Page 1 of 3

Health Data School

Health Data School is an intensive 5-day course focused on working with health data, biostatistics and artificial intelligence using the Python language. Through interactive workshops on real-world datasets, you’ll learn how to clean and merge messy tables, handle dates and time, explore patient-group differences, create regression and AI models, and communicate results clearly.

Location: Bratislava, LF UK Date: 9. - 13. March 2026 Duration: 5 full-time days (45 hours), 8:30 AM - 5:30 PM

Prerequisites: basic knowledge of Excel. Programming knowledge is not required.

The bootcamp is conducted in Slovak language.

Who is the course best for?

Health Analysts who want to transition from Excel to larger datasets, learn to model data and create visualisations.

PhD students and Researchers who work with data in their research and need to use statistics and regression models.

Healthcare Innovators who want to gain practical knowledge in the area of health data science and AI.

Medical Students keen to learn the basics of data processing and AI.

Course Structure

Day 1: Foundations of Health Data Analysis

Fundamentals of healthcare data

Intro to Python and health data analysis

Data analysis using dataframes

Data cleaning, grouping, transforming

Merging multiple datasets

Handling date and time data types

Day 2: Health Data Visualisation

Overview of health data sources

Visualisation theory

Intro to data visualisation in Python

Visual exploration of data

Visualising patient-group differences

Geographical visualisation, maps

Overview of plot types

Publication-ready plots

Day 3: Epidemiology and Biostatistics

Age standardisation

Intro to statistical epidemiology

Foundations of medical statistics

Hypothesis testing

Family-wise error, p-value adjustment

Logistic regression

Overview of data types

Clinical trials vs observational data

Day 4: Survival Models and AI

Visualising Kaplan-Meier curves

Cox proportional hazard models

AI from the ground up: theory and uses

Supervised learning Unsupervised learning

Sensitivity, specificity, ROC-AUC

Interpretation of model decision-making

Neural networks and deep learning

Integration of LLMs into Python

Day 5: Data Hackathon! In cooperation with our partners, we have prepared challenging data tasks using data from the healthcare and education sectors. The goal of the hackathon is for every participant to utilise their new data and programming skills directly in practice and at the same time, learn something new about important social topics. Several teams already came up with interesting findings in both tasks, which provided new insights for stakeholders.

Why Python? Python is considered the most suitable language for health data science and AI innovations today. It is clear, intuitive, and ideal even for beginners. Already on the first day, you will learn its fundamentals and see how to solve real data tasks with just a few lines of code.

Why Codebridge College? We understand that learning data skills isn’t easy. It requires a new way of thinking, patience and the right support. At Codebridge College, we know how overwhelming it can feel to switch from Excel to Python, or from raw tables to statistical models and AI. That’s why our bootcamps are designed to be practical, clear and guided by mentors who walk you through every step. Discussion is encouraged and no question is left unanswered.

Participants consistently describe our courses as the most practical training they’ve attended and with new skills they can apply immediately. -> Feedback on our bootcamps

Graduate Profile

After completing the Health Data School, you will be able to clean, transform, and analyse real-world healthcare datasets in Python. You will move beyond Excel with scalable workflows for large, messy data and document your work in clear, reproducible notebooks.

Participants can explore and communicate findings visually in a way that clinicians and scientific community understand. They can choose the right chart for the question, compare patient groups, highlight patterns and outliers, and build publication-ready plots. They can also create geographic maps (when location is relevant) and create plots showing both signal and uncertainty.

Participants can answer epidemiological and clinical questions with appropriate statistical tools. They can perform hypothesis tests, interpret p-values and confidence intervals, control for multiple comparisons, and use logistic regression to quantify associations and predict binary outcomes. They can also distinguish what can (and cannot) be concluded from observational data compared to clinical trials.

Participants can analyse time-to-event outcomes using Kaplan–Meier curves and Cox proportional hazards models, and build a practical foundation in AI. This includes supervised and unsupervised learning, performance metrics (sensitivity, specificity, ROC-AUC), interpretability, neural networks, and integrating LLM tools into Python workflows.

Lecturers

Imrich Berta A graduate of Applied Mathematics at the University of Cambridge, with experience in machine learning models for disease prediction and clinical data analysis. He currently works as a consultant for government institutions and start-ups. He actively mentors analysts and organises programming and data workshops for students. Imrich enjoys helping people who do not consider themselves "math types" understand and apply mathematical and statistical principles in practice.

Laura Johanesova A bioinformatician studying at the University of Vienna. Skills in shell scripting, R and Python are crucial for her research in regeneration. She designs and leads intensive, practically oriented training sessions that help analysts transition from Excel to Python or R and applied machine learning. Laura's work includes the creation of educational programmes, the development of data skills in science and various projects in the field of healthcare and biomedical data.

Price

Private sector 1 000 EUR | Public sector 900 EUR | Academia 800 EUR

Prices are without VAT.

Contact

Laura Johanesova [email protected]