Department of Radiology, Far Eastern Memorial Hospital, No. 21, Sec. 2, Nanya S. Rd., Banciao Dist., New Taipei City, 220, Taiwan.
Department of Radiology, Taipei Veterans General Hospital, No. 201, Sec. 2, Shipai Rd., Beitou Dist., Taipei City, 112, Taiwan.
Biomed Eng Online. 2023 May 27;22(1):54. doi: 10.1186/s12938-023-01115-w.
Use of an AI system based on deep learning to investigate whether the system can aid in distinguishing malignant from benign calcifications on spot magnification mammograms, thus potentially reducing unnecessary biopsies.
In this retrospective study, we included public and in-house datasets with annotations for the calcifications on both craniocaudal and mediolateral oblique vies, or both craniocaudal and mediolateral views of each case of mammograms. All the lesions had pathological results for correlation. Our system comprised an algorithm based on You Only Look Once (YOLO) named adaptive multiscale decision fusion module. The algorithm was pre-trained on a public dataset, Curated Breast Imaging Subset of Digital Database for Screening Mammography (CBIS-DDSM), then re-trained and tested on the in-house dataset of spot magnification mammograms. The performance of the system was investigated by receiver operating characteristic (ROC) analysis.
We included 1872 images from 753 calcification cases (414 benign and 339 malignant) from CBIS-DDSM. From the in-house dataset, 636 cases (432 benign and 204 malignant) with 1269 spot magnification mammograms were included, with all lesions being recommended for biopsy by radiologists. The area under the ROC curve for our system on the in-house testing dataset was 0.888 (95% CI 0.868-0.908), with a sensitivity of 88.4% (95% CI 86.9-8.99%), specificity of 80.8% (95% CI 77.6-84%), and an accuracy of 84.6% (95% CI 81.8-87.4%) at the optimal cutoff value. Using the system with two views of spot magnification mammograms, 80.8% benign biopsies could be avoided.
The AI system showed good accuracy for classification of calcifications on spot magnification mammograms which were all categorized as suspicious by radiologists, thereby potentially reducing unnecessary biopsies.
利用基于深度学习的人工智能系统，研究该系统是否能辅助鉴别乳腺 X 线摄影术放大点片上的良恶性钙化，从而可能减少不必要的活检。
本回顾性研究纳入了标注有头尾位和内外斜位或头尾位和内外斜位每个乳腺 X 线摄影图像上钙化的公共数据集和内部数据集。所有病变均有病理结果进行相关性分析。我们的系统由一个名为自适应多尺度决策融合模块的基于 You Only Look Once（YOLO）的算法组成。该算法在公共数据集 Curated Breast Imaging Subset of Digital Database for Screening Mammography（CBIS-DDSM）上进行了预训练，然后在内部放大点片乳腺 X 线摄影数据集上进行了重新训练和测试。通过受试者工作特征（ROC）分析来研究系统的性能。
我们纳入了来自 CBIS-DDSM 的 753 例钙化病例（414 例良性，339 例恶性）的 1872 张图像。从内部数据集中，纳入了 636 例（432 例良性，204 例恶性）1269 张放大点片乳腺 X 线摄影图像，所有病变均由放射科医生推荐活检。我们的系统在内部测试数据集上的 ROC 曲线下面积为 0.888（95%CI 0.868-0.908），灵敏度为 88.4%（95%CI 86.9-8.99%），特异性为 80.8%（95%CI 77.6-84%），最佳截断值下的准确性为 84.6%（95%CI 81.8-87.4%）。使用该系统对放大点片乳腺 X 线摄影术的两个视图进行分析，可避免 80.8%的良性活检。
该人工智能系统对放射科医生均归类为可疑的放大点片乳腺 X 线摄影术上的钙化分类具有较好的准确性，从而可能减少不必要的活检。
