MPI Head Scanner (NeuroMap)

A Stroke Surveillance System for the ICU

Time is Brain. A famous quote for stroke treatment. With 1.9 million cells dying every minute with an acute stroke, the on site detection and prevention is a very important goal. However, the treatment of stroke strongly depends on the type of stroke, namely if it is an ischemic or a hemorrhagic stroke. For this decision the patient is typically imaged by the CT within the emergency room. After treatment there is a need of further surveillance to detect a recurrence of the stroke or an induced bleeding due to the treatment.

For the intensive care unit (ICU), no imaging modality is established, as a CT has a lot of infrastructure needs and a mobile ultrasound cannot image the brain within the skull. Magnetic Particle Imaging can fill this gap, as MPI systems can be built as low field systems which can work in unshielded environments and work on a standard power plug.

In 2019 I presented the first prototype of this device together with my colleagues from the University Medical Center Hamburg Eppendorf IBI UKE. The complete system characterization of the Hamburg scanner, including sensitivity, spatial resolution and fast dynamic perfusion imaging, was published in 2024. At the Fraunhofer IMTE, the system was redesigned to be mobile and to match product specifications and clinical requirements (NeuroMap). In parallel, I contributed to the design and validation of a second human brain MPI system in the group of Lawrence Wald at the Massachusetts General Hospital, published in 2025. As the system is so low in demands, it might even be possible to mount it within an emergency vehicle and bring the decision making direct to the patient.

Picture of the built head scanner, a brain phantom with a small stroke (42 ml) and the corresponding reconstruction.

To evaluate the suitability of the system two experiments were carried out: First, a static experiment with two nanoparticle filled hemispheres containing a concentration of 965 ng/ml iron. This concentration corresponds well to the concentration expected within the brain considering both blood and brain matter. On the right hemisphere a small region of 42 ml was left without nanoparticles to represent a very small stroke. This can easily be identified on the reconstructed image.

In the clinic, dynamic images are typically used to calculate perfusion parameter maps. These maps can tell the physician more about the type of stroke, the severity and the best treatment scenario. Therefore, we also performed a dynamic experiment. Two perfusion chambers were constructed where an induced flow enters the chamber from below, then passes inwards, circles around the phantom to the outer sides and exits again at the bottom. A sketch can be seen in the image below.

Picture of the dynamic phantom (upper right), resulting perfusion parameter maps which match nicely the perfusion dynamics of the constructed phantom.

With the image series perfusion parameters were calculated, namely the Time to Peak map (TTP) the mean transit time (MTT), the relative cerebral blood flow (rCBF) and the relative cerebral blood volume (rCBV), which are the most relevant parameters for treatment decision.

At the Fraunhofer IMTE we are now rebuilding the proof of principle into a functional product version. For more information please contact me at matthias.graeser@imte.fraunhofer.de

References

2026

  1. Ackers2026Concepts.png
    Comparison of human-scale MPI system concepts using a 3D arbitrary-waveform magnetic particle spectrometer
    Justin Ackers, and 3 more authors
    In International Workshop on Magnetic Particle Imaging 2026, 2026
  2. Topcu2026CoilArray.png
    Spherical harmonic-based receive coil array for a head MPI system
    Onurcan Topcu, and 10 more authors
    In International Workshop on Magnetic Particle Imaging 2026, 2026
  3. Wang2026JFET.png
    A Low-Noise Differential JFET Amplifier for Magnetic Particle Imaging
    Haozhe Wang, and 5 more authors
    In International Workshop on Magnetic Particle Imaging 2026, 2026

2025

  1. Mattingly2025Brain.jpg
    Design, construction and validation of a magnetic particle imaging (MPI) system for human brain imaging
    Eli Mattingly, and 8 more authors
    Physics in Medicine & Biology, 2025

2024

  1. Thieben2024SystemChar.png
    System characterization of a human-sized 3D real-time magnetic particle imaging scanner for cerebral applications
    Florian Thieben, and 8 more authors
    Communications Engineering, 2024
  2. Mohn2024Coupling.png
    Resonant inductive coupling network for human-sized magnetic particle imaging
    Fabian Mohn, and 6 more authors
    Review of Scientific Instruments, 2024

2021

  1. Ludewig_Wires.jpg
    Magnetic particle imaging for assessment of cerebral perfusion and ischemia
    P. Ludewig, and 10 more authors
    WIREs Nanomedicine and Nanobiotechnology, 2021

2019

  1. Graeser_HumanSized.jpg
    Human-sized Magnetic Particle Imaging for Brain Applications
    Matthias Graeser, and 13 more authors
    Nature Communications, 2019

2017

  1. Ludewig_Stroke.gif
    Magnetic Particle Imaging for Real-Time Perfusion Imaging in Acute Stroke
    Peter Ludewig, and 16 more authors
    ACS Nano, 2017
    PMID: 28976180