EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Optimizing the Mammography AI Pipeline: From Data Filtering to Vision-Language Models

Egor Ushakov, Sofya Zimina, Arsenii Litvinov, Sofia Senotrusova, Kirill Lukianov, Tigran G. Gevorkyan, Evgeny Karpulevich

Informatics · 31.07.2026 · 13 (8) · 125

DOI: 10.3390/informatics13080125 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Informatics, online erschienen am 31.07.2026. Verfasst von Egor Ushakov und 6 weiteren Autor:innen. Die Arbeit nennt 48 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Economic Development of the Russian Federation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The performance of AI solutions in mammography is largely determined by data quality, preprocessing methods, and augmentation strategies. However, systematic evaluation of these factors for models trained on aggregated multicenter datasets remains underexplored. This article presents a comparative assessment of the effects of different stages of the training pipeline on the final diagnostic accuracy. Using a pooled dataset (VinDr-Mammo, INBreast, CMMD, CBIS-DDSM), we evaluated each pipeline step—from filtering to architecture selection (EfficientNet-B3, CLIP). External testing was conducted on the MosMed database. Among the tested preprocessing steps, filtering the darkest 5% of images proved most effective. For EfficientNet-B3, optimal geometric and photometric augmentations increased test AUROC on the prepared MosMed test set from 0.844 to 0.900. Domain-specific pretraining and high resolution yielded the best performance: Mammo-CLIP achieved an AUROC of 0.949 ± 0.012, and EfficientNet-B3 reached 0.934 ± 0.013. Overall, this study developed a standardized pipeline that includes sequential data filtering and harmonization, augmentation optimization, and architecture selection. This approach ensures reliable and reproducible results for automated mammogram classification.

Förderung

  • Ministry of Economic Development of the Russian Federation

Zitieren

APA

Ushakov, E., Zimina, S., Litvinov, A., Senotrusova, S., Lukianov, K., Gevorkyan, T. G., & Karpulevich, E. (2026). Optimizing the Mammography AI Pipeline: From Data Filtering to Vision-Language Models. Informatics, 13(8), 125. https://doi.org/10.3390/informatics13080125

Vancouver

Ushakov E, Zimina S, Litvinov A, Senotrusova S, Lukianov K, Gevorkyan TG, et al. Optimizing the Mammography AI Pipeline: From Data Filtering to Vision-Language Models. Informatics. 2026;13(8):125. doi:10.3390/informatics13080125

BibTeX
@article{Ushakov2026,
  author = {Ushakov, Egor and Zimina, Sofya and Litvinov, Arsenii and Senotrusova, Sofia and Lukianov, Kirill and Gevorkyan, Tigran G. and Karpulevich, Evgeny},
  title = {Optimizing the Mammography AI Pipeline: From Data Filtering to Vision-Language Models},
  journal = {Informatics},
  year = {2026},
  volume = {13},
  number = {8},
  pages = {125},
  doi = {10.3390/informatics13080125}
}
Kostenlos starten →