First positronium imaging using $^{44}$Sc with the J-PET scanner: a case study on the NEMA-Image Quality phantom

This study presents the first experimental demonstration of Positronium Lifetime Imaging (PLI) using the $^{44}$Sc isotope and the plastic scintillator-based J-PET scanner on a NEMA-Image Quality phantom, addressing the prompt gamma limitations of previous isotopes like $^{68}$Ga to advance submolecular tissue characterization.

The Gist

Imagine you are trying to take a photograph of a busy city street at night. A standard Positron Emission Tomography (PET) scan is like taking a regular photo: it tells you where the people (cells) are and how active they are, but it doesn't tell you much about who they are or what they are feeling.

This paper introduces a new, super-powered camera technique called Positronium Lifetime Imaging (PLI). Instead of just taking a picture, this new method tries to listen to the "heartbeat" of the atoms to understand the microscopic environment of the tissue.

Here is the story of how the researchers did it, explained simply:

1. The Mystery Guest: The Positronium Atom

When a radioactive tracer is injected into the body, it shoots out tiny particles called positrons. Usually, these positrons crash into electrons immediately and vanish in a flash of light (the signal a normal PET scanner sees).

However, about 40% of the time, a positron doesn't crash immediately. Instead, it grabs an electron and they dance together for a tiny, tiny moment before vanishing. This dancing pair is called Positronium.

Think of Positronium like a firefly that stays lit for a split second before disappearing.

• In healthy tissue, the firefly might dance for a specific amount of time.
• In diseased tissue (like a tumor or an area with low oxygen), the environment might be "crowded" or "toxic," causing the firefly to get snuffed out faster or slower.

By measuring exactly how long this "dance" lasts, doctors could potentially detect diseases at a molecular level long before a tumor becomes visible on a standard scan.

2. The Problem: The Wrong Flashlight

To measure how long the firefly dances, you need a perfect stopwatch. You need to know exactly when the dance started.

• The Old Way (68Ga): The researchers previously used a radioactive element called Gallium-68. It's like a flashlight that sometimes flashes a signal to say "Start!" but only does so 1 out of 75 times. Most of the time, you miss the start signal, so you can't time the dance accurately.
• The New Way (44Sc): This paper introduces a new element, Scandium-44. This is like a flashlight that flashes a "Start!" signal every single time (100% of the time). It's a much better timekeeper.

3. The Experiment: The "Toy City"

The researchers didn't test this on a human yet. Instead, they built a NEMA IQ Phantom.

• The Analogy: Imagine a giant plastic ball with six smaller plastic balls inside it, like a Russian nesting doll set, but the balls are different sizes.
• They filled the small balls with the "old" radioactive material (Gallium) and the big balls with the "new" material (Scandium).
• They put this whole setup inside a special scanner called J-PET.

What is J-PET?
Most PET scanners are made of expensive crystal blocks. J-PET is built from plastic strips (like long, clear plastic rulers). It's cheaper, lighter, and has a unique superpower: it can detect three flashes of light at once, whereas standard scanners usually only catch two. This is crucial because the "Start" signal (the prompt gamma) is the third flash.

4. The Results: Catching the Dance

The researchers ran the scanner and looked at the data:

• Filtering the Noise: They had to ignore the "old" balls (Gallium) because they didn't send the "Start" signal. They focused only on the "new" balls (Scandium) that sent all three signals.
• The Measurement: They successfully measured how long the Positronium "fireflies" danced inside the plastic balls.
• The Outcome: The results matched what scientists expect to see in water and plastic. The "dance time" was consistent and accurate.

Why This Matters

Think of this as moving from black-and-white photography to high-definition 3D video with sound.

1. Better Timing: Using Scandium-44 gives a much clearer "Start" signal, making the timing measurements much more reliable.
2. New Clues: This technique could eventually help doctors see diseases like cancer, Alzheimer's, or heart disease not just by looking at a lump, but by sensing the chemical "mood" of the cells.
3. Affordable Tech: Because they used the plastic-strip scanner (J-PET), this technology could eventually be made much cheaper than current high-end PET scanners, making advanced diagnostics available to more people.

In a nutshell: The researchers successfully proved that by using a special "super-flashlight" (Scandium-44) and a unique plastic-strip camera (J-PET), they can accurately measure the microscopic "dance" of atoms. This is the first step toward a new kind of medical imaging that sees the invisible chemistry of life.

Technical Summary

1. Problem Statement

Positronium Lifetime Imaging (PLI) is an emerging extension of Positron Emission Tomography (PET) that aims to image the mean lifetime of the positronium atom (Ps) to probe submolecular tissue properties (e.g., oxygen concentration, free radicals, and tissue structure). While conventional PET detects coincident 511 keV annihilation photons, PLI requires measuring the time difference between the formation of the positronium and its annihilation.

Key Challenges:

• Timing Marker: To measure the Ps lifetime ($\Delta T = t_{annihilation} - t_{formation}$), the exact formation time must be known. This requires a radionuclide that emits a "prompt gamma" immediately following positron emission to serve as a start signal.
• Isotope Limitations:
  • $^{22}$Na: Ideal for lab studies (99.94% prompt gamma yield) but has a 2.6-year half-life and bone uptake, making it unsuitable for human clinical use.
  • $^{68}$Ga: Used in the first in-vivo human brain PLI study, but only 1.34% of decays are accompanied by a prompt gamma, severely limiting statistical power.
  • $^{44}$Sc: Promising due to a 4.04-hour half-life (clinically suitable) and a ~100% yield of a 1157 keV prompt gamma following positron emission. However, its feasibility for PLI had not been experimentally demonstrated on a scanner capable of handling high-energy photons without energy restrictions.
• Detector Constraints: Many clinical PET scanners (e.g., Biograph Vision Quadra) have energy detection limits (typically <950 keV) that prevent the efficient detection of the 1157 keV prompt gamma from $^{44}$Sc, leading to significant event loss.

2. Methodology

A. Experimental Setup

• Scanner: The study utilized the Modular J-PET scanner, a total-body PET system based on plastic scintillators. Unlike crystal-based scanners, J-PET operates in a trigger-less mode, allowing simultaneous multiphoton detection without strict energy thresholds, making it ideal for capturing the high-energy prompt gamma.
• Phantom: A NEMA Image Quality (IQ) Phantom was used. It contained six spherical inserts (diameters: 10, 13, 17, 22, 28, and 37 mm).
  • Small spheres (10, 13, 17 mm) were filled with $^{18}$F (no prompt gamma) to test background rejection.
  • Large spheres (22, 28, 37 mm) were filled with $^{44}$Sc (high prompt gamma yield) to test PLI performance.
• Isotope Production: $^{44}$Sc was produced via the $^{44}$Ca(p,n)$^{44}$Sc reaction at the Heavy Ion Laboratory (University of Warsaw) and purified for the experiment.

B. Event Selection and Reconstruction

• Event Classification: The system distinguished between standard PET events (2-hit: two 511 keV photons) and PLI events (3-hit: two 511 keV photons + one 1157 keV prompt gamma).
• Time-Over-Threshold (TOT) Filtering: The TOT signal, proportional to energy deposition, was used to identify photon types:
  • 511 keV photons: TOT range 5.5 – 8.0 ns·V.
  • 1157 keV prompt gammas: TOT range 8.1 – 14.0 ns·V.
• Geometric and Temporal Cuts:
  • Multiplicity: Selected events with exactly 3 hits (2 annihilation + 1 prompt).
  • Angular Constraints: Enforced $\theta_{AA} \ge 60^\circ$ (angle between annihilation photons) and $\theta_{DA} \ge 30^\circ$ (angle between prompt gamma and annihilation photons) to reduce scatter and accidental coincidences.
  • Scatter Test (ST): Applied a time-position consistency check ($ST < -0.5$ ns) to filter out intra-detector Compton scatter.
• Image Reconstruction: Used the CASToR framework with the Maximum Likelihood Expectation Maximization (MLEM) algorithm (10 iterations) to reconstruct both standard PET images (2$\gamma_a$) and annihilation point distributions (2$\gamma_a + \gamma_p$).

C. Lifetime Estimation

• Time Calculation: The positron emission time ($t_p$) was approximated using the prompt gamma registration time ($t_1$) corrected for time-of-flight. The annihilation time ($t_a$) was derived from the two 511 keV hits.
• Fitting Models: The measured lifetime spectrum ($\Delta T$) was fitted using three models to estimate the mean ortho-positronium (oPs) lifetime ($\tau_{oPs}$):
  1. Fixed background.
  2. Constrained background (varying within $\pm 3\sigma$).
  3. Extended parameter range (free lifetimes and intensities).
• Robust Metric: The mean positron lifetime ($\Delta T_{mean}$) was also calculated as a model-independent parameter.

3. Key Contributions

1. First Experimental Demonstration: This is the first study to successfully perform PLI using $^{44}$Sc, validating it as a superior clinical isotope compared to $^{68}$Ga due to its ~75x higher prompt gamma yield.
2. Scanner Validation: Demonstrated the unique capability of the plastic-scintillator-based J-PET to detect high-energy (1157 keV) prompt gammas without the energy cutoff limitations of crystal-based scanners.
3. Methodological Framework: Established a rigorous event selection and reconstruction pipeline for multiphoton PLI, including specific TOT ranges and angular constraints to isolate true Ps events from background.
4. Quantitative Analysis: Provided the first quantitative oPs lifetime measurements in a controlled phantom environment using $^{44}$Sc, comparing different statistical fitting models to address low-statistics challenges.

4. Results

• Imaging Specificity: The reconstructed images of 3-hit events (2$\gamma_a + \gamma_p$) showed a significant reduction in signal from the $^{18}$F spheres (which lack prompt gammas), confirming the system's ability to isolate $^{44}$Sc events. The signal was concentrated in the 22, 28, and 37 mm spheres filled with $^{44}$Sc.
• Lifetime Measurements:
  • Large Spheres (28 mm & 37 mm): The mean oPs lifetime was measured at 1.821 ± 0.061 ns and 1.804 ± 0.042 ns respectively (using Model 1). These values are in excellent agreement with the known oPs lifetime in water/PMMA (~1.839 ns).
  • Small Sphere (22 mm): Initial results showed a lower lifetime (1.413 ns), attributed to low event statistics and background overestimation.
  • Model Sensitivity: Model 2 (constrained background) improved the 22 mm sphere result (1.693 ns), demonstrating that adaptive background handling is crucial for low-statistics datasets.
• Robustness: The mean positron lifetime ($\Delta T_{mean}$) proved to be a highly robust parameter, varying by less than 10 ps across all fitting models and background estimation methods, regardless of the sphere size.

5. Significance

• Clinical Translation: This study bridges the gap between preclinical PLI and clinical application. $^{44}$Sc is a theranostic isotope (used for both imaging and therapy) with a suitable half-life for clinical workflows. Proving its viability for PLI opens the door for new diagnostic biomarkers based on tissue microstructure.
• Technological Advancement: The results validate the J-PET scanner as a cost-effective, scalable platform for PLI. Its ability to detect high-energy prompt gammas makes it uniquely suited for next-generation PLI, potentially outperforming expensive crystal-based systems in this specific modality.
• Future Impact: The combination of $^{44}$Sc and total-body PET systems (with large axial fields of view) promises to significantly enhance sensitivity, enabling PLI to become a routine clinical tool for characterizing diseases at the submolecular level (e.g., tumor hypoxia, inflammation, and fibrosis).

In conclusion, the paper successfully demonstrates that $^{44}$Sc combined with the J-PET scanner is a viable and powerful approach for Positronium Lifetime Imaging, overcoming previous limitations related to isotope yield and detector energy thresholds.