The Effects of External Laser Positioning Systems for MRI Simulation on Image Quality and Quantitative MRI Values

This study demonstrates that while activating external laser positioning systems (ELPS) during MRI simulation generally preserves quantitative values, it significantly degrades image quality—specifically causing a four-fold signal-to-noise ratio drop and geometric distortion errors when using the integrated body coil—necessitating clear clinical guidelines to avoid ELPS interference during imaging.

The Gist

The Big Picture: The "Laser Light Show" Problem

Imagine you are trying to take a super-clear, high-definition photo of a delicate flower inside a very quiet, soundproof room. This is what an MRI machine does: it listens to the tiny whispers of water molecules in your body to create a picture.

Now, imagine that right next to this quiet room, someone turns on a loud, buzzing neon sign that flickers with static electricity. Even though the sign is outside the room, that buzzing noise leaks in, drowning out the flower's whispers. The photo comes out grainy and fuzzy.

This is exactly what this study investigated.

In modern hospitals, doctors use MRI Simulators to plan radiation therapy. To make sure the patient is lying in the exact right spot, technicians use External Laser Positioning Systems (ELPS)—basically, high-tech laser lights that project crosshairs onto the patient.

The manufacturers say: "Turn these lasers off while we take the picture, or the noise will ruin the image."

But, because the switch to turn them off is sometimes hard to reach or people forget, the lasers often stay on. This study asked: "Does leaving the lasers on actually ruin the MRI pictures and the math behind them?"

---

The Experiment: Testing the Noise

The researchers set up a series of tests using "phantoms" (fake bodies made of plastic and liquid) to simulate what happens when the lasers are OFF versus ON.

1. The "Cine" Test (The Movie)

They took a video of the MRI scan while they flipped the laser switch from OFF to ON.

• The Result: It was like watching a clean window get covered in static. As soon as the lasers turned on, a "zipper" of electronic noise appeared on the screen, moving across the image like a glitch in a video game.
• The Analogy: It's like trying to listen to a symphony while someone starts playing a radio with bad reception right next to the orchestra.

2. The "Coil" Test (The Microphones)

MRI machines use different "coils" (like giant headphones or microphones) to pick up the signal. They tested the main body coil and several smaller, specialized coils.

• The Result: The big, built-in "body coil" was the most sensitive. When the lasers were on, the signal quality dropped by 75% (a four-fold decrease). It was like switching from a high-end studio microphone to a cheap, crackling walkie-talkie.
• The Good News: The smaller, specialized coils (like the ones used for shoulders or spines) were much better at blocking the noise. They were like noise-canceling headphones; the lasers were still buzzing, but the headphones mostly ignored it.

3. The "Math" Test (The Numbers)

Radiation doctors don't just look at pictures; they use the MRI to calculate specific numbers (like how fast water moves or how much fat is in a tissue). They needed to know if the laser noise messed up these calculations.

• The Result: The average numbers stayed mostly the same. The lasers didn't trick the computer into thinking the fat content was higher or lower than it really was.
• The Catch: While the average was fine, the consistency got worse.
  • The Analogy: Imagine you are throwing darts at a bullseye.
    • Lasers OFF: You hit the bullseye every time, and all your darts are clustered tightly together.
    • Lasers ON: You still hit the bullseye on average, but your darts are scattered all over the board. Some are high, some are low.
  • This "scatter" (called standard deviation) was most noticeable in the diffusion scans (which measure how water moves). It made the measurements less reliable, even if they weren't "wrong" on average.

4. The "Robot" Test (Automatic Detection)

Modern MRIs use software to automatically find tiny markers (fiducials) inside the machine to check for distortion.

• The Result: When the lasers were on, the software got confused by the noise and missed the markers. It was like a robot trying to find a face in a crowd, but someone kept flashing a strobe light in its eyes.

---

The Takeaway: Why This Matters

This study found that while the lasers don't completely destroy the image or change the main medical numbers, they do introduce unnecessary static that makes the picture grainier and the measurements less consistent.

The Main Lessons:

1. Turn the Whole System Off: You can't just turn off the red laser beams. You have to turn off the entire system (including the monitor and the control box), or the electronic noise will still leak out.
2. The Body Coil is Vulnerable: If you are using the big, built-in coil (often used for quality checks), the lasers are a major problem.
3. Consistency is Key: Even if the numbers look okay, the "scatter" in the data means the doctors might have to be less confident in their measurements.

The Bottom Line

Think of the MRI scanner as a very sensitive ear. The laser system is a buzzing fly. If you leave the fly buzzing while the ear is listening, the ear gets annoyed and the message gets fuzzy.

The authors are urging hospitals to create a simple rule: "Lasers ON for setup, Lasers OFF for scanning." It's a small step that prevents a lot of electronic static, ensuring the clearest possible picture for the patient's treatment.

Technical Summary

1. Problem Statement

Magnetic Resonance Imaging (MRI) simulation is critical for radiation therapy treatment planning, requiring high soft-tissue contrast and geometric accuracy. To ensure reproducible patient positioning, modern MRI suites utilize External Laser Positioning Systems (ELPS), such as the LAP DORADOnova.

• The Issue: While ELPS are "MR Conditional," they must be fully deactivated (including the monitor and control unit, not just the lasers) during scanning to prevent electronic/Radiofrequency (RF) noise leakage.
• The Gap: Despite manufacturer warnings, clinical workflows often result in scans being performed with the ELPS active due to cumbersome switch placement or lack of staff training. It was previously unknown whether this leakage significantly degrades image quality (SNR, artifacts) or compromises the accuracy and repeatability of quantitative MRI sequences (T1, T2, ADC, PDFF, R2*) used in clinical research and practice.

2. Methodology

The study was conducted on a 3T Siemens Vida MRI simulator equipped with a LAP DORADOnova ELPS. The researchers utilized a multi-faceted approach using various phantoms and coil configurations:

• Phantoms Used:
  • Homogeneous Phantom: For Signal-to-Noise Ratio (SNR) assessment and artifact visualization.
  • ACR Large Phantom: For comprehensive image quality testing (uniformity, ghosting, resolution, distortion) per ACR accreditation standards.
  • Geometric Distortion Phantoms: A vendor-provided 3D phantom (Unity/Elekta) and a QUASAR™ GRID3D phantom to test fiducial detection and distortion algorithms.
  • Quantitative Phantoms: CaliberMRI Model 137 (for T1, T2, ADC) and Calimetrix Model 725 (for PDFF and R2*).
• Coil Configurations: Tests were performed using the integrated body coil and various multi-channel array coils (20-channel Head/Neck, 18-channel UltraFlex, 32-channel Spine, 16-channel Shoulder).
• Acquisition Protocols:
  • Cine Acquisition: Real-time monitoring of SNR changes during ELPS activation/deactivation.
  • Quantitative Sequences: Multi-dynamic multi-echo (MDME) for T1/T2, BLADE DWI for ADC, and quantitative Dixon (q-Dixon) for PDFF/R2*.
• Analysis:
  • SNR calculated via NEMA standards.
  • Quantitative values compared against NIST-traceable reference values using Lin's Concordance Correlation Coefficient (LCCC) and Bland-Altman plots.
  • Repeatability assessed via Coefficient of Variation (CoV) across five repetitions.

3. Key Contributions

This study provides the first detailed characterization of ELPS-induced interference on modern MRI simulation workflows, specifically addressing:

1. Coil-Specific Susceptibility: Differentiating the impact of ELPS noise on the integrated body coil versus modern multi-channel array coils.
2. Real-Time Artifact Visualization: Using cine imaging to map the temporal evolution of RF interference artifacts (from broadband noise to "zipper" artifacts).
3. Quantitative Impact: Determining whether ELPS activation introduces bias or reduces the repeatability of clinically critical quantitative maps (T1, T2, ADC, PDFF, R2*).

4. Key Results

A. Image Quality and SNR

• Integrated Body Coil: Suffered the most severe degradation. SNR dropped four-fold (from 73 to 19) when the ELPS was activated. Visible "zipper" artifacts and broadband noise were prominent.
• Multi-Channel Coils: The 20-channel Head/Neck, 18-channel UltraFlex, 32-channel Spine, and 16-channel Shoulder coils showed negligible SNR changes (e.g., Head/Neck SNR remained ~489 vs. 483).
• ACR Accreditation Tests: When using the Head/Neck coil, ELPS activation caused:
  • A 5% decrease in image intensity uniformity (T2-weighted).
  • A 4.5-fold increase in percent signal ghosting.
  • Failure of the low-contrast object detectability test.
• Geometric Distortion: ELPS activation caused automated fiducial detection algorithms to fail (false negatives) due to noise obscuring edges, particularly in the vendor-provided phantom.

B. Quantitative MRI Values

• Accuracy (Bias): There was no significant difference in the mean quantitative values (T1, T2, PDFF, R2*) between ELPS activated and deactivated states. The linear agreement (LCCC) remained high.
• Repeatability (Precision):
  • ADC (Apparent Diffusion Coefficient): Showed the most notable impact. While mean values were stable, the standard deviation and Coefficient of Variation (CoV) increased when the ELPS was active (CoV rose from ~6% to ~7%).
  • T1, T2, PDFF, R2: Showed minimal to no change in repeatability.

5. Significance and Conclusions

• Clinical Workflow Implications: The study confirms that while modern array coils are largely immune to ELPS noise, the integrated body coil is highly susceptible. This is critical because body coils are often used for mandatory quality assurance (QA) and geometric distortion testing. If the ELPS is left on, QA tests may fail unnecessarily, or geometric distortion algorithms may produce erroneous results.
• Quantitative Reliability: While ELPS activation does not introduce significant bias in quantitative values (meaning the average measurement remains accurate), it does degrade repeatability (increasing noise variance), particularly in diffusion-weighted imaging (ADC).
• Actionable Recommendations:
  1. Strict Deactivation: Clinical staff must be trained to deactivate the entire ELPS system (including the monitor), not just the lasers, during scanning.
  2. Workflow Integration: Institutions should implement automated interlocks or clearer protocols to ensure ELPS deactivation during QA and patient scans.
  3. Vendor Communication: The "MR Conditional" requirement for full system shutdown must be emphasized to technologists and physicists to prevent preventable noise artifacts.

In summary, while the impact on mean quantitative values is minimal, the ELPS-induced noise poses a significant risk to image quality (SNR), automated analysis robustness, and the precision of quantitative metrics, necessitating strict adherence to deactivation protocols.