Improving timing resolution of BGO for TOF-PET: a comparative analysis with and without deep learning
Abstract Background The renewed interest in BGO scintillators for TOF-PET is driven by the improved Cherenkov photon detection with new blue-sensitive SiPMs. However, the slower scintillation light from BGO causes significant time walk with leading edge discrimination (LED), which degrades the coinc...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2025-01-01
|
| Series: | EJNMMI Physics |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s40658-024-00711-6 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832594408715845632 |
|---|---|
| author | Francis Loignon-Houle Nicolaus Kratochwil Maxime Toussaint Carsten Lowis Gerard Ariño-Estrada Antonio J. Gonzalez Etiennette Auffray Roger Lecomte |
| author_facet | Francis Loignon-Houle Nicolaus Kratochwil Maxime Toussaint Carsten Lowis Gerard Ariño-Estrada Antonio J. Gonzalez Etiennette Auffray Roger Lecomte |
| author_sort | Francis Loignon-Houle |
| collection | DOAJ |
| description | Abstract Background The renewed interest in BGO scintillators for TOF-PET is driven by the improved Cherenkov photon detection with new blue-sensitive SiPMs. However, the slower scintillation light from BGO causes significant time walk with leading edge discrimination (LED), which degrades the coincidence time resolution (CTR). To address this, a time walk correction (TWC) can be done by using the rise time measured with a second threshold. Deep learning, particularly convolutional neural networks (CNNs), can also enhance CTR by training with digitized waveforms. It remains to be explored how timing estimation methods utilizing one (LED), two (TWC), or multiple (CNN) waveform data points compare in CTR performance of BGO scintillators. Results In this work, we compare classical experimental timing estimation methods (LED, TWC) with a CNN-based method using the signals from BGO crystals read out by NUV-HD-MT SiPMs and high-frequency electronics. For $${2 \times 2 \times 3}\,\hbox {mm}^{3}$$ 2 × 2 × 3 mm 3 crystals, implementing TWC results in a CTR of 129 ± 2 ps FWHM, while employing the CNN yields 115 ± 2 ps FWHM, marking improvements of 18 % and 26 %, respectively, relative to the standard LED estimator. For $${2 \times 2 \times 20}\,\hbox {mm}^{3}$$ 2 × 2 × 20 mm 3 crystals, both methods yield similar CTR (around 240 ps FWHM), offering a $$\sim$$ ∼ 15 % gain over LED. The CNN, however, exhibits better tail suppression in the coincidence time distribution. Conclusions The higher complexity of waveform digitization needed for CNNs could potentially be mitigated by adopting a simpler two-threshold approach, which appears to currently capture most of the essential information for improving CTR in longer BGO crystals. Other innovative deep learning models and training strategies may nonetheless contribute further in a near future to harnessing increasingly discernible timing features in TOF-PET detector signals. |
| format | Article |
| id | doaj-art-5e68ab3533ba41e499e329a614bf662d |
| institution | Kabale University |
| issn | 2197-7364 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | EJNMMI Physics |
| spelling | doaj-art-5e68ab3533ba41e499e329a614bf662d2025-01-19T12:39:08ZengSpringerOpenEJNMMI Physics2197-73642025-01-0112111510.1186/s40658-024-00711-6Improving timing resolution of BGO for TOF-PET: a comparative analysis with and without deep learningFrancis Loignon-Houle0Nicolaus Kratochwil1Maxime Toussaint2Carsten Lowis3Gerard Ariño-Estrada4Antonio J. Gonzalez5Etiennette Auffray6Roger Lecomte7Instituto de Instrumentación para Imagen Molecular, Centro Mixto CSIC-Universitat Politècnica de ValènciaDepartment of Biomedical Engineering, University of California DavisSherbrooke Molecular Imaging Center and Department of Nuclear Medicine and Radiobiology, Université de SherbrookeCERN, Department EP-CMXDepartment of Biomedical Engineering, University of California DavisInstituto de Instrumentación para Imagen Molecular, Centro Mixto CSIC-Universitat Politècnica de ValènciaCERN, Department EP-CMXSherbrooke Molecular Imaging Center and Department of Nuclear Medicine and Radiobiology, Université de SherbrookeAbstract Background The renewed interest in BGO scintillators for TOF-PET is driven by the improved Cherenkov photon detection with new blue-sensitive SiPMs. However, the slower scintillation light from BGO causes significant time walk with leading edge discrimination (LED), which degrades the coincidence time resolution (CTR). To address this, a time walk correction (TWC) can be done by using the rise time measured with a second threshold. Deep learning, particularly convolutional neural networks (CNNs), can also enhance CTR by training with digitized waveforms. It remains to be explored how timing estimation methods utilizing one (LED), two (TWC), or multiple (CNN) waveform data points compare in CTR performance of BGO scintillators. Results In this work, we compare classical experimental timing estimation methods (LED, TWC) with a CNN-based method using the signals from BGO crystals read out by NUV-HD-MT SiPMs and high-frequency electronics. For $${2 \times 2 \times 3}\,\hbox {mm}^{3}$$ 2 × 2 × 3 mm 3 crystals, implementing TWC results in a CTR of 129 ± 2 ps FWHM, while employing the CNN yields 115 ± 2 ps FWHM, marking improvements of 18 % and 26 %, respectively, relative to the standard LED estimator. For $${2 \times 2 \times 20}\,\hbox {mm}^{3}$$ 2 × 2 × 20 mm 3 crystals, both methods yield similar CTR (around 240 ps FWHM), offering a $$\sim$$ ∼ 15 % gain over LED. The CNN, however, exhibits better tail suppression in the coincidence time distribution. Conclusions The higher complexity of waveform digitization needed for CNNs could potentially be mitigated by adopting a simpler two-threshold approach, which appears to currently capture most of the essential information for improving CTR in longer BGO crystals. Other innovative deep learning models and training strategies may nonetheless contribute further in a near future to harnessing increasingly discernible timing features in TOF-PET detector signals.https://doi.org/10.1186/s40658-024-00711-6Time-of-flight PETFast timingBGOTime resolutionCherenkovDeep learning |
| spellingShingle | Francis Loignon-Houle Nicolaus Kratochwil Maxime Toussaint Carsten Lowis Gerard Ariño-Estrada Antonio J. Gonzalez Etiennette Auffray Roger Lecomte Improving timing resolution of BGO for TOF-PET: a comparative analysis with and without deep learning EJNMMI Physics Time-of-flight PET Fast timing BGO Time resolution Cherenkov Deep learning |
| title | Improving timing resolution of BGO for TOF-PET: a comparative analysis with and without deep learning |
| title_full | Improving timing resolution of BGO for TOF-PET: a comparative analysis with and without deep learning |
| title_fullStr | Improving timing resolution of BGO for TOF-PET: a comparative analysis with and without deep learning |
| title_full_unstemmed | Improving timing resolution of BGO for TOF-PET: a comparative analysis with and without deep learning |
| title_short | Improving timing resolution of BGO for TOF-PET: a comparative analysis with and without deep learning |
| title_sort | improving timing resolution of bgo for tof pet a comparative analysis with and without deep learning |
| topic | Time-of-flight PET Fast timing BGO Time resolution Cherenkov Deep learning |
| url | https://doi.org/10.1186/s40658-024-00711-6 |
| work_keys_str_mv | AT francisloignonhoule improvingtimingresolutionofbgofortofpetacomparativeanalysiswithandwithoutdeeplearning AT nicolauskratochwil improvingtimingresolutionofbgofortofpetacomparativeanalysiswithandwithoutdeeplearning AT maximetoussaint improvingtimingresolutionofbgofortofpetacomparativeanalysiswithandwithoutdeeplearning AT carstenlowis improvingtimingresolutionofbgofortofpetacomparativeanalysiswithandwithoutdeeplearning AT gerardarinoestrada improvingtimingresolutionofbgofortofpetacomparativeanalysiswithandwithoutdeeplearning AT antoniojgonzalez improvingtimingresolutionofbgofortofpetacomparativeanalysiswithandwithoutdeeplearning AT etiennetteauffray improvingtimingresolutionofbgofortofpetacomparativeanalysiswithandwithoutdeeplearning AT rogerlecomte improvingtimingresolutionofbgofortofpetacomparativeanalysiswithandwithoutdeeplearning |