Milena A. Keren

Milena A. Keren

Auto-generated profile, based on publicly available information

Milena A. Keren

Bakulev National Medical Research Center for Cardiovascular Surgery

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 February 2023 in Russian Journal of Cardiology
Verified
Journal Article
Open access
PDF file
EGE. Z. Golukhova
E. Z. Golukhova
MKM. A. Keren
M. A. Keren
TZT. V. Zavalikhina
T. V. Zavalikhina
NBNaida Bulaeva
Naida Bulaeva
DAD. S. Akatov
D. S. Akatov
DKD. A. Kolesnikov
D. A. Kolesnikov
Aim. To develop and evaluate the effectiveness of models for predicting mortality after coronary bypass surgery, obtained using machine learning analysis of preoperative data.Material and methods. As part of a cohort study, a retrospective prediction of in-hospital mortality after coronary artery bypass grafting (CABG) was performed in 2182 patients with stable coronary artery disease. Patients were divided into 2 following samples: learning (80%, n=1745) and training (20%, n=437). The initial ratio of surviving (n=2153) and deceased (n=29) patients in the total sample indicated a pronounced class imbalance, and therefore the resampling method was used in the training sample. Five machine learning (ML) algorithms were used to build predictive risk models: Logistic regression, Random Forrest, CatBoost, LightGBM, XGBoost. For each of these algorithms, cross-validation and hyperparameter search were performed on the training sample. As a result, five predictive models with the best parameters were obtained. The resulting predictive models were applied to the learning sample, after which their performance was compared in order to determine the most effective model.Results. Predictive models implemented on ensemble classifiers (CatBoost, LightGBM, XGBoost) showed better results compared to models based on logistic regression and random forest. The best quality metrics were obtained for CatBoost and LightGBM based models (Precision — 0,667, Recall — 0,333, F1-score — 0,444, ROC AUC — 0,666 for both models). There were following common high-ranking parameters for deciding on the outcome for both models: creatinine and blood glucose levels, left ventricular ejection fraction, age, critical stenosis (>70%) of carotid arteries and main lower limb arteries.Conclusion. Ensemble machine learning methods demonstrate higher predictive power compared to traditional methods such as logistic regression. The prognostic models obtained in the study for preoperative prediction of in-hospital mortality in patients referred for CABG can serve as a basis for developing systems to support medical decision-making in patients with coronary artery disease.

3 July 2023 in Annals of the Russian academy of medical sciences
Verified
Journal Article
Closed access
PDF file
EGElena Z. Golukhova
Elena Z. Golukhova
MKMilena A. Keren
Milena A. Keren
TZTatiana V. Zavalikhina
Tatiana V. Zavalikhina
NBNaida Bulaeva
Naida Bulaeva
DMDamir Marapov
Damir Marapov
АДАлшибая Михаил Дурмишханович
Алшибая Михаил Дурмишханович
Background. The importance of predictors of adverse outcomes after coronary bypass surgery needs to be regularly review in view of changing standards of practice. Predicting complications will help reduce current mortality and morbidity among patients who have undergone coronary bypass surgery (CABG). Aims prediction and determination of independent predictors of in-hospital mortality after CABG based on the results of preoperative examination. Methods. In a retrospective cohort single-center study, we analyzed the results of 2239 patients who underwent isolated CABG for the period from September 2018 to August 2021. The primary endpoint of the study was hospital death after CABG. Also, the frequency of: non-fatal myocardial infarction, non-fatal stroke, acute heart failure, postoperative rhythm and conduction disorders, acute renal failure, multiple organ failure, infectious complications was analyzed. Next, we searched for independent predictors and built a prognostic model for assessing the risk of death after CABG. Results. Results. Hospital mortality among patients was 1.34% (n = 30). Independent predictors of death after CABG were: ejection fraction (OR 0.91; 95% CI: 0.880.93; p 0.001), body mass index (OR 0.90; 95% CI: 0.820.99; p = 0.035), glomerular filtration rate (OR 0.97; 95% CI: 0.960.99; p = 0.005), repeated CABG (OR 9.3; 95% CI: 2.7431.24; p 0.001), critical atherosclerotic lesion of the arteries of the lower extremities (OR 4.5; 95% CI: 1.93310.268; p 0.001). The model for predicting the probability of death after CABG had high quality metrics: ROC-AUC 0.838 (95% CI: 0.7470.929; p 0.001), sensitivity 72.4%, specificity 84%. The threshold value of the logistic function P at the cut-off point, which corresponded to the highest value of the Yuden index, was 0.015. The outcome was predicted when the value of the logistic function P was higher than or equal to this value. Conclusion. We have developed a model for predicting in-hospital mortality after CABG based on independent linear predictors: left ventricular ejection fraction, body mass index, glomerular filtration rate, repeated CABG, critical atherosclerotic lesion of the main arteries of the lower extremities.

20 May 2025 in Russian Journal of Cardiology
Verified
Journal Article
Open access
PDF file
EGE. Z. Golukhova
E. Z. Golukhova
ISI. Yu. Sigaev
I. Yu. Sigaev
MKM. A. Keren
M. A. Keren
TZT. V. Zavalikhina
T. V. Zavalikhina
KPK. V. Petrosyan
K. V. Petrosyan
SAS. A. Avakova
S. A. Avakova
Aim. To assess the long-term all-cause mortality after percutaneous coronary intervention (PCI) and coronary artery bypass grafting (CABG) in patients with stable coronary artery disease (CAD) under various clinical and anatomical scenarios.Material and methods. This single-center cohort retrospective study assessed the outcomes of CABG and PCI with implantation of second-generation drugeluting stents in 4177 patients with stable CAD. Inhospital, 30-day and remote 5-year all-cause mortality (mean follow-up period — 38 months) was assessed. Also, the influence of the initial severity of CAD, the presence/absence of diabetes, myocardial contractility on remote all-cause mortality after myocardial revascularization was assessed.Results. Inhospital and 30-day risks of death in patients who underwent PCI and CABG, after comparing the initial clinical characteristics, did not differ significantly. In the long-term follow-up period, PCI compared with CABG was associated with an increased all-cause mortality in the main groups (PCI vs CABG: odds ratio (OR) 1,84, 95% confidence interval (CI) 1,30-2,62, p50% (OR 5,04, 95% CI 1,72-14,76, p=0,003), but not in patients with single-vessel disease (OR 2,084, 95% CI 0,9964,361, p=0,051). Diabetes in the main study groups did not affect the difference in mortality as follows: CABG had an advantage over PCI regardless of diabetes. However, CABG in patients with multivessel disease and diabetes, in contrast to patients without diabetes, led to a significant decrease in all-cause death risk (OR 2,29, 95% CI 1,173-4,47, p=0,015). Also, PCI compared with CABG was accompanied by an increase in the 5-year death risk in patients with an not reduced left ventricular ejection fraction (LVEF) >40% (OR 1,74, 95% CI 1,205-2,536, p=0,003), but not in patients with LVEF 50%, lesion of two or more coronary arteries), with concomitant diabetes in multivessel disease, and patients with non-reduced LVEF (>40%).

30 August 2022 in Kardiologiia
Verified
Journal Article
Closed access
PDF file
ISI. Yu. Sigaev
I. Yu. Sigaev
MKM. A. Keren
M. A. Keren
ISInessa Viktorovna Slivneva
Inessa Viktorovna Slivneva
ZSZ. D. Shonia
Z. D. Shonia
DMDamir Marapov
Damir Marapov
Aim To develop an algorithm for using ultrasonic flowmetry (USF) and epicardial ultrasonic scanning (EpiUSS) for intraoperative assessment of anatomic and functional viability of conduits. Material and methods  For viability assessment of 460 coronary grafts in 150 patients who were operated at the Bakulev National Medical Research Center for Cardiovascular Surgery (2018–2021 г.), markers of graft failure were analyzed using the USF and EpiUSS data confirmed by results of graft angiography. According to RОС analysis, the Qmean and PI values indicative of the graft failure were determined. A CHAID decision tree was developed for assessing the prognostic significance of the analyzed parameters. Based on this prognostic model, an algorithm was developed for intraoperative diagnosis of anatomic and functional graft viability during coronary bypass surgery. Results The Qmean ≤20.5 ml/min values were associated with an increased relative risk (RR) of detecting graft failure (RR, 8.2; 95 % confidence interval, CI, 4.4–15.2). The developed model shows a high accuracy of predicting the graft failure (AUC = 0.906±0.03). The RR of graft failure at PI ≥2.65 was 3.3 (95 % CI, 2.17–5.08). The prognostic model for PI (AUC = 0.745±0.042) was sufficiently accurate with respect of possible graft failure. Nodes of high and low risk for graft failure were determined in the developed decision tree. The obtained model was characterized by high sensitivity and specificity (100 and 84.3 %, respectively). Conclusion The combined use of USF and EpiUSS allows a highly accurate assessment of both morphological and functional characteristics of graft flow. The developed algorithm for the intraoperative diagnosis of anatomic and functional graft viability can be recommended for clinical use.


29 May 2026 in Kreativnaya kardiologiya
Verified
Journal Article
Closed access
КЯК.б. Яхяева
К.б. Яхяева
МКМ.а. Керен
М.а. Керен
ИСИ.ю. Сигаев
И.ю. Сигаев
ИВИ.в. Волковская
И.в. Волковская
MMMaxim Mamalyga
Maxim Mamalyga
ЕКЕ.а. Космачева
Е.а. Космачева