Meta menu:

From here, you can access the Emergencies page, Contact Us page, Accessibility Settings, Language Selection, and Search page.

Open Menu

You are here:

Holistisches AutoML für FPGAs - 2 (HALF2)

One aim of the project is to improve automatic ECG analysis through optimized artifact detection while simultaneously using energy-saving hardware. However, such detection cannot be achieved with sufficient predictive accuracy using conventional signal processing methods. In contrast, machine learning algorithms and, in particular, artificial neural networks (KNNs) are capable of successfully solving such complex problems. In terms of hardware implementation, this project analyzes two approaches regarding the integration of the FPGA into the existing hardware. 

The project builds on the results of the competition "Energy Efficient AI Systems" of the German Federal Ministry of Education and Research (BMBF)". This was won in the category FPGA (English: Field Programmable Gate Array, a programmable logic circuit) with the project: "Holistic approach to optimize FPGA architectures for deep neural networks via AutoML - Automatic Machine Learning (HALF)" by Fraunhofer Institute for Industrial Mathematics ITWM and TU Kaiserslautern. 


Fraunhofer Institute for Industrial Mathematics ITWM
Dr. Jens Krüger 
Abteilung High Performance Computing 
Fraunhofer-Platz 1
67663 Kaiserslautern 

Further project partners

  • Technische Universität Kaiserslautern - Fachbereich Elektrotechnik und Informationstechnik, Lehrstuhl Entwurf Mikroelektronischer Systeme (EMS), contact Prof. Dr. Norbert Wehn
  • GETEMED Medizin- und Informationstechnik AG, contact: Tilo Bochardt
  • Charité - Universitätsmedizin Berlin, CharitéCentrum 11 für Herz-, Kreislauf- und Gefäßmedizin - Arbeitsbereich kardiovaskuläre
  • Telemedizin, contact: Prof. Dr. med. Friedrich Köhler