Lab

Premedix Academy - Precision Medicine

About the lab

A new era in the evolution of medicine is approaching - the era of precision medicine. It combines the newest scientific knowledge and technology to maximally tailor preventive and therapeutic strategy. Our mission is to contribute to this exciting field by high quality education, intensive research and innovative technology development.

Main research areas:
Cardiovascular System, Thrombosis, Atrial Fibrillation, Biomarkers, MicroRNA, Artificial Intelligence, Telemedicine

Main projects:
AFISBIO I & II
APAF ESUS
BIOGUARD
STOP SHOCK

For more information about our projects, please see the Featured projects section below.

Featured projects (8)

Project
Premedix Academy provides research and technological background for the sister organization Premedix Clinic. Premedix Academy is testing new digital medical devices and developing a tele-health infrastructure for patients. Furthermore, we are doing intensive research in genomics to improve genotyping reports and personalize the preventive and therapeutic strategies delivered in Premedix Clinic.
Project
I. Identify novel plasmatic biomarkers associated with prevalent/incident atrial fibrillation in patients with high risk for AF and stroke; II. Assess predictive ability of novel plasmatic biomarkers (especially Apelin and miRNAs) on prevalent/incident atrial fibrillation in patients with high risk for AF and stroke; III. Validate predictive models from previous studies based on comorbidities, age, sex, BMI, NT-proBNP, FGF-23, IGF-1 and IGFBP-1 on prevalent/incident AF in patients with high risk for AF and stroke.
Project
Main goal of the project is to study the ability of plasma apelin to predict atrial fibrillation (AF) in patients after an embolic stroke of unknown source (ESUS) who will be monitored using an implantable loop recorder. Primary goal is to determine predictive value, specificity and sensitivity of apelin in the diagnosis of AF in patients after ESUS. Secondary goal is to develop a scoring system based on patient history and plasma apelin levels in order to more accurately determine the risk of AF after ESUS.
Project
The main goal of this project is to create a smartphone application which, on the basis of a simple measurement, will enable the identification of high-risk patients who can then benefit from targeted diagnostic and therapy process. The cornerstones of this project are the use of machine learning and unique algorithm for pulse wave photoplethysmographic detection, using only a smartphone camera. BioGuard is being developed in cooperation with Dutch company Happitech, creators of currently world’s first CE Certified heart rhythm Software Development Kit (SDK).
Project
The aim of this study is to develop a predictive model for Postoperative Atrial Fibrillation (POAF) based on machine learning algorithms and to compare its predictive power with existing scoring systems on a large population of patients post cardiac surgery.

Featured research (13)

Purpose Oxidative stress is an important contributor to the etiology of atrial fibrillation (AF). Our aim was to study oxidative stress biomarkers in patients undergoing pulmonary vein isolation (PVI) for paroxysmal AF with radiofrequency catheter ablation and to assess its prognostic value in predicting long-term PVI outcome. Methods In this prospective cohort study, we included 62 patients (mean age 55±8 years, 12 females and 50 males) with paroxysmal AF and implanted ECG loop recorders who underwent PVI. Plasmatic concentrations of advanced glycation end-products (AGEs), fructosamine, advanced oxidation protein products, and thiobarbituric-acid reacting substances were measured before PVI. AF burden (percentage of time spent in AF) was continually assessed during the follow-up period (1063±271 days). Results Nineteen patients (31%) were defined as optimal responders (oR) with AF burden < 0.5% after PVI. Remaining 43 patients (69%) were defined as sub-optimal responders. Concentration of AGEs was significantly lower in oR by 3.7 g/g (CI: −6.5 to −1.7; P=0.0003). After adjustment for age, sex, BMI, left atrial size, arterial hypertension, and AF burden before PVI, only low concentration of AGEs remained significantly associated with oR (odds ratio: 1.3; P=0.04). AGEs concentration achieved area under the curve of 0.78 for predicting optimal long-term PVI response. Conclusions AGEs concentration before PVI was associated with long-term PVI outcome in patients with paroxysmal AF. Further research will show if this biomarker could contribute to optimal patient selection for catheter ablation.
Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Ministry of Education, Science, Research and Sport of the Slovak Republic VEGA Background Cardiogenic shock (CS) is a serious life-threatening condition affecting almost 10% of patients suffering from acute coronary syndrome (ACS). Despite recent treatment improvements such as mechanical circulatory support devices (MCS), mortality remains 50%. It is hypothesized that early implantation of MCS before hemodynamic deterioration could prevent CS. Therefore, knowing which ACS patient will progress into CS would be of paramount importance. Purpose We aimed to develop a model based on machine learning algorithms for CS prediction in patients with ACS. Method Over 40 000 patients from critical care units of the Beth Israel Deaconess Medical Center database were extensively analyzed and sorted. Patients suffering from acute coronary syndrome undergoing cardiac catheterizations were selected and divided into two groups based on the development of CS. Patients in CS at the time of admission were excluded. The study population consisted of 3056 patients who didn’t develop CS and of 176 patients who did develop CS. Important information was also extracted manually from textual summaries of hospital stays. Potentially relevant clinical variables for shock prediction were selected using supervised feature selection, and missing values were supplemented using imputation methods. Seven well-known and established machine learning algorithms were used. Based on preliminary evaluation of classifier performance (AUC on random train-test split with 30% test data), we selected two best-performing algorithms: Logistic Regression and Gaussian Process classifier with Radial Basis Function (RBF) kernel. Both models were subsequently validated using Repeated Stratified K-Fold cross-validation with 5 folds and 20 repeats each. Results Age, heart rate, mean arterial pressure, respiratory rate, oxygen flow (liters per minute), peripheral oxygen saturation, blood glucose, pain type, heart rhythm and ectopy frequency were chosen as input variables. Both models showed good discrimination. The Logistic Regression model scored AUC of 0.76 ± 0.04. The Gaussian Process classifier with Radial Basis Function (RBF) kernel scored AUC of 0.77 ± 0.03. Conclusion According to our knowledge this is the first study that uses machine learning algorithms to predict CS in patients with ACS based on easily obtainable clinical variables. Further explorations and refinement of used imputation, feature selection, machine learning techniques and validation on an external cohort might result in even better performance of the proposed prediction models. Furthermore, these prediction models could be transformed into a simple predictive scoring system available in clinical practice.
Artificial intelligence (AI) is here to stay. It is not a future anymore, and there are many particular problems in cardiology that are already being solved via machine learning (ML), and many more are to come. AI cannot solve complex tasks yet, and probably this will not change in the upcoming years. Therefore, cardiologists do not have to be afraid that computers will replace them. However, cardiologists who will not be able to use ML algorithms in their clinical practice will be replaced by those who will. (Fig. 2, Ref. 50). Keywords: artificial intelligence, cardiology, potential machine learning, survival models, classification algorithms, computer vision, automated analysis of various imaging examinations, ECG interpretation, phenotype clustering, pathophysiological mechanisms.
Background: Catecholamines are recommended as first-line drugs to treat hemodynamic instability after out-of-hospital cardiac arrest (OHCA). The benefit-to-risk ratio of catecholamines is dose dependent, however, their effect on metabolism and organ function early after OHCA has not been investigated. Methods: The Post-Cardiac Arrest Syndrome (PCAS) pilot study was a prospective, observational, multicenter study. The primary outcomes of this analysis were association between norepinephrine/cumulative catecholamines doses and neuron specific enolase (NSE)/lactate concentration over the first 72 hours after resuscitation. The association was adjusted for proven OHCA mortality predictors and verified with propensity score matching (PSM). Results: Overall 148 consecutive OHCA patients; aged 18–91 (62.9 ± 15.27), 41 (27.7%) being female, were included. Increasing norepinephrine and cumulative catecholamines doses were significantly associated with higher NSE concentration on admission (r = 0.477, p < 0.001; r = 0.418, p < 0.001) and at 24 hours after OHCA (r = 0.339, p < 0.01; r = 0.441, p < 0.001) as well as with higher lactate concentration on admission (r = 0.404, p < 0.001; r = 0.280, p < 0.01), at 24 hours (r = 0.476, p < 0.00; r = 0.487, p < 0.001) and 48 hours (r = 0.433, p < 0.01; r = 0.318, p = 0.01) after OHCA. The associations remained significant up to 48 hours in non-survivors after PSM. Conclusions: Increasing the dose of catecholamines is associated with higher lactate and NSE concentration, which may suggest their importance for tissue oxygen delivery, anaerobic metabolism, and organ function early after OHCA.

Lab head

Allan Böhm
Department
  • Acute cardiology

Members (7)

Ljuba Bacharova
  • International Laser Centre
Nikola Jajcay
  • Technische Universität Berlin
Marianna Vachalcová
  • Vychodoslovensky ústav srdcových a cievnych chorôb, a.s
Marta Kollárová
  • Premedix Academy
Katarina Petrikova
  • Premedix Academy - Precision Medicine
Tomáš Uher
  • Comenius University Bratislava
Juliana Haráková
  • University of Trnava
Peter Michalek
Peter Michalek
  • Not confirmed yet
Peter Michalek
Peter Michalek
  • Not confirmed yet