Topic outline

  • General

    Welcome to the Decision Support in Healthcare course!

    This course is about understanding the needs, requirements, challenges and possibilities of data-analysis techniques for helping with decision making in healthcare. Decision making relates to, e.g., diagnosis of patients, treatment planning, and treatment follow-up (by doctors), but it can also relate to understanding risks or making decisions about lifestyle changes or disease management (for patients or people who want to stay healthy), or it can relate to making decision regarding investments in e.g. healthcare resources, legislations.

    The key idea is that we use different data sources, and process them  to extract essential information to help with those complex decisions. As technical tools we have many options, and the course will concentrate on them. They include signal processing, AI and machine learning, statistics, rule-based systems and many more. We will introduce their main properties and look at when and how to use what kind of technique.

    After successfully completing the course, you understand and can independently formulate data-analysis requirements associated with different healthcare decision making tasks (risk assessment, early diagnosis, differential diagnosis, treatment planning, and treatment follow-up). You 

    • Can recognize what kind of approaches (data-driven, rule-based, mechanistic models etc) are most appropriate for what decision-making challenge
    • Have knowledge of, and know how to select and apply methods for data curation and quality assurance
    • Have an understanding of the most common feature extraction and feature selection methods
    • Have gained knowledge of the most common AI/ML methods for advanced decision support
    • Understand how to objectively assess the performance of AI/ML methods in common healthcare decision support settings
    • Understand considerations such as explainability, privacy-preservation, bias and ethics in computer-based decision support in health and know which tools are available to address these.

    For more information, just ask. Using Mark's e-mail address at Tampere University (mark.vangils@tuni.fi) is the fastest way to get a response!