Objective Long-axial field-of-view (LAFOV) positron emission tomography (PET) systems allow higher sensitivity, with an increased number of detected lines of response induced by a larger angle of acceptance. However this extended angle increases the number of multiple scatters and the scatter contribution within oblique planes. As scattering affects both quality and quantification of the reconstructed image, it is crucial to correct this effect with more accurate methods than the state-of-the-art single scatter simulation (SSS) that can reach its limits with such an extended field-of-view (FOV). In this work, which is an extension of our previous assessment of deep learning-based scatter estimation (DLSE) carried out on a conventional PET system, we aim to evaluate the DLSE method performance on LAFOV total-body PET. Approach The proposed DLSE method based on an convolutional neural network (CNN) U-Net architecture uses emission and attenuation sinograms to estimate scatter sinogram. The network was trained from Monte-Carlo (MC) simulations of XCAT phantoms [<InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="259_2025_7120_Article_IEq1.gif" Format="GIF" Height="10" Rendition="HTML" Resolution="72" Type="Linedraw" Width="13" /> </InlineMediaObject> <EquationSource Format="TEX">$^{\text {18}}$</EquationSource> <EquationSource Format="MATHML"><math> <mmultiscripts> <mrow /> <mrow /> <mtext>18</mtext> </mmultiscripts> </math></EquationSource> </InlineEquation>F]-FDG PET acquisitions using a Siemens Biograph Vision Quadra scanner model, with multiple morphologies and dose distributions. We firstly evaluated the method performance on simulated data in both sinogram and image domain by comparing it to the MC ground truth and SSS scatter sinograms. We then tested the method on seven [<InlineEquation ID="IEq2"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="259_2025_7120_Article_IEq1.gif" Format="GIF" Height="10" Rendition="HTML" Resolution="72" Type="Linedraw" Width="13" /> </InlineMediaObject> <EquationSource Format="TEX">$^{\text {18}}$</EquationSource> <EquationSource Format="MATHML"><math> <mmultiscripts> <mrow /> <mrow /> <mtext>18</mtext> </mmultiscripts> </math></EquationSource> </InlineEquation>F]-FDG and [<InlineEquation ID="IEq3"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="259_2025_7120_Article_IEq1.gif" Format="GIF" Height="10" Rendition="HTML" Resolution="72" Type="Linedraw" Width="13" /> </InlineMediaObject> <EquationSource Format="TEX">$^{\text {18}}$</EquationSource> <EquationSource Format="MATHML"><math> <mmultiscripts> <mrow /> <mrow /> <mtext>18</mtext> </mmultiscripts> </math></EquationSource> </InlineEquation>F]-PSMA clinical datasets, and compare it to SSS estimations. Results DLSE showed superior accuracy on phantom data, greater robustness to patient size and dose variations compared to SSS, and better lesion contrast recovery. It also yielded promising clinical results, improving lesion contrasts in [<InlineEquation ID="IEq4"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="259_2025_7120_Article_IEq1.gif" Format="GIF" Height="10" Rendition="HTML" Resolution="72" Type="Linedraw" Width="13" /> </InlineMediaObject> <EquationSource Format="TEX">$^{\text {18}}$</EquationSource> <EquationSource Format="MATHML"><math> <mmultiscripts> <mrow /> <mrow /> <mtext>18</mtext> </mmultiscripts> </math></EquationSource> </InlineEquation>F]-FDG datasets and performing consistently with [<InlineEquation ID="IEq5"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="259_2025_7120_Article_IEq1.gif" Format="GIF" Height="10" Rendition="HTML" Resolution="72" Type="Linedraw" Width="13" /> </InlineMediaObject> <EquationSource Format="TEX">$^{\text {18}}$</EquationSource> <EquationSource Format="MATHML"><math> <mmultiscripts> <mrow /> <mrow /> <mtext>18</mtext> </mmultiscripts> </math></EquationSource> </InlineEquation>F]-PSMA datasets despite no training with [<InlineEquation ID="IEq6"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="259_2025_7120_Article_IEq1.gif" Format="GIF" Height="10" Rendition="HTML" Resolution="72" Type="Linedraw" Width="13" /> </InlineMediaObject> <EquationSource Format="TEX">$^{\text {18}}$</EquationSource> <EquationSource Format="MATHML"><math> <mmultiscripts> <mrow /> <mrow /> <mtext>18</mtext> </mmultiscripts> </math></EquationSource> </InlineEquation>F]-PSMA. Significance LAFOV PET scatter can be accurately estimated from raw data using the proposed DLSE method.

Evaluation of deep learning-based scatter correction on a long-axial field-of-view PET scanner

期刊论文
发表日期：

2025年2月7日

查看全文

Baptiste Laurent,
Alexandre Bousse,
Thibaut Merlin,
Axel Rominger,
Kuangyu Shi,
Dimitris Visvikis

摘要

Objective

Long-axial field-of-view (LAFOV) positron emission tomography (PET) systems allow higher sensitivity, with an increased number of detected lines of response induced by a larger angle of acceptance. However this extended angle increases the number of multiple scatters and the scatter contribution within oblique planes. As scattering affects both quality and quantification of the reconstructed image, it is crucial to correct this effect with more accurate methods than the state-of-the-art single scatter simulation (SSS) that can reach its limits with such an extended field-of-view (FOV). In this work, which is an extension of our previous assessment of deep learning-based scatter estimation (DLSE) carried out on a conventional PET system, we aim to evaluate the DLSE method performance on LAFOV total-body PET.

Approach

The proposed DLSE method based on an convolutional neural network (CNN) U-Net architecture uses emission and attenuation sinograms to estimate scatter sinogram. The network was trained from Monte-Carlo (MC) simulations of XCAT phantoms [ $^{\text {18}}$ $^{18}$ F]-FDG PET acquisitions using a Siemens Biograph Vision Quadra scanner model, with multiple morphologies and dose distributions. We firstly evaluated the method performance on simulated data in both sinogram and image domain by comparing it to the MC ground truth and SSS scatter sinograms. We then tested the method on seven [ $^{\text {18}}$ $^{18}$ F]-FDG and [ $^{\text {18}}$ $^{18}$ F]-PSMA clinical datasets, and compare it to SSS estimations.

Results

DLSE showed superior accuracy on phantom data, greater robustness to patient size and dose variations compared to SSS, and better lesion contrast recovery. It also yielded promising clinical results, improving lesion contrasts in [ $^{\text {18}}$ $^{18}$ F]-FDG datasets and performing consistently with [ $^{\text {18}}$ $^{18}$ F]-PSMA datasets despite no training with [ $^{\text {18}}$ $^{18}$ F]-PSMA.

Significance

LAFOV PET scatter can be accurately estimated from raw data using the proposed DLSE method.