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 <InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="40658_2024_711_Article_IEq1.gif" Format="GIF" Height="17" Rendition="HTML" Resolution="72" Type="Linedraw" Width="103" /> </InlineMediaObject> <EquationSource Format="TEX">${2 \times 2 \times 3}\,\hbox {mm}^{3}$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mrow> <mn>2</mn> <mo>×</mo> <mn>2</mn> <mo>×</mo> <mn>3</mn> </mrow> <mspace width="0.166667em" /> <msup> <mtext>mm</mtext> <mn>3</mn> </msup> </mrow> </math></EquationSource> </InlineEquation> 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 <InlineEquation ID="IEq2"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="40658_2024_711_Article_IEq2.gif" Format="GIF" Height="17" Rendition="HTML" Resolution="72" Type="Linedraw" Width="111" /> </InlineMediaObject> <EquationSource Format="TEX">${2 \times 2 \times 20}\,\hbox {mm}^{3}$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mrow> <mn>2</mn> <mo>×</mo> <mn>2</mn> <mo>×</mo> <mn>20</mn> </mrow> <mspace width="0.166667em" /> <msup> <mtext>mm</mtext> <mn>3</mn> </msup> </mrow> </math></EquationSource> </InlineEquation> crystals, both methods yield similar CTR (around 240 ps FWHM), offering a <InlineEquation ID="IEq3"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="40658_2024_711_Article_IEq3.gif" Format="GIF" Height="6" Rendition="HTML" Resolution="72" Type="Linedraw" Width="17" /> </InlineMediaObject> <EquationSource Format="TEX">$\sim$</EquationSource> <EquationSource Format="MATHML"><math> <mo>∼</mo> </math></EquationSource> </InlineEquation>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.

Improving timing resolution of BGO for TOF-PET: a comparative analysis with and without deep learning

期刊论文
出版模式：

开放获取
发表日期：

2025年1月17日

查看全文

Francis Loignon-Houle,
Nicolaus Kratochwil,
Maxime Toussaint,
Carsten Lowis,
Gerard Ariño-Estrada,
Antonio J. Gonzalez,
Etiennette Auffray,
Roger Lecomte

摘要

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 ${2 \times 2 \times 3}\,\hbox {mm}^{3}$ $2 \times 2 \times 3 {mm}^{3}$ 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 ${2 \times 2 \times 20}\,\hbox {mm}^{3}$ $2 \times 2 \times 20 {mm}^{3}$ crystals, both methods yield similar CTR (around 240 ps FWHM), offering a $\sim$ $\sim$ 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.