Back to the list
Congress: ECR24
Poster Number: C-14045
Type: EPOS Radiologist (scientific)
Authorblock: S. Kasai1, H. Tamori2, C. Kai1; 1Niigata city/JP, 2Yokohama city/JP
Disclosures:
Satoshi Kasai: Research/Grant Support: Konica Minolta Inc. Consultant: Konica Minolta Inc.
Hideaki Tamori: Nothing to disclose
Chiharu Kai: Nothing to disclose
Keywords: Breast, Computer applications, Oncology, CAD, Digital radiography, Mammography, CAD, Computer Applications-Detection, diagnosis, Cancer
Purpose This study aimed to identify patients who should be treated with early intervention by classifying molecular subtypes based on biological characteristics on mammograms.  Moreover, to ascertain the capability of subtype classification with high accuracy, the combination of feature extraction and identification was explored. Vision Transformer, a cutting-edge deep learning model, and the Gradient Boosting method, known for its efficacy in data classification, were employed for this purpose.
Read more Methods and materials <Background>Breast cancer ranks as the most prevalent cancer among women globally. However, early detection contributes to a favorable prognosis for this form of cancer.  According to Cancer Information Service by National cancer center in Japan, While the incidence of cancer in women is the highest among other cancers, the 5-year survival rate is as high as 92.3% (2009-2011), and this underscores the favorable prognosis associated with early detection and treatment of the disease.  The continuum from identifying breast cancer to...
Read more Results The discrimination accuracy on the hold-out test for classifying subtypes on mammograms was 69.1% (47/68), when four ViT features, four global features, age, and orientation were used.  The discrimination accuracy for each of the four subtypes was 64.7% (11/17) for Luminal A, 76.5% (13/17) for Luminal B, 64.7% (11/17) for HER2-enriched, and 70.1% (12/17) for triple-negative (Fig. 5).  Figures 6 to 9 illustrate instances of accurately classified cases within each category. Consequently, the application of a model combining deep learning...
Read more Conclusion It was suggested that subtypes could be classified from mammograms based on ViT and global features.
Read more References 1. Robertson S, et al: Re-testing of predictive biomarkers on surgical breast cancer specimens is clinically relevant. Breast Cancer Res Treat, 174(3), 795-805, 2019.2. Chen J, et al: Comparison of Core Needle Biopsy and Excision Specimens for the Accurate Evaluation of Breast Cancer Molecular Marker: a Report of 1003 Cases. Pathol Oncol Res, 23(4), 769-775, 2017.3. Nakamura S, et al: Subtype Classification of Breast Cancer based on MRI Image. The 33rd Annual Conference of the Japanese Society for Artificial Intelligence,...
Read more
GALLERY