Feasibility of Volumetric Analysis using Bedside Ultra-Low-Field Portable Magnetic Resonance Imaging in Patients receiving Extracorporeal Membrane Oxygenation

This study demonstrates that ultra-low-field portable MRI is feasible for generating accurate volumetric brain measurements in patients receiving extracorporeal membrane oxygenation (ECMO), with results comparable to conventional MRI despite the presence of ECMO circuitry.

The Gist

Imagine a patient in the Intensive Care Unit (ICU) who is so sick that their heart and lungs need a machine called ECMO to keep them alive. This machine acts like an external heart-lung bypass, pumping blood and oxygen for them. However, this machine is a giant, complex tangle of tubes and metal wires that usually makes it impossible to use a standard MRI scanner. Why? Because the powerful magnets in a normal MRI would act like a giant magnet on a fridge, potentially ripping the metal parts of the ECMO machine off the patient or causing them to overheat. It's like trying to take a photo of a person wearing a suit of armor while standing next to a massive industrial magnet—it just doesn't work.

For a long time, doctors had to guess what was happening inside the brains of these critical patients because they couldn't safely get a clear picture.

The New Solution: The "Gentle Giant" Scanner
Researchers at Johns Hopkins Hospital tested a new, tiny, portable MRI scanner. Think of a standard MRI as a giant, roaring freight train that is too powerful to get close to the ECMO machine. This new scanner is more like a quiet, gentle bicycle. It uses a very weak magnetic field (ultra-low-field) that is safe to bring right up next to the patient and the ECMO machine without causing any trouble.

The Big Question: Can a "Bicycle" Do the Work of a "Freight Train"?
The study, called SAFE MRI ECMO, already proved that this "bicycle" scanner is safe and can spot brain injuries (like a bump on the head) just as well as the big machines. But the researchers wanted to know something more specific: Can this weak scanner measure the size of different parts of the brain accurately?

Imagine trying to measure the volume of water in a bucket. A standard MRI is like a high-tech laser scale that gives you a perfect number. The question was: Can this new, weaker scanner give us a number that is just as close to the truth, even with all the ECMO tubes and wires buzzing around the patient?

How They Tested It
The team looked at scans from 30 patients who were on ECMO. They used a special computer program to slice the brain images into tiny 3D pieces, calculating the size of:

• The whole brain
• The "gray matter" (the thinking part) and "white matter" (the wiring)
• The fluid-filled spaces (ventricles)
• The left and right sides of the brain

The Results: A Resounding "Yes!"
The findings were exciting. The measurements taken by the tiny, portable scanner were almost identical to measurements taken by standard, powerful MRI machines (and even scans taken when patients weren't on ECMO).

It's as if you used a simple ruler to measure a room, and it turned out to be just as accurate as using a high-tech 3D laser scanner, even though there was a messy construction site (the ECMO machine) right next to the room.

Why This Matters
The study also found a tiny difference between patients on two different types of ECMO support (one that helps the heart and lungs vs. one that just helps the lungs), showing the scanner is sensitive enough to catch subtle changes.

The Bottom Line
This paper proves that we can now use these portable, weak-magnet scanners to not only see brain injuries in critically ill patients but also to measure the brain's volume accurately. It's like finally being able to take a high-definition, 3D map of a patient's brain while they are hooked up to life support, giving doctors a powerful new tool to monitor and treat the sickest patients in the hospital.

Technical Summary

Technical Summary: Feasibility of Volumetric Analysis using Bedside Ultra-Low-Field Portable MRI in ECMO Patients

1. Problem Statement

Patients receiving Extracorporeal Membrane Oxygenation (ECMO) are critically ill and at high risk for acquired brain injuries. However, continuous neurological monitoring is significantly hindered by the logistical and safety challenges of transporting these unstable patients to conventional MRI suites. Furthermore, the presence of ECMO circuitry (pumps, oxygenators, and tubing) often introduces magnetic interference or safety concerns in standard high-field MRI environments. While the SAFE MRI ECMO study previously established that ultra-low-field (64 mT) portable MRI is safe and sensitive for detecting brain injuries in this population, it remained unclear whether these low-field images could support quantitative volumetric analysis, which is essential for tracking subtle neurological changes over time.

2. Methodology

• Study Design: This was a retrospective analysis of prospectively collected observational data.
• Cohort: The study analyzed data from 30 patients who underwent ECMO support at Johns Hopkins Hospital.
• Imaging Protocol: Patients were scanned using a portable ultra-low-field (64 mT) MRI system while remaining in the intensive care unit (ICU).
• Data Processing:
  • Sequence: T2-weighted scans were utilized.
  • Pipeline: A specialized volumetric pipeline was applied to the images to segment and quantify specific brain structures.
  • Metrics Calculated: The analysis determined volumes for:
    • Whole brain volume.
    • Tissue types: Total grey matter, total white matter, subcortical grey matter, and cerebellum.
    • Ventricular system: Total intracranial volume, total ventricles, and specific left/right lateral ventricles.
    • Hemispheric and regional volumes: Left/right hemispheres and telencephalon.
• Comparative Analysis: The segmented volumes from ECMO patients were compared against:
  1. Measurements from conventional field-strength MRI.
  2. Measurements from ultra-low-field MRI scans taken without ECMO instrumentation.
  3. Subgroup analysis was performed to compare Venoarterial (VA) vs. Veno-Venous (VV) ECMO support.

3. Key Contributions

• First Evidence of Volumetric Feasibility: This study provides the first empirical evidence that ultra-low-field (64 mT) MRI can generate reliable volumetric data comparable to conventional high-field MRI, even in the complex electromagnetic environment of an ECMO circuit.
• Validation of Portable Neuroimaging: It bridges the gap between qualitative injury detection (established in the SAFE MRI ECMO study) and quantitative structural monitoring, enabling the tracking of brain atrophy or edema in situ.
• ECMO-Specific Insights: The study successfully differentiated volumetric nuances between different ECMO support types (VA vs. VV), suggesting the technology's sensitivity to physiological variations in critical care settings.

4. Results

• Volume Comparability: Segmented brain volumes obtained from patients with ECMO instrumentation were comparable to those obtained from conventional MRI and ultra-low-field MRI without ECMO. This indicates that the ECMO circuitry does not significantly degrade the accuracy of volumetric segmentation at 64 mT.
• Subgroup Differences: A subtle but distinct volumetric difference was observed between patients supported by Venoarterial (VA) ECMO versus those on Veno-Venous (VV) ECMO, highlighting the potential for the technology to detect hemodynamic or perfusion-related structural variations.
• Comprehensive Segmentation: The pipeline successfully quantified a wide range of anatomical structures, including deep grey matter, white matter tracts, and ventricular systems, demonstrating the robustness of the ultra-low-field data for complex morphometric analysis.

5. Significance

The findings fundamentally shift the paradigm for neurological monitoring in critical care. By proving that bedside ultra-low-field MRI can provide accurate volumetric data, this study validates the technology as a viable tool for:

• Continuous Monitoring: Allowing for repeated, non-invasive neurological assessments without the risks associated with patient transport.
• Early Intervention: Enabling the detection of subtle structural changes (e.g., ventricular expansion or tissue loss) that might precede clinical deterioration.
• Personalized Care: Facilitating tailored management strategies for ECMO patients based on quantitative neuroimaging biomarkers, ultimately improving outcomes for this vulnerable population.