D.S. Makhov1, A.V. Samorodov2

1,2 Bauman Moscow State Technical University (Moscow, Russia)
1 dennismak@yandex.ru, 2 avs@bmstu.ru

Fluorescence in situ hybridization is required in cases of equivocal immunohistochemical assessment. Visual analysis of such images is labor-intensive and subjective due to specimen heterogeneity, as well as technical and biological artifacts. To ensure reproducibility of results and reduce the physician’s workload, there is a need for an interpretable automated algorithm based on guidelines for manual analysis. The goal of this study is to investigate the specific features of applying manual analysis guidelines in automated clinical decision support using binary segmentation of nuclei and signals in fluorescence in situ hybridization images, and to assess the influence of additional features of the segmented regions on algorithm performance. In a dataset of 155 images from 44 patients, nuclei and signal segmentation was performed using pretrained deep learning models. It was shown that the basic adaptation of manual analysis guidelines provides agreement with expert assessment at the level of κ = 0.36. The use of a weighted signal ratio and optimization of the threshold value increased agreement to κ = 0.49. The use of a meta-classifier based on the number, area, and intensity features of the segmented objects further increased agreement to κ = 0.58. The proposed approach makes it possible to improve the objectivity and reproducibility of fluorescence in situ hybridization image interpretation in breast cancer and to reduce the labor intensity of the analysis for the physician.

Makhov D.S., Samorodov A.V. Automated clinical decision support for determining HER2 gene amplification from FISH images in breast cancer // Biomedicine Radioengineering. 2026. V. 29. № 4. P. 18–22. DOI: https:// doi.org/10.18127/j15604136-202604-02

1. Tsang J.Y.S., Tse G.M. Molecular classification of breast cancer. Advances in Anatomic Pathology. 2020. V. 27. № 1. P. 27–35.
2. Volchenko N.N. i dr. Sovremennaya citologicheskaya diagnostika zabolevanij molochnoj zhelezy. M.: MGTU im. Baumana, 2014. 198 s. (In Russian)
3. Ly H. et al. Response to anti-HER2 neoadjuvant chemotherapy in HER2-positive invasive breast cancers with different HER2 FISH patterns. Journal of Clinical Pathology. 2025. V. 78. № 8. P. 540–547.
4. Nitta H., Kelly B.D. et al. The assessment of HER2 status in breast cancer: the past, the present, and the future. Pathology International. 2016. V. 66. № 6. P. 313–324.
5. Wolff A.C. et al. Epidermal Growth Factor Receptor 2 Testing in Breast Cancer. Archives of Pathology & Laboratory Medicine. 2023. V. 147. № 9. P. 993–1000.
6. Viale G. et al. Assessment of HER2 amplification status in breast cancer using a new automated HER2 IQFISH pharmDx™ (Dako Omnis) assay. Pathology – Research and Practice. 2016. V. 212. № 8. P. 735–742.
7. Zakrzewski F. et al. Automated detection of the HER2 gene amplification status in fluorescence in situ hybridization images for the diagnostics of cancer tissues. Scientific Reports. 2019. V. 9. № 1. Art. 8231. DOI: 10.1038/s41598-019-44821-4.
8. Mahov D.S., Razmahaev G.S., Slavnova E.N., Samorodov A.V. Razrabotka biotekhnicheskoj sistemy avtomatizirovannogo opredeleniya HER2-statusa pri rake molochnoj zhelezy metodom fluorescentnoj in situ gibridizacii (FISH) . Biomedicinskaya radioelektronika. 2022. T. 25. № 5. S. 58–69 (In Russian).
9. Ronneberger O., Fischer P., Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation. arXiv. 2015. Art. arXiv:1505.04597.
10. van der Logt E. M. J. et al. Fully automated fluorescent in situ hybridization (FISH) staining and digital analysis of HER2 in breast cancer: a validation study. PloS one. 2015. V. 10. № 4. Art. e0123201.
11. Xue T. et al. Deep learning to automatically evaluate HER2 gene amplification status from fluorescence in situ hybridization images. Scientific Reports. 2023. V. 13. № 1. Art. 9746.