Book a demo

Magnetic particle imaging technology landscape 2026

Magnetic Particle Imaging Technology Landscape 2026 — PatSnap Insights
Medical Imaging Technology

Magnetic Particle Imaging has evolved from a 2005 theoretical construct into a preclinical platform on the cusp of clinical deployment — this landscape maps the patent and literature signals across hardware, tracers, reconstruction, and application domains from 2013 to 2025, identifying where the field’s most consequential bottlenecks and opportunities lie.

PatSnap Insights Team Innovation Intelligence Analysts 14 min read
Share
Reviewed by the PatSnap Insights editorial team ·

What MPI is and why it matters in 2026

Magnetic Particle Imaging is a radiation-free, tomographic imaging modality that directly detects and quantifies superparamagnetic iron oxide nanoparticles (SPIONs) by exploiting their nonlinear magnetization response to applied magnetic fields, producing zero-background tissue signal with high temporal and spatial resolution. Unlike MRI, which detects proton resonance, MPI exploits the saturable, nonlinear magnetization of SPIONs governed by Langevin theory — an inherent zero-background imaging contrast that MRI and CT cannot replicate.

2005
Year MPI was first proposed by Gleich & Weizenecker
46/s
Volumetric frames per second in cardiovascular MPI
263 pmol
Fe/mL detection limit in human-brain-scale scanner
200 cells
In vitro detection sensitivity demonstrated by UC Berkeley
87 days
Neural graft monitoring duration in rat brain

A spatially structured magnetic field incorporating a field-free region (FFR) — either a field-free point (FFP) or field-free line (FFL) — is scanned across the imaging volume. Only particles within the FFR contribute a time-varying signal detectable by receive coils; particles outside are magnetically saturated and silent. First proposed by Gleich and Weizenecker in 2005, MPI has matured from a theoretical construct into an established preclinical platform now approaching clinical translation across cardiovascular, oncological, neurological, and interventional domains.

Field-Free Region (FFR) — core MPI concept

The FFR is the spatial zone within an MPI scanner where the magnetic field is near zero, allowing SPIONs located there to respond to the applied excitation field and generate a detectable signal. All particles outside the FFR are saturated and contribute no signal, creating the modality’s characteristic zero-background contrast. The FFR can be configured as a field-free point (FFP) or a field-free line (FFL), each with distinct SNR and scanning trade-offs.

This landscape analyses the innovation landscape across hardware systems, tracer design, image reconstruction, and application verticals, drawing on patent filings and literature records spanning 2013–2025. It represents a snapshot of innovation signals within this dataset only and should not be interpreted as a comprehensive view of the full industry.

Magnetic Particle Imaging (MPI) produces zero-background tissue signal by exploiting the nonlinear magnetization response of superparamagnetic iron oxide nanoparticles (SPIONs) to a spatially structured magnetic field, a physical principle fundamentally distinct from MRI’s proton resonance detection.

Three phases of MPI innovation: from theory to clinical threshold

MPI innovation from 2013 to 2025 progressed through three discernible phases — foundational, diversification, and clinical-translation engineering — each defined by distinct milestones in hardware capability, tracer performance, and application breadth.

Figure 1 — MPI Innovation Phase Timeline: Key Milestones by Year
Magnetic Particle Imaging Innovation Milestones Timeline 2013–2025 FOUNDATIONAL DEVELOPMENT & DIVERSIFICATION CLINICAL TRANSLATION 2013–2015 2016–2020 2021–2025 SPION design principles (2013) 200-cell detect. UC Berkeley (2015) 46 vol/s aneurysm hemodynamics (2016) Human-brain MPI 263 pmol (2019) OpenMPIData initiative (2020) Intraoperative freehand MPI (2021) Aselsan calib. patents EP/JP (2023–24) Mitsubishi FFL patent JP (2025)
MPI innovation progressed through three phases: foundational (2013–2015), development and diversification (2016–2020), and clinical-translation engineering (2021–2025), with each phase introducing distinct hardware, tracer, and application milestones.

The foundational phase (2013–2015) established theoretical and experimental baselines. Key milestones include first demonstrations of SPION design principles for MPI at the University of Hong Kong (2013), the first MPI cell tracking study demonstrating 200-cell detection in rat brain at UC Berkeley (2015), and multi-color MPI feasibility demonstration by Philips GmbH (2015). The University of Lübeck’s 2015 review summarized the field’s state at that inflection point.

The development and diversification phase (2016–2020) saw consolidation of preclinical platforms and expansion into new application domains. Human-scale brain imaging hardware was demonstrated by the Hamburg-Eppendorf group in 2019, detecting iron concentrations as low as 263 pmol Fe/mL. Cardiovascular MPI was validated for aneurysm hemodynamics at 46 volumes per second in 2016. The OpenMPIData initiative from Hamburg-Eppendorf in 2020 signaled growing community infrastructure by democratizing access to experimental datasets. Point-of-care MPI systems emerged from ETRI, South Korea, in 2020.

The clinical-translation and engineering maturity phase (2021–2025) reflects deliberate clinical-pathway engineering. Active patents from the University of California (EP, 2023) and Mitsubishi Electric (JP, 2025) address hardware optimization and FFL-based reconstruction respectively. Aselsan’s calibration methodology patents (EP 2024, JP 2023) address a critical bottleneck — system matrix acquisition time — that has constrained routine clinical deployment. Intraoperative applications from Leiden University Medical Center (2021) and breast-conserving surgery margin assessment from Harvard Medical School (2021) represent the frontier of translational activity in this dataset.

“Human-scale brain imaging hardware demonstrated iron detection at 263 pmol Fe/mL, operable in unshielded environments such as intensive care units — a threshold that reframes MPI from laboratory instrument to bedside tool.”

Hardware architectures and SPION tracer design: the twin rate-limiters

MPI’s clinical viability depends on two interlocking engineering challenges: generating and scanning a field-free region at clinically safe dB/dt and SAR levels in human-scale bore geometries, and synthesising SPION tracers whose physical properties are optimised for MPI rather than MRI.

FFR Hardware Architectures

The dominant patent-level innovation in this dataset focuses on excitation waveform engineering and FFR translation efficiency. The University of California’s pulsed MPI system (EP, 2023) introduces a pulse sequence generator driving FFR location shifts via excitation magnetic fields, enabling optimised scan time, amplifier heating, and dB/dt management. Their complementary patent on improved MPI techniques (EP, 2023) incorporates high-Q receive coils and multi-resolution scanning via intermodulated low and radio-frequency excitation signals. Mitsubishi Electric’s 2025 FFL-based device (JP, 2025) generates corrected projection data using sensitivity correction applied to system-function-derived projection data, with the FFL region scanned and rotated for full volumetric coverage. According to WIPO filing records, FFL geometries offer SNR advantages over FFP configurations at equivalent gradient strengths, and Mitsubishi’s filing signals industrial-level commitment to FFL as the preferred clinical architecture.

A self-shielded human-brain-scale MPI scanner developed by the University Medical Center Hamburg-Eppendorf achieved iron detection at 263 pmol Fe/mL and was designed to operate in unshielded environments such as intensive care units, as reported in 2019.

SPION Tracer Design

SPION performance is the single largest determinant of MPI sensitivity, resolution, and quantitativeness. Langevin theory predicts that the optimal SPION core diameter for MPI signal generation is approximately 25–30 nm — larger than typical commercial iron oxides — with narrow size distribution, low anisotropy, and controlled surface chemistry. University of Hong Kong researchers established how core size, size distribution, and surface modification govern spatial resolution and sensitivity (2013). Multicore nanoparticle formulations (MCP 3) demonstrated superior in vivo MPI performance over commercial Resovist in rat angiography at dose reductions, as reported by Charité Berlin in 2020. Philips GmbH’s multi-color MPI work established that particles in different environments or of different types produce separable spectral signatures, enabling simultaneous multi-tracer imaging (2015).

Figure 2 — MPI SPION Optimal Core Diameter vs. Commercial Iron Oxide Range
Optimal SPION Core Diameter for Magnetic Particle Imaging vs. Commercial Iron Oxide Range MPI Signal (relative) SPION Core Diameter (nm) 0 50 100 5 10 15 20 25 30 35 40 Typical commercial iron oxides (5–15 nm) Optimal MPI 25–30 nm MPI signal response curve
Langevin theory predicts optimal MPI signal at approximately 25–30 nm SPION core diameter — substantially larger than the 5–15 nm range of typical commercial iron oxides — underscoring the need for MPI-specific tracer development.

Bimodal marker fabrication combining MPI and MRI visibility was demonstrated using KLB and Bayoxide E8706 nanoparticle coatings on non-metallic guidewires (University Hospital Schleswig-Holstein, 2022). The absence of filed SPION composition patents in this dataset suggests either concentration in literature through academic disclosure, or a gap in prosecution strategy — an opportunity for pharmaceutical and nanomaterial companies to build IP position around optimised MPI-specific tracers.

Analyse the full MPI patent landscape and SPION tracer IP landscape with PatSnap Eureka.

Explore MPI Patents in PatSnap Eureka →

Image reconstruction and system calibration: the clinical deployment bottleneck

MPI reconstruction is fundamentally an ill-posed inverse problem, and the time required to acquire system matrices represents the primary commercial bottleneck preventing routine clinical deployment of MPI scanners today.

Two paradigms coexist: system matrix (SM)-based reconstruction, which requires time-consuming empirical calibration scans, and model-based reconstruction, which derives reconstruction operators from mathematical forward models. The SM approach is reviewed extensively by the Beijing Institute of Technology (2021), identifying background signal contamination and field-of-view limitations as primary challenges. Mathematical model-based approaches are characterised by the University of Bremen (2018), framing the forward operator as a Fredholm integral of the first kind. Quality enhancement through merging multi-angle scan data is addressed by Hochschule Darmstadt (2022).

Key finding: Aselsan’s coded calibration scene

The most operationally significant calibration advance in this dataset is Aselsan’s coded calibration scene, which distributes multiple nanoparticle samples in a volume larger than the field of view, moved linearly and rotationally to populate the system matrix efficiently. Active patents were granted in Europe (EP 2024) and Japan (JP 2023). This approach represents a freedom-to-operate risk for any commercial MPI scanner developer deploying similar calibration architectures.

Aselsan (Turkey) holds two active patents on MPI system matrix calibration methodology — EP 2024 and JP 2023 — covering a coded calibration scene approach that distributes nanoparticle samples across a volume larger than the field of view to populate the system matrix more efficiently, addressing the primary commercial bottleneck for clinical MPI deployment.

Temperature monitoring during liver tumour ablation using MPI has been validated with 1°C mean absolute deviation accuracy (University Medical Center Hamburg-Eppendorf, 2020). This level of thermometric precision — enabled by the temperature-dependent shift in SPION magnetisation curves — is a direct product of accurate system calibration and is clinically relevant for ablation margin control, a domain where standards bodies such as ISO are actively developing thermal therapy guidance frameworks.

Figure 3 — MPI Reconstruction Approaches: System Matrix vs. Model-Based Methods
MPI Image Reconstruction Approaches: System Matrix vs. Model-Based Comparison System Matrix (SM) Reconstruction Approach: Empirical calibration scans Advantage: High fidelity to real scanner physics Challenge: Time-consuming; background signal contamination and FOV limitations (Beijing IT, 2021) Model-Based Reconstruction Approach: Fredholm integral forward operator (Bremen, 2018) Advantage: No lengthy calibration scans required Enhancement: Multi-angle scan data merging improves quality (Hochschule Darmstadt, 2022)
Two reconstruction paradigms coexist in MPI: system matrix methods offer high fidelity but require time-consuming calibration, while model-based approaches avoid lengthy scans but require accurate forward models — Aselsan’s coded calibration scene patents (EP 2024, JP 2023) directly address the SM bottleneck.

Application domains: where MPI is winning and where it is emerging

MPI’s zero-background contrast and high temporal resolution have translated into demonstrable advantages across six application domains, each at a different stage of preclinical validation and clinical readiness.

Cardiovascular and Vascular Imaging

MPI’s combination of high temporal resolution — 46 volumes per second — and zero-background signal makes it particularly suited for dynamic blood pool imaging. Aneurysm hemodynamics assessment was demonstrated against DSA and 4D phase-contrast MRI benchmarks in 2016. Moving table MPI extending the field of view for whole-vasculature imaging was demonstrated at Hamburg-Eppendorf in 2018. In vivo rat aorta and inferior vena cava angiography confirmed dose-dependent sensitivity with novel multicore particles (Charité Berlin, 2020). Standards bodies including IEEE have begun addressing medical imaging system performance benchmarks that will be relevant as MPI cardiovascular systems approach regulatory submission.

Neuroimaging and Stroke

The human-brain-scale MPI scanner operating at intensive-care-unit conditions (2019) represents the most advanced hardware demonstration for neurological applications in this dataset. A dedicated neuroimaging review articulates use cases spanning stroke perfusion monitoring, traumatic brain injury, and real-time cerebral blood flow quantification (2019). MPI’s absence of ionising radiation and susceptibility artefacts positions it as a candidate for bedside stroke monitoring — a gap not currently addressable by CT (radiation) or MRI (susceptibility artefacts, incompatibility with metallic implants).

Oncology and Theranostics

Tumour-specific SPION accumulation was imaged in CT26 and MC38 murine colon carcinoma models via both intratumoral and intravenous delivery using a point-of-care MPI platform (Eulji University, 2022). Tumour self-homing of circulating tumour cells was visualised in a murine breast cancer model (University of Western Ontario, 2020). Intraoperative margin assessment in breast-conserving surgery was proposed with both a handheld detector and small-bore scanner configuration (Harvard Medical School, 2021). MPI-guided hyperthermia with temperature monitoring validated for liver tumour ablation used Lissajous scanning MPI as a multifunctional platform providing nanoparticle localisation, remote thermometry, and focused hyperthermia delivery (Charité Berlin, 2020), with 1°C mean absolute deviation thermometric accuracy (Hamburg-Eppendorf, 2020).

Cell Tracking and Regenerative Medicine

MPI’s direct SPION detection enables quantitative longitudinal cell tracking not achievable with SPION-based MRI, which detects particles only indirectly via proton relaxation effects. Neural graft clearance was monitored over 87 days in rat brain with 200-cell in vitro detection sensitivity (UC Berkeley, 2015). Iron-labelled tumour cell metastasis and cell death were tracked in vivo using combined bioluminescence and MPI (Michigan State University, 2022). Research published in journals indexed by Nature has highlighted quantitative cell tracking as one of MPI’s most differentiated capabilities relative to existing modalities.

Interventional Radiology and Intraoperative Navigation

Real-time 4D catheter tracking at 46 volumes per second for endovascular procedures was demonstrated in vitro in 2016. Freehand 3D MPI for sentinel lymph node biopsy in penile cancer, using SPION tracers combined with fluorescence guidance, was demonstrated clinically (Leiden University Medical Center, 2021). Nanoparticle swarm navigation — steering magnetite clouds at 8 mm/s with real-time MPI imaging feedback — opens robotic drug delivery pathways (Fraunhofer IMTE, 2021).

Fraunhofer IMTE demonstrated nanoparticle swarm navigation at 8 mm/s with real-time MPI imaging feedback in 2021, opening a robotically guided drug delivery paradigm in which MPI serves simultaneously as navigation sensor and delivery confirmation modality.

Map the full MPI application patent landscape across cardiovascular, oncology, and interventional domains with PatSnap Eureka.

Search MPI Application Patents in PatSnap Eureka →

Geographic concentration and strategic implications for IP teams

MPI innovation within this dataset is concentrated in fewer than 10 primary institutional nodes, with Germany, the United States, and Asia representing the three dominant geographies — each with distinct IP profiles and strategic postures.

Figure 4 — MPI Innovation Contributions by Geography (2013–2025 Dataset)
Magnetic Particle Imaging Geographic Innovation Concentration 2013–2025 Low Med High V.High Innovation Volume Very High Germany High USA Med-High Japan Medium China Medium S. Korea Low-Med Turkey Hardware/Systems IP dominant Literature/Applications dominant Calibration/Methods IP
Germany leads in MPI innovation volume within this dataset, driven by the Hamburg-Eppendorf ecosystem; the University of California holds the most substantive active US patent portfolio; Aselsan (Turkey) holds active calibration IP in both EP and JP jurisdictions; and Mitsubishi Electric (Japan) filed the most recent hardware patent in 2025.

Germany is the dominant geography for MPI systems research in this dataset, led by the University Medical Center Hamburg-Eppendorf (UKE), which contributes across angiography, catheter tracking, moving table imaging, IVOCT integration, OpenMPI data infrastructure, and human-scale scanner development. The Charité – Universitätsmedizin Berlin contributes hyperthermia theranostics and temperature-resolved MPI. Fraunhofer IMTE (Lübeck) contributes nanoswarm actuation. The University of Lübeck and University Hospital Schleswig-Holstein contribute tracer development and bimodal marker work.

United States: The Regents of the University of California hold two active European patent grants covering pulsed MPI architectures and improved scanning techniques (both EP, 2023) — the only large-entity US-origin MPI patents with URLs in this dataset. UC Berkeley’s 2015 cell tracking work and Harvard Medical School’s 2021 intraoperative margin assessment work represent key translational contributions.

Asia: Mitsubishi Electric (Japan) filed the most recent MPI hardware patent in this dataset (JP, 2025), covering FFL-based imaging with sensitivity-corrected reconstruction. South Korean contributions include ETRI’s point-of-care 3D MPI system (2020) — a 20×33×45 cm³ device weighing under 100 kg — and Eulji University’s preclinical tumour imaging studies (2022). Chinese institutions — Beijing Institute of Technology, Beijing You’an Hospital/Capital Medical University, and Zhuhai People’s Hospital — contribute primarily to review literature covering reconstruction, applications, and liver imaging. Turkey: Aselsan holds two active patents on system matrix calibration methodology (EP 2024, JP 2023), representing an unexpected geography for MPI system-level IP.

“Aselsan’s coded calibration scene patents (EP 2024, JP 2023) represent freedom-to-operate risk for any commercial MPI scanner developer — IP strategists should conduct FTO analysis before deploying similar calibration architectures.”

Six forward-looking signals are discernible from 2021–2025 records: pulsed and multi-resolution excitation architectures (University of California, EP 2023); FFL-based human-scale scanners with sensitivity correction (Mitsubishi Electric, JP 2025); efficient system matrix calibration (Aselsan, EP 2024, JP 2023); intraoperative and point-of-care miniaturisation toward bedside and operating-room-compatible devices; nanoparticle swarm navigation and robotic theranostics at 8 mm/s (Fraunhofer IMTE, 2021); and immunotherapy and CAR T-cell tracking for months-long quantitative MPI monitoring of cancer immune responses (A*STAR Singapore, 2021). The NIH has identified quantitative cell tracking and theranostic imaging as priority areas in its biomedical imaging roadmap, directly aligning with MPI’s core technical differentiators. IP teams at medical device companies and pharmaceutical developers should monitor CN, JP, and KR filings closely as Asian industrial and academic actors are expected to be significant IP generators in the next filing cycle, per PatSnap’s innovation intelligence platform analysis.

Frequently asked questions

Magnetic Particle Imaging — key questions answered

Still have questions? Let PatSnap Eureka answer them for you.

Ask PatSnap Eureka for a Deeper Answer →

References

  1. Applications of Magnetic Particle Imaging in Biomedicine: Advancements and Prospects — Beijing You’an Hospital, Capital Medical University, 2022
  2. Magnetic particle imaging: current developments and future directions — University of Lübeck, 2015
  3. Comprehensive Evaluation of Magnetic Particle Imaging (MPI) Scanners for Biomedical Applications — Gebze Technical University, 2022
  4. Magnetic particle imaging: tracer development and biomedical applications — University College London, 2022
  5. A Review of Magnetic Particle Imaging and Perspectives on Neuroimaging, 2019
  6. Lissajous scanning magnetic particle imaging as a multifunctional platform for magnetic hyperthermia therapy — Charité Berlin, 2020
  7. First experimental evidence of the feasibility of multi-color magnetic particle imaging — Philips GmbH, 2015
  8. Human-sized magnetic particle imaging for brain applications — University Medical Center Hamburg-Eppendorf, 2019
  9. Pulsed magnetic particle imaging systems and methods — The Regents of the University of California, EP 2023
  10. Improved techniques for magnetic particle imaging — The Regents of the University of California, EP 2023
  11. Magnetic particle imaging device, method, and program — Mitsubishi Electric, JP 2025
  12. Method of calibrating magnetic particle imaging system — Aselsan, EP 2024
  13. Method for calibrating a magnetic particle imaging system — Aselsan, JP 2023
  14. Magnetic Particle Imaging tracks the long-term fate of in vivo neural cell implants — UC Berkeley, 2015
  15. Magnetic Particle Imaging for High Temporal Resolution Assessment of Aneurysm Hemodynamics, 2016
  16. The Reconstruction of Magnetic Particle Imaging: Current Approaches Based on the System Matrix — Beijing Institute of Technology, 2021
  17. Mathematical models for magnetic particle imaging — University of Bremen, 2018
  18. Advancing intraoperative magnetic tracing using 3D freehand magnetic particle imaging — Leiden University Medical Center, 2021
  19. Concept for using MPI for intraoperative margin analysis in breast-conserving surgery — Harvard Medical School, 2021
  20. Selective Actuation and Tomographic Imaging of Swarming Magnetite Nanoparticles — Fraunhofer IMTE, 2021
  21. WIPO — World Intellectual Property Organization
  22. NIH — National Institutes of Health
  23. IEEE — Institute of Electrical and Electronics Engineers
  24. ISO — International Organization for Standardization
  25. Nature — Nature Portfolio journals

All data and statistics in this article are sourced from the references above and from PatSnap‘s proprietary innovation intelligence platform. This landscape is derived from a targeted set of patent and literature records spanning 2013–2025 and represents a snapshot of innovation signals within this dataset only; it should not be interpreted as a comprehensive view of the full industry.

Your Agentic AI Partner
for Smarter Innovation

PatSnap fuses the world’s largest proprietary innovation dataset with cutting-edge AI to
supercharge R&D, IP strategy, materials science, and drug discovery.

Book a demo