Журнал «Современная Наука»

Russian (CIS)English (United Kingdom)
 MOSCOW +7(495)-142-86-81

INNOVATIVE APPROACHES TO NON-INVASIVE ENDOMETRIOSIS DIAGNOSIS: CONSTRUCTION AND SELECTION OF FACTORS USING MACHINE LEARNING METHODS IN CLINICAL PRACTICE

Rusinova Anastasia Konstantinovna  (FSBOU VSMU named after N.N. Burdenko Ministry of Health of Russia)

Endometriosis is a widely prevalent pathological condition; however, its clinical manifestations and underlying mechanisms remain insufficiently studied. The time gap between the onset of the first symptoms and the establishment of a diagnosis can sometimes exceed ten years, significantly hindering early diagnosis and adequate treatment. Currently, there is no universal treatment capable of completely eradicating endometriosis, underscoring the necessity for developing new diagnostic approaches. This study examines the construction and selection of factors associated with the risk of developing endometriosis using modern machine learning techniques to formulate an optimal mathematical model. An analysis of the significance of selected features was conducted, allowing for the reduction of the factor set to those that do not degrade the dynamic characteristics of the model, including accuracy, responsiveness, and stability. As a result, a risk prediction algorithm for endometriosis based on logistic regression was developed, incorporating 30 significant features. The effectiveness of the developed model was evaluated using standard metrics such as accuracy, sensitivity, specificity, F1-score, and the area under the ROC curve. The best results were achieved with an AUC value of 0.950, indicating a high predictive ability of the model.

Keywords:endometriosis, non-invasive diagnostics, machine learning, forecasting, logistic regression
 

Read the full article …

Citation link:
Rusinova A. K. INNOVATIVE APPROACHES TO NON-INVASIVE ENDOMETRIOSIS DIAGNOSIS: CONSTRUCTION AND SELECTION OF FACTORS USING MACHINE LEARNING METHODS IN CLINICAL PRACTICE // Современная наука: актуальные проблемы теории и практики. Серия: Естественные и Технические Науки. -2025. -№06/2. -С. 283-288 DOI 10.37882/2223-2966.2025.06-2.33
LEGAL INFORMATION:
Reproduction of materials is permitted only for non-commercial purposes with reference to the original publication. Protected by the laws of the Russian Federation. Any violations of the law are prosecuted.
© ООО "Научные технологии"

Terms Publications

Issues of the journals (Archive)

ECONOMICS AND LAW
HUMANITIES
NATURAL AND TECHNICAL SCIENCES
COGNITION

Серия - Natural and Technical Sciences