Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Manish Das, Sushil Sharma, Aleksander Bilewicz, Jarosław Choiński, Neha Chug, Catalina Curceanu, Eryk Czerwiński, Jakub Hajduga, Sharareh Jalali, Krzysztof Kacprzak, Tevfik Kaplanoglu, Łukasz Kapłon, Kamila Kasperska, Aleksander Khreptak, Grzegorz Korcyl, Tomasz Kozik, Karol Kubat, Deepak Kumar, Anoop Kunimmal Venadan, Edward Lisowski, Filip Lisowski, Justyna Medrala-Sowa, Simbarashe Moyo, Wiktor Mryka, Szymon Niedźwiecki, Piyush Pandey, Szymon Parzych, Alessio Porcelli, Bartłomiej Rachwał, Elena Perez del Rio, Martin Rädler, Axel Rominger, Kuangyu Shi, Magdalena Skurzok, Anna Stolarz, Tomasz Szumlak, Pooja Tanty, Keyvan Tayefi Ardebili, Satyam Tiwari, Kavya Valsan Eliyan, Rafał Walczak, Ermias Yitayew Beyene, Ewa Ł. Stępień, Paweł Moskal

TL;DR

This work demonstrates the first positronium lifetime imaging (PLI) using $^{44}$Sc with the Modular J-PET scanner on a NEMA IQ phantom. By exploiting the 1157 keV prompt gamma emitted after $\beta^+$ decay, the authors implement 3-hit event selection (two 511 keV annihilation photons plus one prompt gamma) and TOT-based discrimination to reconstruct conventional PET images and Ps-enhanced images, followed by lifetime estimation from the ΔT spectrum. The results show that large-diameter spheres filled with $^{44}$Sc yield mean ortho-positronium lifetimes $\tau_{oPs}$ in the water-like range ($\approx 1.8$ ns) and that the mean positron lifetime $\Delta T_{mean}$ is a robust descriptor across background models, validating the approach. This study establishes $^{44}$Sc as a strong candidate for PLI in clinical settings and underscores the potential of modular, plastic-scintillator PET systems to enable Ps lifetime measurements, particularly when paired with next-generation total-body PET platforms.

Abstract

Positronium Lifetime Imaging (PLI), an emerging extension of conventional positron emission tomography (PET) imaging, offers a novel window for probing the submolecular properties of biological tissues by imaging the mean lifetime of the positronium atom. Currently, the method is under rapid development in terms of reconstruction and detection systems. Recently, the first in vivo PLI of the human brain was performed using the J-PET scanner utilizing the $^{68}$Ga isotope. However, this isotope has limitations due to its comparatively low prompt gamma yields, which is crucial for positronium lifetime measurement. Among alternative radionuclides, $^{44}$Sc stands out as a promising isotope for PLI, characterized by a clinically suitable half-life (4.04 hours) emitting 1157 keV prompt gamma in 100% cases after the emission of the positron. This study reports the first experimental demonstration of PLI with $^{44}$Sc, carried out on a NEMA-Image Quality (IQ) phantom using the Modular J-PET tomograph-the first plastic scintillators-based PET scanner.

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

TL;DR

This work demonstrates the first positronium lifetime imaging (PLI) using Sc with the Modular J-PET scanner on a NEMA IQ phantom. By exploiting the 1157 keV prompt gamma emitted after decay, the authors implement 3-hit event selection (two 511 keV annihilation photons plus one prompt gamma) and TOT-based discrimination to reconstruct conventional PET images and Ps-enhanced images, followed by lifetime estimation from the ΔT spectrum. The results show that large-diameter spheres filled with Sc yield mean ortho-positronium lifetimes in the water-like range ( ns) and that the mean positron lifetime is a robust descriptor across background models, validating the approach. This study establishes Sc as a strong candidate for PLI in clinical settings and underscores the potential of modular, plastic-scintillator PET systems to enable Ps lifetime measurements, particularly when paired with next-generation total-body PET platforms.

Abstract

Positronium Lifetime Imaging (PLI), an emerging extension of conventional positron emission tomography (PET) imaging, offers a novel window for probing the submolecular properties of biological tissues by imaging the mean lifetime of the positronium atom. Currently, the method is under rapid development in terms of reconstruction and detection systems. Recently, the first in vivo PLI of the human brain was performed using the J-PET scanner utilizing the Ga isotope. However, this isotope has limitations due to its comparatively low prompt gamma yields, which is crucial for positronium lifetime measurement. Among alternative radionuclides, Sc stands out as a promising isotope for PLI, characterized by a clinically suitable half-life (4.04 hours) emitting 1157 keV prompt gamma in 100% cases after the emission of the positron. This study reports the first experimental demonstration of PLI with Sc, carried out on a NEMA-Image Quality (IQ) phantom using the Modular J-PET tomograph-the first plastic scintillators-based PET scanner.

Paper Structure

This paper contains 10 sections, 5 equations, 5 figures, 1 table.

Table of Contents

  1. Introduction
  2. Methods
  3. Isotope preparation
  4. Phantom Study
  5. Event Selection for positronium imaging
  6. Image reconstruction
  7. Positronium Lifetime estimation
  8. Results
  9. Discussion
  10. Conclusion

Figures (5)

  • Figure 1: (A) The decay schemes for the $^{44}$Sc isotope. $\beta^{+}$ denotes the positron yield, EC indicates electron capture contributions, and $\gamma$ represents the prompt gamma with energy indicated in the paranthesis. Additionally, the delay time is presented in blue text for clarity. The delay time denotes the average time between a positron's emission and a prompt gamma's emission. (B) The event definition of one prompt gamma ($t_1,\vec{r}_1$) and two annihilation photons (($t_2,\vec{r}_2$) and ($t_3,\vec{r}_3$)) is depicted in the modular J-PET, requisite for positronium lifetime estimation. (C) Experimental Time-Over-Threshold (TOT) spectra for $^{68}$Ga and $^{44}$Sc, showing an increased prompt gamma yield for $^{44}$Sc in the modular J-PET scanner.
  • Figure 2: (A) Transaxial CT scan of the NEMA IQ phantom schematically depicting the radiotracer distribution in the spheres, with those highlighted in red containing $^{18}$F and those in blue filled with $^{44}$Sc. (B) The NEMA IQ phantom positioned inside the modular J-PET detector.
  • Figure 3: (A) Distribution of time-over-threshold (TOT$_{Hit}$) for photon identification, with annihilation photons (red) and prompt gammas (blue) marked by distinct ranges. (B) Hit multiplicity ($\mu$) distribution for events, with the red-shaded histogram highlighting selected events containing exactly two annihilation photons and one prompt gamma. For $\mu = 3$, selected three hits correspond to two photons in the annihilation region and one in the prompt region of the TOT$_{Hit}$ distribution. For $\mu > 3$, events are selected with exactly two hits in the annihilation region and one hit in the prompt region, while additional hits in $\mathrm{TOT}_{\mathrm{hit}} < 5.5 \mathrm{ns\cdot V}$ or $\mathrm{TOT}_{\mathrm{hit}}\in(8-8.1) \mathrm{ns\cdot V}$ are discarded. (C) Distribution of the relative angle ($\theta_{AA}$) between annihilation photon vectors $\vec{r}_2$ and $\vec{r}_3$ (per Fig. \ref{['introduction']}B), with $\theta_{AA} \geq 60^\circ$ (red) as the selection criterion. (D) Distribution of the relative angle ($\theta_{DA}$) between prompt gamma vector $\vec{r}_1$ and annihilation photon vectors $\vec{r}_2$, $\vec{r}_3$ (per Fig. \ref{['introduction']}B), with $\theta_{DA} \geq 30^\circ$ (red) as the restriction.
  • Figure 4: (A) Transaxial view of the conventional PET image ($2\gamma_a$) obtained from the modular J-PET, reconstructed with CASToR and overlaid on the CT image. (B) Line profile along the indicated dashed line in the images. The red line represents the profile for the conventional PET image, showing that the $^{18}$F activity concentration is more than two times higher than $^{44}$Sc along the image slice. The blue line corresponds to the $2\gamma_a + \gamma_p$ image. (C) Transaxial view of the $2\gamma_a$ image for $2\gamma_a + \gamma_p$ events, reconstructed using two 511 keV photons in CASToR and overlaid on the CT image. (D) Transaxial view of the annihilation point distribution ($\vec{r}_a$) for $2\gamma_a + \gamma_p$ events overlaid on the CT image of the NEMA IQ phantom, with voxels having a relative intensity greater than 10% displayed.
  • Figure 5: (A) Transaxial view of the NEMA-IQ phantom with selected ROIs from the spheres, overlaid on the CT image of the NEMA IQ phantom. (B–D) Distributions of positron annihilation lifetimes ($\Delta T$) for the 22 mm (B), 28 mm (C), and 37 mm (D) diameter spheres. The black histograms represent the experimental data, while the overlaid curves correspond to the fitted components: pPs (C$_\text{pPs}$), direct annihilations (C$_\text{direct}$), oPs (C$_\text{oPs}$), and background from accidental coincidences. The red curve represents the total fit, obtained as the sum of all contributions. (E–F) Visualization of the estimated mean oPs lifetime ($\tau_\text{oPs}$) (E) and mean positron lifetime ($\Delta T_{\mathrm{mean}}$) (F) from Fitting Model 1, where the activity counts within each selected ROI are replaced by their respective lifetime values.