Spring til indhold.
HomeResearch

Artificial Intelligence in General Practice

AI inGenerelPractice

Purpose:

AI-POWER: PHYSICS-INFORMED ARTIFICIAL INTELLIGENCE FOR NEXT-GENERATION POWER ELECTRONICS

The AI-Power project aims to consolidate the Danish industry and university competitive edge in the energy sector by addressing the challenges of applying artificial intelligence (AI) to next-generation cognitive power electronics systems. Denmark stands in a leading position in the green energy transition, in which power electronics, as one of the critical technologies, have played a significant role. The evolving electrical energy systems require new principles and operational practices. It, in turn, means we must develop and implement new and updated system tools. This project will address three significant challenges in applying AI for power electronics industry applications:

  1. How to obtain and understand the most relevant database?
  2. How to develop computational-efficient and affordable AI solutions?
  3. How to quantify the uncertainty and interpret the results from AI?

The project will develop tailored physics-informed AI tools for designing, operating, and maintaining power electronic systems.

The main purpose of this research project, "Artificial Intelligence (AI) in General Practice," is to investigate how AI can improve efficiency and the quality of care in general practice. The project aims to identify legal and ethical concerns regarding the implementation of machine learning (ML) technology into general practice and explore whether such concerns might be mitigated through the use of novel privacy-preserving approaches, such as Federated Learning; with this kind of approach, the patient data used to train the algorithms remains locally stored and several numerical weights are instead sent to a central server, which will then process the weights received from the decentralized units resulting in an averaged ML model (as depicted below).

Participants:

Kasper Elmholdt, Christoffer Florczak, Jeppe Agger Nielsen (Department of Politics and Society), Roman Jurowetzki & Daniel Hein (AAU Business School), CLAAUDIA.