Abstract

ObjectiveComputer-aided diagnosis using deep learning algorithms has been initially applied in the field of mammography, but there is no large-scale clinical application.MethodsThis study proposed to develop and verify an artificial intelligence model based on mammography. Firstly, mammograms retrospectively collected from six centers were randomized to a training dataset and a validation dataset for establishing the model. Secondly, the model was tested by comparing 12 radiologists’ performance with and without it. Finally, prospectively enrolled women with mammograms from six centers were diagnosed by radiologists with the model. The detection and diagnostic capabilities were evaluated using the free-response receiver operating characteristic (FROC) curve and ROC curve.ResultsThe sensitivity of model for detecting lesions after matching was 0.908 for false positive rate of 0.25 in unilateral images. The area under ROC curve (AUC) to distinguish the benign lesions from malignant lesions was 0.855 [95% confidence interval (95% CI): 0.830, 0.880]. The performance of 12 radiologists with the model was higher than that of radiologists alone (AUC: 0.852 vs. 0.805, P=0.005). The mean reading time of with the model was shorter than that of reading alone (80.18 s vs. 62.28 s, P=0.032). In prospective application, the sensitivity of detection reached 0.887 at false positive rate of 0.25; the AUC of radiologists with the model was 0.983 (95% CI: 0.978, 0.988), with sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of 94.36%, 98.07%, 87.76%, and 99.09%, respectively. ConclusionsThe artificial intelligence model exhibits high accuracy for detecting and diagnosing breast lesions, improves diagnostic accuracy and saves time.

Highlights

  • Computer-aided diagnosis using deep learning algorithms has been initially applied in the field of mammography, but there is no large-scale clinical application

  • Several imaging methods are used to identify suspicious malignant breast lesions, while mammography is the only screening method that has been proved to reduce the mortality of breast cancer [3,4], which can reduce the risk of breast cancer death up to 40% [5,6]

  • Of the remaining 4,367 participants assessed for quality control, 97 (2.22%) were excluded due to poor quality of mammography and 151 (3.46%) were excluded due to inconsistency in anatomical location and pathological report

Read more

Summary

Introduction

Computer-aided diagnosis using deep learning algorithms has been initially applied in the field of mammography, but there is no large-scale clinical application. Breast cancer is the most common malignant tumor in women [1], and the leading cause of cancer death in women worldwide. Several imaging methods are used to identify suspicious malignant breast lesions, while mammography is the only screening method that has been proved to reduce the mortality of breast cancer [3,4], which can reduce the risk of breast cancer death up to 40% [5,6]. The large number of breast cancer screening population results in heavy mammography load, and uneven distribution of breast specialists makes difference in the level of mammography diagnosis. A number of studies have pointed out that about 75% of breast biopsies caused by suspicious mammography results are confirmed as benign changes [7]. It is highly essential to effectively and accurately detect breast lesions

Objectives
Methods
Results
Discussion
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call