MEMoRIAL-M1.2 | Under-sampled MRI for percutaneous intervention
Projektleiter:
Projektbearbeiter:
Dipl.-Phys. Mario Breitkopf
Finanzierung:
Forschergruppen:
Background
Undersampling MR images leads to an insufficient amount of data for conventional reconstruction techniques, making it an ill posed inverse problem. Deep neural networks provide promising solutions to the problem, but lack explainability.
Objective
MRI acceleration, especially golden angle radial sampling, in the process making real time MRI possible.
Methods
>> Utilizing and improving data-driven neural network approaches and their analysis
Results
>> Up-to-date deep learning reconstruction methods for undersampled radial MR signal data in image and signal domain with competitive results in that field of research
Conclusions
Current methods still mark the starting point since they are still missing key points like holoporphic activation functions for computing complex gradients throughout neural nets.
Orignality
>> Problem specific methods that are tailored to the underlying complex valued MR problem
Keywords
>> MRI, undersampling, reconstruction, deep learning, unblackboxing
Undersampling MR images leads to an insufficient amount of data for conventional reconstruction techniques, making it an ill posed inverse problem. Deep neural networks provide promising solutions to the problem, but lack explainability.
Objective
MRI acceleration, especially golden angle radial sampling, in the process making real time MRI possible.
Methods
>> Utilizing and improving data-driven neural network approaches and their analysis
Results
>> Up-to-date deep learning reconstruction methods for undersampled radial MR signal data in image and signal domain with competitive results in that field of research
Conclusions
Current methods still mark the starting point since they are still missing key points like holoporphic activation functions for computing complex gradients throughout neural nets.
Orignality
>> Problem specific methods that are tailored to the underlying complex valued MR problem
Keywords
>> MRI, undersampling, reconstruction, deep learning, unblackboxing
Anmerkungen
Wiss. Co-Betreuende / Scientific Co-Supervisors: Prof. Dr. Georg Rose (OVGU:FEIT/IMT, STIMULATE)
Geräte im Projekt
Kooperationen im Projekt
Publikationen
Die Daten werden geladen ...
Die Daten werden geladen ...
Kontakt
Prof. Dr. habil. Oliver Speck
Otto-von-Guericke-Universität Magdeburg
Fakultät für Naturwissenschaften
Leipziger Str. 44
39120
Magdeburg
Tel.:+49 391 6756113
weitere Projekte
Die Daten werden geladen ...