Alexander Preuhs

Dr.-Ing. Alexander Preuhs

Researcher

Department of Computer Science
Chair of Computer Science 5 (Pattern Recognition)

Room: Room 09.130
Martensstr. 3
91058 Erlangen

Academic CV

Education

  • 05/2020 – present:
    Researcher,
    Siemens Healthineers
  • 03/2017 – 04/2020:
    PhD Student, Pattern Recognition Lab,
    Friedrich-Alexander Universtität Erlangen-Nürnberg
    • Data-driven and model-based rigid motion compensation for cone-beam computed tomography
  • 11/2014 – 03/2017:
    M.Sc.  in Medical Engineering, Major “Medical Image and Data-processing
    • Friedrich-Alexander Universtität Erlangen-Nürnberg
  • 05/2011 – 11/2014:
    B.Sc. 
    in Medical Engineering
  • 09/2002 – 04/2011:
    Abitur,
    Gymnasium Neubiberg, Munich

Professional Employment

  • 06/2016 – 12/2016:
    Master Student, Innovation at Siemens Healthineers
  • 05/2016 – 06/2016:
    Internship, Patent Attorneys Kraus&Weisert
  • 01/2016 – 06/2016:
    Working Student, DFG Project “Flatdetector-CT Consistency” at Pattern Recognition Lab, FAU
  • 11/2014 – 05/2016:
    Working Student, Innovation at Siemens Healthineers
  • 11/2013 – 11/2014:
    Executive Board, Krakadu e.V.
  • 06/2013 – 07/ 2013:
    Internship, Siemens Healthineers, X-ray Tube Manufacturing
  • 02/2013 – 03/2014:
    Internship, University Hospital, Erlangen

 

Projects

Motion Compensation for DynaCT Acquisitions

With modern flat panel C-arm systems, soft tissue imaging is possible during intervention. This can be of high potential in the field of neuroradiologic interventions, when a stroke is to be classified as either ischemic or hemorrhagic. Within the interventional stroke therapy (Mechanical Thrombectomy), DynaCT has the potential to substitute CT and MRT as standard diagnostic modalities. By this substitution, imaging and intervention could be done within one room, making the therapy much faster and thus increasing the quality of stroke therapy.

 

To reach this goal, the detectability of cerebral bleedings under DynaCT acquisitions must be as reliable as it is using CT. The acquisition of projection data for reconstruction is performed by rotating the C-arm around the patient. One rotation is typically performed in 20 seconds, increasing the risk of motion induced artifacts compared to CT, where a single rotation is performed in less than a second. Thus, a major drawback of DynaCT compared to CT is patient motion. This project aims at compensating motion artifacts in DynaCT acquisitions with the goal of a reliable detectability of cerebral bleedings. 

 

Rigid Motion Creator

A python tool to generate rigid motion is available on Opens external link in new windowGitHub.

 

 

 

 

Interactive Fan-Beam Reconstruction

An interactive python prototy

pe based on Opens external link in new windowPyConrad for simulating fan-beam projections and fan-beam reconstruction is available on Opens external link in new windowgithub. The reconstruction is performed analytically in a filtered-backprojection scheme.

 

Publications

2021

Journal Articles

Conference Contributions

2020

Journal Articles

Book Contributions

Conference Contributions

Thesis

2019

Journal Articles

Conference Contributions

2018

Journal Articles

Conference Contributions

2015

Conference Contributions