Reducing the number of unnecessary biopsies of US-BI-RADS 4a lesions through a deep learning method for residents-in-training: a cross-sectional study

Abstract

ObjectiveThe aim of the study is to explore the potential value of S-Detect for residents-in-training, a computer-assisted diagnosis system based on deep learning (DL) algorithm.MethodsThe study was designed as a cross-sectional study. Routine breast ultrasound examinations were conducted by an experienced radiologist. The ultrasonic images of the lesions were retrospectively assessed by five residents-in-training according to the Breast Imaging Report and Data System (BI-RADS) lexicon, and a dichotomic classification of the lesions was provided by S-Detect. The diagnostic performances of S-Detect and the five residents were measured and compared using the pathological results as the gold standard. The category 4a lesions assessed by the residents were downgraded to possibly benign as classified by S-Detect. The diagnostic performance of the integrated results was compared with the original results of the residents.ParticipantsA total of 195 focal breast lesions were consecutively enrolled, including 82 malignant lesions and 113 benign lesions.ResultsS-Detect presented higher specificity (77.88%) and area under the curve (AUC) (0.82) than the residents (specificity: 19.47%–48.67%, AUC: 0.62–0.74). A total of 24, 31, 38, 32 and 42 identified as BI-RADS 4a lesions by residents 1, 2, 3, 4 and 5 were downgraded to possibly benign lesions by S-Detect, respectively. Among these downgraded lesions, 24, 28, 35, 30 and 40 lesions were proven to be pathologically benign, respectively. After combining the residents' results with the results of the software in category 4a lesions, the specificity and AUC of the five residents significantly improved (specificity: 46.02%–76.11%, AUC: 0.71–0.85, p<0.001). The intraclass correlation coefficient of the five residents also increased after integration (from 0.480 to 0.643).ConclusionsWith the help of the DL software, the specificity, overall diagnostic performance and interobserver agreement of the residents greatly improved. The software can be used as adjunctive tool for residents-in-training, downgrading 4a lesions to possibly benign and reducing unnecessary biopsies.

The Added Value of a Computer‐Aided Diagnosis System in Differential Diagnosis of Breast Lesions by Radiologists With Different Experience

“…With BI‐RADS 4a as the cutoff value for benign and malignant breast masses, S‐Detect software was more effective for less experienced radiologists and can significantly improve the diagnostic capabilities of less experienced radiologists 20,33 . Zhao et al explored the added value of CAD software for five less experienced radiologists and found that CAD software played an important role in reducing unnecessary biopsy, which was consistent with our study 21 . Therefore, S‐Detect can be considered as an auxiliary tool to improve diagnostic capabilities and reduce unnecessary biopsy of less experienced radiologists.…”

The Establishment of Transformer-Based Computer-Aided Diagnosis Model to Improve the Classification Consistency of BI-RADS-US3–5 Nodules Among Radiologists: A Multi-Center Study

“…[20] Similar studies have shown that young radiologists and radiologists lacking experience in breast ultrasound diagnosis have bene ted signi cantly from the use of a CAD model, especially for category 4a breast nodules, thereby minimizing unnecessary biopsies. [21,22] The radiologists participating in this study each had several years of experience and classi cation consistency among them reached values greater than 0.6 after referencing the CAD model. In addition, the sensitivity, speci city, and accuracy of classi cation improved with the use of CAD, informing adjustments to the initial diagnoses.…”

Artificial Intelligence System Reduces False-Positive Findings in the Interpretation of Breast Ultrasound Exams

“…Despite these limitations, we believe this study is a meaningful contribution to the emerging field of AI-based decision support systems for interpreting breast US exams. On a clinically realistic population, our AI system achieved a higher diagnostic accuracy (AUROC: 0.976, 95% CI: 0.972, 0.980) than prior AI systems for breast US lesion classification (AUROC: 0.82-0.96) [32,34,48,49,50,51,52], though we acknowledge these systems can be compared only approximately as they were evaluated on different datasets. Key features that contributed to our AI system's high level performance were the large dataset used in training, along with utilization of the weakly supervised learning paradigm that enables the system to learn from automatically extracted labels.…”