Did you ever wonder how you can improve your data management and handling? Have you ever hoped for a clean and indisputable database that you could easily share with your collaborators? Understand the art of Data Stewardship and get a handle on your data! Data could yield great value when processed intelligently for medical data science but holds great risks when processing is lost in complexity. This online Data Stewardship course will guide you to steward your clinical data to aid data science developments, facilitate collaborative research, comply with privacy regulations, and ensure data integrity and quality. Together we will put data stewardship in practice with clinical decision-making tools, predictive models, and Artificial Intelligence (AI).
Clinical-decision support aims at advancing care to achieve preventive, predictive, personalized, and participatory (P4) medicine. Data scientists can aid physicians by developing artificial intelligence techniques, but insight from clinical practice is needed to steward care data for data science and research purposes.
In this course we will take you along the journey of a data steward in medical science. You will experience the variety of problems that can arise upon unstewarded data and you will discover the tools to prevent but also clean-up and validate your data. You will tackle the hurdles in clinically relevant examples and delve into making your data FAIR: Findable, Accessible, Interoperable, and Reusable. Also, you will learn how to handle your data, from personal to anonymous, and become familiar with General Data Protection Regulation (GDPR) and Confidentiality, Integrity, and Accessibility (CIA), to assess legal and privacy concerns. Together we will put data stewardship in practice with clinical decision-making tools, predictive models, and AI.
Target audience
Healthcare professionals
Aim of the course
During this course, you will develop skills regarding Communication, Scientific reporting, Research, and Implementation. By the end of the course, you should be able to:
- Discuss the issues of retrospective and prospective data verification with both domain experts and data experts.
- Explain data research approaches to the data scientist.
- Formulate questions and issues on data with data scientists.
- Use basic principles of scientific medical background when reporting about data science results.
- Use basic principles of the clinical decision-making process at the individual patient level when reporting about personalized medicine.
- Use basic principles of the limitations of the model with respect to medical, clinical, technical, and model exceptions when reporting about data science models.
- Make a technical file for CE marking that explains basic principles of reporting.
- Design data validation rules compliant to GDPR to limit data entry for analytics.
- Retrospectively check data usefulness for research by verifarting parameter values at both intra and inter-parameter levels dependent on a specific domain.
- Translates care data to high quality structured data for research.
- Recognize errors in data and deletes, alters, or excludes data from study.
- Validate data using semi-automated approaches fit for large data sets.
- Explain to the data scientist that sensitivity analysis is needed when the model cannot be validated manually.
- Search for understanding of putative mechanisms of found correlations in clinical decision-making tools, within domain knowledge, personal interactions, and literature search.
- Develop Data privacy impact assessment (DPIA) and data management plan (DMP).
Study load: 40 hours
Self-paced - 100% flexible
For more information click "LINK TO ORIGINAL" below.