Breast cancer classification based on the integration of diagnostic algorithms for calcifications and masses using a mixture of experts
PLOS One · 2025
DOI: 10.1371/journal.pone.0331017 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, 2025. Verfasst von Yuma Konaka und 9 weiteren Autor:innen. Beteiligte Einrichtung: Department of Diagnostic Radiology, Tohoku University Graduate School of Medicine, 2-1 Seiryo-Machi, Aoba-ku, Sendai, Miyagi, Japan. Hauptthemen laut MEDLINE: Breast Neoplasms, Algorithms, Calcinosis. Die Arbeit nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Japan Science and Technology (JST) Agency, as core research for evolutional science and technology.
Abstract (Originalsprache)
Purpose
To investigate the effectiveness of an integrated deep-learning (DL) algorithm, the Mixture of Radiological Findings Specific Experts (MoRFSE) , in breast cancer classification by imitating the diagnostic decision-making process of radiologists.
Methods
A total of 2,764 mammographic images (1,462 breast cancer, 248 benign lesions, and 1,054 normal breast tissue) from the TOMPEI-CMMD were used. The MoRFSE comprises three DL models: a gate network for categorization (gNet) and two classification expert networks (cExp and mExp) specialized in capturing the distinct characteristics of calcifications and masses, respectively. This structure imitates radiologists’ comprehensive diagnostic process by applying distinct algorithms for different lesion types. The classification performance of MoRFSE was compared with the Conventional ResNet18 model. 5-fold cross-validation was used, and performance was assessed using the area under the receiver operating characteristic curve (AUC). DeLong’s test was performed to evaluate statistical significance.
Results
MoRFSE achieved a significantly higher AUC (0.9616) compared to Conventional ResNet18 (0.9577, p = 0.001348).
Conclusion
By integrating specialized algorithms for calcifications and masses, the MoRFSE model effectively emulates radiologists’ diagnostic process. These findings suggest that MoRFSE has the potential to improve the accuracy of breast cancer diagnosis based on mammograms.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Department of Diagnostic Radiology, Tohoku University Graduate School of Medicine, 2-1 Seiryo-Machi, Aoba-ku, Sendai, Miyagi, Japan.
- Department of Diagnostic Imaging, Tohoku University Graduate School of Medicine, 2-1 Seiryo-Machi, Aoba-ku, Sendai, Miyagi, Japan.
- AI Lab, Tohoku University Hospital, 1-1 Seiryo-machi, Aoba-ku, Sendai, Miyagi, Japan.
ORCID-Profile: Yuma Konaka, Ryusei Inamori, Eichi Takaya
Förderung
- Japan Science and Technology (JST) Agency, as core research for evolutional science and technology
Zitieren
APA
Konaka, Y., Ueda, T., Inamori, R., Sato, J., Sugawara, K., Shiratori, Y., Hario, F., Takaya, E., Kobayashi, T., & Okamoto, Y. (2025). Breast cancer classification based on the integration of diagnostic algorithms for calcifications and masses using a mixture of experts. PLOS One. https://doi.org/10.1371/journal.pone.0331017
Vancouver
Konaka Y, Ueda T, Inamori R, Sato J, Sugawara K, Shiratori Y, et al. Breast cancer classification based on the integration of diagnostic algorithms for calcifications and masses using a mixture of experts. PLOS One. 2025. doi:10.1371/journal.pone.0331017
BibTeX
@article{Konaka2025,
author = {Konaka, Yuma and Ueda, Takuya and Inamori, Ryusei and Sato, Jumpei and Sugawara, Keisuke and Shiratori, Yuta and Hario, Fumihito and Takaya, Eichi and Kobayashi, Tomoya and Okamoto, Yoshikazu},
title = {Breast cancer classification based on the integration of diagnostic algorithms for calcifications and masses using a mixture of experts},
journal = {PLOS One},
year = {2025},
doi = {10.1371/journal.pone.0331017}
}
EUVIMED