Informatics and AI in Medicine

Course guarantor:

Anna Jelínek Michaelidesová

Lecturer:

Anna Jelínek Michaelidesová

Tutor:

Anna Jelínek Michaelidesová

Supervisor:

Department of Dosimetry and Application of Ionizing Radiation

Synopsis:

The course introduces the fundamentals of artificial intelligence and its applications in medicine, with a particular focus on radiology and radiotherapy. It emphasizes understanding the clinical workflow, handling medical data, and principles of data processing. Students will gain a basic overview of machine learning methods and their use in medical image analysis, including related ethical and safety aspects.

Requirements:

During the semester, students complete five practical assignments focused on working with data and images in Python. The assignments cover data visualization and basic analysis, medical image processing, implementation of a simple machine learning model, model evaluation (metrics and interpretation), and a mini-project involving an AI or image processing application in medicine. The tasks are completed individually based on provided materials and are continuously discussed during practical sessions. Assessment is carried out throughout the semester and concludes with an individual oral examination, during which the student demonstrates understanding of the solution steps, applied methods, and obtained results. To obtain course credit (graded), students must submit all assignments and successfully pass the oral examination. Evaluation takes into account the quality of the assignments, depth of understanding, and the ability to critically discuss the results.

Syllabus of lectures:

1.AI in Medicine: Overview of AI applications in healthcare.

2.AI in the Clinical Workflow of Radiology: Data flow from image acquisition to diagnosis, AI role.

3.Fundamentals of Health Informatics: Types of medical data, data storage and transfer.

4.Data Security and Privacy in Healthcare: Encryption, anonymization, protection of personal data.

5.DICOM and PACS Systems: Structure of medical imaging data, image storage and sharing.

6.Digital Imaging and Basics of CT Reconstruction: Pixel, voxel, noise, basic reconstruction principles.

7.Medical Image Processing: Filtering, contrast enhancement, basic operations.

8.Fundamentals of Machine Learning: Model, training, model error.

9.Neural Networks and Convolutional Approaches: Principles of neural networks and their application in imaging.

10.Applications of AI in Radiology: Detection, classification, low-dose CT, reconstruction.

11.Applications of AI in Radiotherapy and Radiobiology: Segmentation, treatment planning, response prediction.

12.Model Validation, Bias, and Ethics of AI in Healthcare: Limitations of AI and safe use in clinical practice.

Syllabus of tutorials:

The practical sessions are conducted in Python using Jupyter Notebooks. Students work with prepared (anonymized) medical datasets, including DICOM data, and progressively modify and extend provided code. Emphasis is placed on practical handling of patient data in a clinical context, including secure data processing and interpretation of results.

Weeks 12: Basics of Python programming and data structures

Weeks 34: Loading, visualization, and quality assessment of medical imaging data (including artifacts)

Weeks 56: Working with DICOM metadata, data extraction, anonymization, and secure data handling

Weeks 78: Basic image processing techniques and introduction to segmentation

Weeks 910: Implementation and basic evaluation of a machine learning model (including performance metrics)

Week 11: Independent project work (data/image analysis and simple pipeline design)

Week 12: Project presentations, discussion of results, and interpretation in a clinical context

Study Objective:

The course provides students with a comprehensive introduction to health informatics and the fundamentals of artificial intelligence, with a focus on radiology, radiotherapy, and medical data processing. Emphasis is placed on understanding the structure and flow of medical data, the principles of digital imaging, and basic machine learning methods, as well as their practical application in a clinical setting. The course also includes introductory work with data in Python, enabling students to gain hands-on experience with image processing principles and simple models.

Study materials:

[1] Goodfellow, I.; Bengio, Y.; Courville, A. Deep Learning. MIT Press, 2016.

[2] Huang: PACS and Imaging Informatics: Basic Principles And Applications; John Wiley & Sons, 2009. ISBN 9780470373729.

[3] STEPHAN, Thompson (ed.). Artificial intelligence in medicine. Boca Raton: CRC Press, 2025. ISBN 9781032438344.

Note:

Time-table for winter semester 2026/2027:

Time-table is not available yet

Time-table for summer semester 2026/2027:

Time-table is not available yet

The course is a part of the following study plans:

• Radiologická technika (compulsory course in the program)