A. V. Samorodov1

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

Cytological analysis in oncology has a number of advantages: efficiency, high accuracy, low cost – however, its results are highly dependent on the experience of the cyto-pathologist. The issues of methodology for designing systems intended for automated cytological analysis are partially covered in the scientific and technical literature. But the degree of their development does not allow creating systems suitable for use in routine practice for solving most tasks of oncocytology, including those solved in the case of the most common women cancer – breast cancer.

The aim of the study is further development of issues of the biotechnical systems for automated microscopy of cytological preparations (BTS AMCP) design methodology, aimed primarily at ensuring the reliability of the results obtained.

Three levels of information transformation in BTS AMCP are described. At the technical level, the tasks of cells sampling and assessing their differential and group quantitative characteristics are solved; on the biological – the tasks of classifying cells, assessing and classifying laboratory parameters with the formation of laboratory signs; on the nosological – the tasks of assessing non-quantitative and quantitative indicators with the formation of a cytological conclusion. The reliability of the analysis results is determined by the quality of cytological preparation, of cells sampling, by algorithms for assessing quantitative characteristics, by classification errors, by the presence of biological variation and the uncertainty of the relationship between laboratory signs as well as group characteristics of cells with the patient's condition.

The identification and detailing of the main levels of information transformation in the BTS AMCP, as well as the analysis of the uncertainty propagation diagram, showed that the formation of the results of the biological level in an explicit form is a necessary condition for increasing the reliability of the results of automated analysis and ensuring the possibility of their rapid verification and pathomorphological interpretation.

Samorodov A.V. Aspects of the methodology for designing biotechnical systems for automated microscopy of cytological preparations. Biomedicine Radioengineering. 2025. V. 28. № 7. P. 33–41. DOI: https:// doi.org/10.18127/j15604136-202507-04 (In Russian)

1. Field A.S. et al. The International Academy of Cytology Yokohama system for reporting breast fine needle aspiration biopsy cytopathology. Acta Cytologica. 2019. V. 63. P. 257–273.
2. De Rosa F. et al. The continuing role of breast fine-needle aspiration biopsy after the introduction of the IAC Yokohama system for reporting breast fine needle aspiration biopsy cytopathology. Diagnostic Cytopathology. 2020. V. 48. Is. 12. P. 1244–1253.
3. Mitra S. et al. Cytology image analysis techniques towards automation: systematically revisited. DOI: 10.48550/arXiv.2003.07529. URL: https://arxiv.org/abs/2003.07529v1.
4. Field A.S., Raymond W.A., Schmitt F.C. The International Academy of Cytology Yokohama system for reporting breast fine needle aspiration biopsy cytopathology (Preface). Acta Cytologica. 2019. V. 63. P. 255–256.
5. Yin K. et al. Breast imaging, breast surgery, and cancer genetics in the age of COVID-19. Cancer. 2020. V. 126. Is. 20. P. 4466–4472.
6. Pinto D., Schmitt F. The role of breast FNA during and post- COVID-19 pandemic: a fast and safe alternative to needle core biopsy. Cytopathology. 2020. V. 31. № 6. P. 627–629.
7. Spiridonov I.N. Osobennosti proektirovaniya sistem deshifrirovaniya medicinskih izobrazhenij. Biomedicinskaya radioelektronika. 1999. № 6. S. 26–38 (In Russian).
8. Buttner J. Laboratory findings: structure, validity and significance for medical cognitive processes. Eur. J. Clin. Chem. Clin. Biochem. 1991. V. 29. № 8. P. 507–519.
9. Samorodov A.V. Biotekhnicheskie sistemy avtomatizirovannoj mikroskopii citologicheskih preparatov. Medicinskaya tekhnika. 2018. № 6. S. 17–19 (In Russian).
10. Semwal S. et al. Validation and modification of the Masood Scoring Index for the diagnosis of atypical breast lesions. Acta Cytologica. 2017. V. 61. № 2. P. 111–116.
11. Robinson I.A. et al. Prognostic value of cytologic grading of fine‐needle aspirates from breast carcinomas. Lancet. 1994. V. 343. P. 947–949.
12. Sinha A., Gill S.S. Correlative study of cytological features in grading of invasive breast carcinoma. Journal of Cytology. 2018. V. 35. Is. 3. P. 149–152.
13. Dobrolyubova D.A., Kravtsova T.A., Samorodova O.A., Samorodov A.V., Slavnova E.N., Volchenko N.N. Automatic image analysis algorithm for quantitative assessment of breast cancer estrogen receptor status in immunocytochemistry. Pattern Recognition and Image Analysis. 2016. V. 26. № 3. P. 552–557.
14. Cuzick J. et al. Prognostic value of a combined estrogen receptor, progesterone receptor, Ki-67, and human epidermal growth factor receptor 2 immunohistochemical score and comparison with the Genomic Health recurrence score in early breast cancer. Journal of Clinical Oncology. 2011. V. 29. № 32. P. 4273–4278.
15. Samorodov A.V. Building intelligent systems for the analysis of microscopic images in medicine and biology. Pattern Recognition and Image Analysis. 2013. V. 23. № 4. P. 508–511.
16. Schuh F. et al. Histopathological grading of breast ductal carcinoma in situ: validation of a web-based survey through intra-observer reproducibility analysis. Diagnostic Pathology. 2015. V. 10. Article № 93. 10 p.
17. Huang J., Wang D., Da J. Automated classification of cancer from fine needle aspiration cytological image use neural networks: A meta-analysis. Diagnostic Cytopathology. 2020. V. 48. Is. 11. P. 1027–1033.
18. Artyuhova O.A., Samorodov A.V. Sravnitel'noe issledovanie harakteristik rezkosti mikroskopicheskih izobrazhenij mediko-biologicheskih preparatov. Medicinskaya tekhnika. 2011. № 1. S. 15–22 (In Russian).
19. Mahov D.S., Sagatelyan G.R., Samorodov A.V. Cvetovaya mera dlya kalibrovki sistem cifrovoj mikroskopii. Izmeritel'naya tekhnika. 2021. № 10. C. 60–64 (In Russian).
20. Samorodov A.V. Formirovanie trebovanij k razmeru vyborki kletok v zadachah citologicheskoj diagnostiki raka molochnoj zhelezy. Biomedicinskaya radioelektronika. 2023. T. 26. № 1. S. 27–34 (In Russian).
21. Guan Z. et al. Combining breast cancer risk prediction models. Cancers. 2023. V. 15. Art. № 1090. https://doi.org/10.3390/cancers15041090.
22. Khan S. et al. A novel deep learning based framework for the detection and classification of breast cancer using transfer learning. Pattern Recognition Letters. 2019. V. 125. P. 1–6.
23. Subbaiah R.M., Dey P., Nijhawan R. Artificial neural network in breast lesions from fine-needle aspiration cytology smear. Diagnostic Cytopathology. 2014. V. 42. Is. 3. P. 218–224.
24. Nikitaev V.G. et al. Methods and means of diagnostics of oncological diseases on the basis of pattern recognition: intelligent morphological systems – problems and solutions. Journal of Physics: Conference Series. 2017. V. 798. Is. 1. Art. № 012131.