I.G. Gataullin1, A.R. Savinova2
1Kazan State Medical Academy ― Branch Campus of the FSBEI FPE RMACPE MOH Russia, Kazan
2Tatarstan Cancer Center, Kazan
Savinova Aygul R. ― oncologist of the Department №9 of the Tatarstan Cancer Center
29 Sibirskiy tract, Kazan, 420029, Russian Federation, tel. (843) 233-86-79, e-mail: aigulkazan@mail.ru, ORCID ID: 0000-0001-7048-4125
Abstract. Ovarian cancer continues to be a major public health problem due to its high mortality and recurrence rate. Because of this, and especially during the past three decades, scientists have taken various steps to identify factors that contribute to development, progression, drug sensitivity, and overall survival.
In order to introduce the accumulated scientific data into clinical practice and to present them to patients in the simplest form, as well as to choose one or another method of treatment depending on the individual prognosis of patients, prognostic and predictive models calculating the probability of one or another event were developed.
The purpose of this study was to review existing data regarding prognostic and predictive models that predict the risk of ovarian cancer, as well as the effectiveness of chemotherapy or surgery.
As part of this review, models based on various statistical modeling methods were considered: logistic regression, artificial neural networks, mathematical modeling, Cox proportional hazards method, etc.
To ease perception of the material, existing models were grouped into the following groups: models predicting individual risk of ovarian cancer development, models for differential diagnosis of ovarian neoplasia, predictive models for treatment selection, and models predicting disease progression.
Key words: prediction of individual risk of ovarian cancer, predictive models of ovarian cancer, prognostic models of ovarian cancer, ovarian cancer recurrence.