PROGNOSIS FOR ENDOMETRIAL HYPERPLASIA PROGRESSION IN PREMENOPAUSAL AND MENOPAUSAL WOMEN BASED ON THE ANALYSIS OF CELLULAR IMMUNITY INDICATORS USING MULTIPARAMETRIC NEURAL NETWORK CLUSTERING

Petro  Selskyy; Olena  Hladii; Svitlana  Heryak; Andrii  Sverstiuk; Andrii  Slyva; Anatolii  Televiak; Iryna  Parahnyuk; Tetyana  Golovata; Yurii  Orel; Tetiana  Adam

doi:10.21272/eumj.2025;13(3):674-682

Petro Selskyy Department of Pathologic Anatomy, Autopsy Course and Forensic Pathology, I. Horbachevsky Ternopil National Medical University, Ternopil, Ukraine https://orcid.org/0000-0001-9778-2499
Olena Hladii Department of Pathologic Anatomy, Autopsy Course and Forensic Pathology, I. Horbachevsky Ternopil National Medical University, Ternopil, Ukraine https://orcid.org/0000-0003-1800-9591
Svitlana Heryak Obstetrics and Gynecology Department #2, I. Horbachevsky Ternopil National Medical University, Ternopil, Ukraine https://orcid.org/0000-0002-7894-1009
Andrii Sverstiuk Department of Medical Informatics, I. Horbachevsky Ternopil National Medical University, Ternopil, Ukraine https://orcid.org/0000-0001-8644-0776
Andrii Slyva Department of Pathologic Anatomy, Autopsy Course and Forensic Pathology, I. Horbachevsky Ternopil National Medical University, Ternopil, Ukraine https://orcid.org/0000-0003-2385-4001
Anatolii Televiak Human Anatomy Department, I. Horbachevsky Ternopil National Medical University, Ternopil, Ukraine https://orcid.org/0000-0001-7173-400X
Iryna Parahnyuk MNE "Ternopil Regional Pathological Anatomical Bureau" of the Ternopil Regional Council, Ternopil, Ukraine
Tetyana Golovata Department of Pathologic Anatomy, Autopsy Course and Forensic Pathology, I. Horbachevsky Ternopil National Medical University, Ternopil, Ukraine https://orcid.org/0000-0001-9989-6510
Yurii Orel Department of Pathologic Anatomy, Autopsy Course and Forensic Pathology, I. Horbachevsky Ternopil National Medical University, Ternopil, Ukraine https://orcid.org/0000-0002-5871-5397
Tetiana Adam Obstetrics and Gynecology Department #2, I. Horbachevsky Ternopil National Medical University, Ternopil, Ukraine https://orcid.org/0009-0001-1002-7579

DOI: https://doi.org/10.21272/eumj.2025;13(3):674-682

Keywords: endometrial hyperplasia, cellular immunity, neural network clustering

Abstract

Many factors play a role in the progression of endometrial hyperplasia and the increased risk of malignant transformation. One of the important factors influencing pathological tissue remodeling is the immune response. However, changes in cellular immunity have not yet been systematized into specific patterns of immunological response in hyperplasia. Therefore, the implementation of easy-to-use and relatively inexpensive information technologies and risk factor analysis techniques is particularly important.

The objective of the study was to develop methods for predicting endometrial hyperplasia progression based on the analysis of morphological markers and indicators of cellular immunity using multiparametric neural network clustering.

Materials and Methods. The indicators of the cellular component of general immunity were determined in 43 pre- and menopausal women, of whom 31 patients were diagnosed with endometrial hyperplasia without atypia, and 12 women were otherwise healthy and formed the control group. For deeper analysis, we applied an approach based on multiparameter neural network clustering using NeuroXL Classifier for Microsoft Excel.

Results. In patients with endometrial hyperplasia, suppression of cellular immunity with a significant decrease in the percentage of all lymphocyte subpopulations was detected, whereas no significant changes in the immunoregulatory index were observed. It can indicate sufficient compensatory capabilities of the immune defense. The results of cluster analysis showed that in order to predict the progression of endometrial hyperplasia based on the analysis of the cellular immunity, it is important to consider the combination of reduced levels of CD3+ T-lymphocytes, CD4+ T-lymphocytes, and CD8+ T-lymphocytes, and increased levels of CD3+CD56+ NKT-like cells and CD56+ NK cells.

Conclusions. Neural network clustering was used to objectively classify patients into risk groups for progression of endometrial hyperplasia based on the results of clustering the studied indicators, which allows determining the significance of combined changes in certain parameters for disease progression prognosis.

Downloads

Download data is not yet available.

References

Batko Yu, Batryn N, Melnyk G, et al. Development of algorithms for biomedical image segmentation based on preliminary markup and texture attributes. Eastern-European Journal of Enterprise Technologies 2017; 90 (6/4): 35-44. https://doi.org/10.15587/1729-4061.2017.119299

Epstein RM, Franks P, Fiscella K, et al. Measuring patient-centered communication in patient-physician consultations: theoretical and practical issues. Soc Sci Med 2005; 61: 1516-1528.

Mintzer OP, Woronenko YV, Babintseva LY, et al. Health Informatization Concept in Ukraine. Medical Informatics and Engineering 2012; 3: 5-29. https://doi.org/10.1016/j.socscimed.2005.02.001

Hartzband P, Groopman J. Off the record − avoiding the pitfalls of going electronic. N Engl J Med 2008; 358: 1656-1658. https://doi.org/10.1056/nejmp0802221

Florczak B, Scheurich A, Croghan J, et al. An observational study to assess an electronic point-of-care wound documentation and reporting system regarding user satisfaction and potential for improved care. Ostomy Wound Manage 2012; 58(3): 46-51. https://pubmed.ncbi.nlm.nih.gov/22391956/

Nees LK, Heublein S, Steinmacher S, et al. Endometrial hyperplasia as a risk factor of endometrial cancer. Arch Gynecol Obstet 2022; 306(2): 407-421. https://doi.org/10.1007/s00404-021-06380-5

Karnezis AN, Leung S, Magrill J, et al. Evaluation of endometrial carcinoma prognostic immunohistochemistry markers in the context of molecular classification. J Pathol Clin Res 2017; 3(4):279–93. https://doi.org/10.1002/cjp2.82

Slyva A. Endometrial hyperplastic processes in perimenopausal women. Clinical and anamnestic analysis by using neural network clustering. Journal of Education, Health and Sport 2019; 9:1014–1024.

Alper E, Coskun A, Vural F. Comparison of nonspecific inflammatory markers in endometrial cancer and hyperplasia. Rev Assoc Med Bras 2021; 67 (7): 966-970 https://doi.org/10.1590/1806-9282.20210318

Vanderstraeten A, Tuyaerts S and Amant F. The immune system in the normal endometrium and implications for endometrial cancer development. J Reprod Immunol 2015; 109: 7–16. https://doi.org/10.1016/j.jri.2014.12.006

Versluis MA, Marchal S, Plat A, et al. The prognostic benefit of tumour-infiltrating Natural Killer cells in endometrial cancer is dependent on concurrent overexpression of Human Leucocyte Antigen-E in the tumour microenvironment. Eur J Cancer. 2017; 86: 285–295. https://doi.org/10.1016/j.ejca.2017.09.008

Zidan AA, Hassan AA, Seadah SS, et al. Selected immunohistochemical prognostic factors in endometrial hyperplasia versus carcinoma. Am J Med Sci 2015; 11(4):14–22. https://www.jofamericanscience.org/journals/am-sci/am110415/003_28215am110415_14_22.pdf

Sobczuk K., Sobczuk A. New classification system of endometrial hyperplasia WHO 2014 and its clinical implications. Prz Menopauzalny 2017; 16 (3): 107–111. https://doi.org/10.5114/pm.2017.70589

Travaglino A, Raffone A, Saccone G, M et al. Congruence Between 1994 WHO Classification of Endometrial Hyperplasia and Endometrial Intraepithelial Neoplasia System. Am J Clin Pathol 2020; 153(1):40-48. https://doi.org/10.1093/ajcp/aqz132

Boichuk AV, Shadrina VS, Vereshchahina TV. Hyperplasia of endometrium – a modern systemic-pathogenetic view of the problem (literature review). Actual Problems of Pediatrics, Obstetrics and Gynecology 2019; 1: 67-72.

Hapangama DK, Kamal АМ, Bulmer JN. Estrogen receptor b: the guardian of the endometrium. Hum Reprod Update 2015; (21):174–93. https://doi.org/10.1093/humupd/dmu053

Martsenyuk V. On an indirect method of exponential estimation for a neural network model with discretely distributed delays. Electronic Journal of Qualitative Theory of Differential Equations 2017; 23: 1-16. http://dx.doi.org/10.14232/ejqtde.2017.1.23

Selskyy PR, Sverstiuk AS, Slyva AF, Selskyy BP. Prediction of the progression of endometrial hyperplasia in women of premenopausal and menopausal age based on an analysis of clinical and anamnestic indicators using multiparametric neural network clustering. FM & PCR journal 2023. Vol. 25 (2). P. 184-189. https://doi.org/10.5114/fmpcr.2023.127679

Martsenyuk V. Indirect method of exponential convergence estimation for neural network with discrete and distributed delays. Electronic Journal of Differential Equations 2017; 246: 1-12. http://ejde.math.txstate.edu