Cancer Research and Treatment

Original Articles

Breast cancer

Molecular Classification of Breast Cancer Using Weakly Supervised Learning: Wooyoung Jang, Jonghyun Lee, Kyong Hwa Park, Aeree Kim, Sung Hak Lee, Sangjeong Ahn; Cancer Res Treat. 2025;57(1):116-125. Published online June 25, 2024; DOI: https://doi.org/10.4143/crt.2024.113

Abstract PDF Supplementary Material PubReader ePub: Purpose
The molecular classification of breast cancer is crucial for effective treatment. The emergence of digital pathology has ushered in a new era in which weakly supervised learning leveraging whole-slide images has gained prominence in developing deep learning models because this approach alleviates the need for extensive manual annotation. Weakly supervised learning was employed to classify the molecular subtypes of breast cancer.
Materials and Methods
Our approach capitalizes on two whole-slide image datasets: one consisting of breast cancer cases from the Korea University Guro Hospital (KG) and the other originating from The Cancer Genomic Atlas dataset (TCGA). Furthermore, we visualized the inferred results using an attention-based heat map and reviewed the histomorphological features of the most attentive patches.
Results
The KG+TCGA-trained model achieved an area under the receiver operating characteristics value of 0.749. An inherent challenge lies in the imbalance among subtypes. Additionally, discrepancies between the two datasets resulted in different molecular subtype proportions. To mitigate this imbalance, we merged the two datasets, and the resulting model exhibited improved performance. The attentive patches correlated well with widely recognized histomorphologic features. The triple-negative subtype has a high incidence of high-grade nuclei, tumor necrosis, and intratumoral tumor-infiltrating lymphocytes. The luminal A subtype showed a high incidence of collagen fibers.
Conclusion
The artificial intelligence (AI) model based on weakly supervised learning showed promising performance. A review of the most attentive patches provided insights into the predictions of the AI model. AI models can become invaluable screening tools that reduce costs and workloads in practice.

2,026 View
173 Download

Quantitation of AgNORs in Breast Lesions Using Image Analysis: In Sun Kim, Young Sik Kim, Aeree Kim; J Korean Cancer Assoc. 1994;26(5):756-764.

Abstract PDF: The argyrophilic nucleolar organizer regions(AgNORs) staining is one of the useful methods for the estimation of proliferative activity in conventional histologic sections in various benign and malignant lesions. Because of difficulties in AgNORs counting, there is a relucance to accept the technique as a reliable diagnostic tooL We measured the area and numbers of ARNORs in 22 surgically resected infiltrating ductal carcinomas and 5 fibroadenomas using image analyzer and also visually enumerated the numbers at x 1000 magnification. The number of AgNORs by visual conuting was correlated with the area measured by Image counting(r=0.56). The mean area of AgNORs in the malignant group(4.36um(2)) was significantly different(p<0.0001) from that of the benikn group(0.79pm). As the number of AgNORs increased, the area increased(r=0.7091 The result of this study suggests that measurement using image analyzer can give more objective and reproduccible measurement of ARNORs than visual counting.

1,862 View
12 Download

First
Prev
Page of 1
Next
Last

Search