of age on PPV of sonographic BI-RADS categories 3, 4, and 5

Ultrasound-based deep learning in the establishment of a breast lesion risk stratification system: a multicenter study

OBJECTIVES

To establish a breast lesion risk stratification system using ultrasound images to predict breast malignancy and assess Breast Imaging Reporting and Data System (BI-RADS) categories simultaneously.

METHODS

This multicenter study prospectively collected a dataset of ultrasound images for 5012 patients at thirty-two hospitals from December 2018 to December 2020. A deep learning (DL) model was developed to conduct binary categorization (benign and malignant) and BI-RADS categories (2, 3, 4a, 4b, 4c, and 5) simultaneously. The training set of 4212 patients and the internal test set of 416 patients were from thirty hospitals. The remaining two hospitals with 384 patients were used as an external test set. Three experienced radiologists performed a reader study on 324 patients randomly selected from the test sets. We compared the performance of the DL model with that of three radiologists and the consensus of the three radiologists.

RESULTS

In the external test set, the DL model achieved areas under the receiver operating characteristic curve (AUCs) of 0.980 and 0.945 for the binary categorization and six-way categorizations, respectively. In the reader study set, the DL BI-RADS categories achieved a similar AUC (0.901 vs. 0.933, p = 0.0632), sensitivity (90.98% vs. 95.90%, p = 0.1094), and accuracy (83.33% vs. 79.01%, p = 0.0541), but higher specificity (78.71% vs. 68.81%, p = 0.0012) than those of the consensus of the three radiologists.

CONCLUSIONS

The DL model performed well in distinguishing benign from malignant breast lesions and yielded outcomes similar to experienced radiologists. This indicates the potential applicability of the DL model in clinical diagnosis.

KEY POINTS

• The DL model can achieve binary categorization for benign and malignant breast lesions and six-way BI-RADS categorizations for categories 2, 3, 4a, 4b, 4c, and 5, simultaneously. • The DL model showed acceptable agreement with radiologists for the classification of breast lesions. • The DL model performed well in distinguishing benign from malignant breast lesions and had promise in helping reduce unnecessary biopsies of BI-RADS 4a lesions.

An Improved Nomogram to Reduce False-Positive Biopsy Rates of Breast Imaging Reporting and Data System Ultrasonography Category 4A Lesions

BACKGROUND

The NCCN clinical guidelines recommended core needle biopsy for breast lesions classified as Breast Imaging Reporting and Data System (BI-RADS) 4, while category 4A lesions are only 2-10% likely to be malignant. Thus, a large number of biopsies of BI-RADS 4A lesions were ultimately determined to be benign, and those unnecessary biopsies may incur additional costs and pains. However, it is important to emphasize that the current risk prediction model focuses primarily on the details and complex risk features of US or MG findings, which may be difficult to apply in order to benefit from the model. To stratify and manage BI-RADS 4A lesions effectively and efficiently, a more effective and practical predictive model must be developed.

METHODS

We retrospectively analyzed 465 patients with BI-RADS ultrasonography (US) category 4A lesions, diagnosed between January 2019 and July 2019 in Tianjin Medical University Cancer Institute and Hospital and National Clinical Research Center for Cancer. Univariate and multivariate logistic regression analyses were conducted to identify risk factors. To stratify and predict the malignancy of BI-RADS 4A lesions, a nomogram combining the risk factors was constructed based on the multivariate logistic regression results. In order to determine the predictive performance of our predictive model, we used the concordance index (C-index), calibration curve, and receiver operating characteristic (ROC), and the decision curve analysis (DCA) to assess the clinical benefits.

RESULTS

Based on our analysis, 16.3% (76 out of 465) of patients were pathologically diagnosed with malignant lesions, while 83.6% (389 out of 465) were diagnosed with benign lesions. According to univariate and multivariate logistic regression analysis, age (OR = 3.414, 95%CI:1.849-6.303), nipple discharge (OR = .326, 95%CI:0.157-.835), palpable lesions (OR = 1.907, 95%CI:1.004-3.621), uncircumscribed margin (US) (OR = 1.732, 95%CI:1.033-2.905), calcification (mammography, MG) (OR = 2.384, 95%CI:1.366-4.161), BI-RADS(MG) (OR = 5.345, 95%CI:2.934-9.736) were incorporated into the predictive nomogram (C-index = .773). There was good agreement between the predicted risk and the observed probability of recurrence. Furthermore, we determined that 153 was the best cutoff score for distinguishing between patients in the low- and high-risk groups. Malignant lesions were significantly more prevalent in high-risk patients than in low-risk patients.

CONCLUSION

Based on clinical, US, and MG features, we present a predictive nomogram to reliably predict the malignancy risk of BI-RADS(US) 4A lesions, which may assist clinicians in the selection of patients at low risk of malignancy and reduce the number of false-positive biopsies.

Application of a second opinion ultrasound in Breast Imaging Reporting and Data System 4A cases: can immediate biopsy be avoided?

Objective

The probability of malignancy in women who are diagnosed with a Breast Imaging Reporting and Data System (BI-RADS) 4A score is low. Application of a second opinion ultrasound (SOUS), which is low in cost and minimally invasive, may lower the biopsy rate for patients who fall into this category. This study aimed to apply SOUS to patients with a BI-RADS score of 4A and predict the pathological results of a biopsy.

Methods

One hundred seventy-eight patients were analyzed. Univariate and multivariate analyses were performed to screen for predictive factors that are associated with malignancy. Categorical alteration of downgraded, unchanged, or upgraded was made after SOUS results. Changes in category were compared with biopsies to determine their predictive value of benignancy or malignancy.

Results

Independent factors associated with malignancy were age (>50 years), tumor size (≥20 mm), margin (not circumscribed), orientation (not parallel), and peripheral location, and an upgraded categorical alteration from SOUS. Downgraded categorical alterations were associated with benignancy.

Conclusions

In BI-RADS 4A cases, a biopsy is recommended when independent factors are associated with malignancy. A downgraded result from an SOUS examination is a protective factor, supporting the likelihood of benignancy in these patients.

Prediction for Breast Cancer in BI-RADS Category 4 Lesion Categorized by Age and Breast Composition of Women in Songklanagarind Hospital

Background:

Older age and dense breast are the important risk factors for breast cancer. The ACR BI-RADS lexicon 5^th edition does not mention how patient age and breast density may affect the category assessment. The aim of this study was to investigate whether patient age and breast density influence the positive predictive value (PPV) of mammographic and ultrasonographic findings categorized as BI-RADS category 4 and subcategories 4a, 4b, and 4c among female patients.

Materials and Methods:

A retrospective study was conducted in Songklanagarind Hospital between January 1, 2016 and December 31, 2017 in female patients older than 18 years who had breast lesions categorized as BI-RADS category 4 and subcategories 4a, 4b, 4c. A total of 961 breast lesions consisted of 772 (80.33%) benign lesions and 189 (19.67%) malignant lesions. Categorization was done in each lesion based on age ranges of ≤35 years, >35 to 60 years, and >60 years and breast density according to mammographic breast composition. The PPV for each BI-RADS category was calculated based on the pathological diagnoses and were compared using the chi-square test.

Results:

The overall PPV in each subcategory was in the reference range. The PPV increased with increasing age: 4% vs. 22.63% vs. 36.67% for category 4 (p-value=0.01); 0% vs. 5.81% vs. 6.88% for subcategory 4a (p-value=0.002); 6.67% vs. 26.62% vs. 51.35% for subcategory 4b (p-value=0.001); and 33.33% vs. 76.92% vs. 81.82% for subcategory 4c (p-value=0.02). An association was not found between PPV and breast density.

Conclusion:

A significantly positive association was found between PPV and age in patients in BI-RADS subcategories 4a, 4b, and 4c. This study could not determine that mammographic breast composition according to the ACR BI-RADS 5th edition was associated with PPV due to improper sample distribution.

Does patient age affect the PPV3 of ACR BI-RADS Ultrasound categories 4 and 5 in the diagnostic setting?

OBJECTIVES

To calculate the positive predictive value of biopsies performed (PPV₃) of the Ultrasound section of the American College of Radiology Breast Imaging Reporting and Data System (ACR BI-RADS US) atlas categories 4 and 5 in different age groups and to determine whether patient age influences the PPV₃ of each category in the diagnosis of breast lesions.

METHODS

We identified 2,433 ACR BI-RADS US categories 4 and 5 lesions with a known pathological diagnosis in 2,433 women. The patients were classified into three age groups (<35, 35-50, and >50 years). The age-related PPV₃ of each category in the three age groups were calculated based on the pathological diagnoses and compared using the chi-squared test.

RESULTS

The overall PPV₃ of each category was within the reference range provided by the ACR in 2013. PPV₃ gradually increased with increasing age in patients with category 4 lesions. PPV₃ in the oldest group with subcategories 4A and 4B lesions were close to or exceeded the reference values.

CONCLUSIONS

PPV₃ and age were significantly associated in patients with category 4 lesions according to the newest edition of ACR BI-RADS US in the diagnostic setting. Closer attention should be given to older patients when assigning a final assessment category.

KEY POINTS

• In patients with category 4 lesions , the likelihood of malignancy is associated with age. • In patients with category 5 lesions, the association is not definite. • Closer attention should be given to older patients in applying the ACR BI-RADS US.

Can unenhanced breast MRI be used to decrease negative biopsy rates?

PURPOSE

We aimed to determine whether low-risk breast masses can be effectively managed with unenhanced magnetic resonance imaging (MRI) combining T2-weighted sequences with diffusion-weighted imaging (DWI) instead of immediate biopsy to decrease negative biopsy rates.

METHODS

After institutional review board and patient approvals, 141 consecutive women with 156 low-risk breast masses, who underwent unenhanced MRI and later on received a final diagnosis, were included in the study. There were 72 BI-RADS 3 masses in women with relative risk factors and 84 BI-RADS 4A masses, all referred for biopsy. Apparent diffusion coefficient (ADC) cutoff was 0.90×10-3 mm2/s. According to ADC values and T2-weighted imaging characteristics, masses were classified as either malignant or benign. Unenhanced MRI results were compared with final diagnoses obtained by histopathology or imaging surveillance, and diagnostic values were calculated.

RESULTS

Of 156 masses, 112 underwent biopsy. Four malignancies were diagnosed, three of which having ADC values lower than the cutoff. In women who rejected the biopsy, masses were stable during a follow-up of at least two years (n=44). MRI revealed 91% specificity and 99% negative predictive value (NPV) for detection of breast cancer.

CONCLUSION

Combination of T2-weighted imaging with DWI is a feasible method to further characterize breast masses with a low probability of malignancy. With the use of unenhanced MRI instead of immediate biopsy, it might be possible to decrease negative biopsy rates of low-risk breast masses.

Comparison of clinicopathological findings among patients whose mammography results were classified as category 4 subgroups of the BI-RADS

OBJECTIVE:

Our aim is to compare mammographic, demographic and clinicopathological characteristics of patients whose mammographies were classified as subgroups of BI-RADS 4 category (Breast Imaging – Reporting and Data System).

METHODS:

In total, 103 patients with mammography (Senographe 600t Senix HF; General Electric, Moulineaux, France) results classified as BI-RADS 4 were included in the study. Demographic data (age, menopause, and family history) were recorded. All data were compared among BI-RADS 4 subgroups.

RESULTS:

In all, 68.9% (71/103), 7.8% (8/103) and 23.3% (24/103) the patients were in groups BI-RADS 4A, 4B and 4C, respectively. The incidence of malignancy was higher in Groups 4B and 4C than in Group 4A (p<0.05), but similar in Groups 4B and 4C (p>0.05). Mean age was lower in Group 4B than in Groups 4A and 4C (p<0.05). A positive family history was more common in Group 4A than in Group 4B (p=0.025). The frequency of menopausal patients was greater in Groups 4A and 4C than in Group 4B (p=0.021, and 0.003, respectively).

METHODS:

The rate of malignancy was higher in Groups 4B, and 4C than in Group 4A. A positive family history was more common in Group 4A than in Group 4C. Groups 4A, and 4C patients tended to be older and were more likely to be menopausal than Group 4B patients.

Bayesian probability of malignancy with BI-RADS sonographic features

OBJECTIVES

The purpose of this study was to develop a quantitative approach for combining individual American College of Radiology Breast Imaging Reporting and Data System (BI-RADS) sonographic features of breast masses for assessing the overall probability of malignancy.

METHODS

Sonograms of solid breast masses were analyzed by 2 observers blinded to patient age, mammographic features, and lesion pathologic findings. BI-RADS sonographic features were determined by using American College of Radiology criteria. A naïve Bayes model was used to determine the probability of malignancy of all the sonographic features together and with age and BI-RADS mammographic features. The diagnostic performance for various combinations was evaluated by using the area under the receiver operating curve (Az).

RESULTS

Sonographic features had high positive and negative predictive values. The Az values for BI-RADS sonographic features for the 2 observers ranged from 0.772 to 0.884, which increased to 0.866 to 0.924 when used with patient age and BI-RADS mammographic features. The benefit of adding age and mammographic information was more marked for the observer with lower initial diagnostic performance. Age-specific analysis showed that diagnostic performance varied with age, with higher performance for patients aged 45 years and younger and patients older than 60 years compared to those aged 46 to 60 years. In 85% of cases, the diagnosis of the observers matched. When the consensus between the observers was used for diagnostic decisions, a high level of diagnostic performance (Az, 0.954) was achieved.

CONCLUSIONS

A naïve Bayes model provides a systematic approach for combining sonographic features and other patient characteristics for assessing the probability of malignancy to differentiate malignant and benign breast masses.