Dataset Information

Improved amyloid burden quantification with nonspecific estimates using deep learning.

ABSTRACT:

Purpose

Standardized uptake value ratio (SUVr) used to quantify amyloid-β burden from amyloid-PET scans can be biased by variations in the tracer's nonspecific (NS) binding caused by the presence of cerebrovascular disease (CeVD). In this work, we propose a novel amyloid-PET quantification approach that harnesses the intermodal image translation capability of convolutional networks to remove this undesirable source of variability.

Methods

Paired MR and PET images exhibiting very low specific uptake were selected from a Singaporean amyloid-PET study involving 172 participants with different severities of CeVD. Two convolutional neural networks (CNN), ScaleNet and HighRes3DNet, and one conditional generative adversarial network (cGAN) were trained to map structural MR to NS PET images. NS estimates generated for all subjects using the most promising network were then subtracted from SUVr images to determine specific amyloid load only (SAβ_L). Associations of SAβ_L with various cognitive and functional test scores were then computed and compared to results using conventional SUVr.

Results

Multimodal ScaleNet outperformed other networks in predicting the NS content in cortical gray matter with a mean relative error below 2%. Compared to SUVr, SAβ_L showed increased association with cognitive and functional test scores by up to 67%.

Conclusion

Removing the undesirable NS uptake from the amyloid load measurement is possible using deep learning and substantially improves its accuracy. This novel analysis approach opens a new window of opportunity for improved data modeling in Alzheimer's disease and for other neurodegenerative diseases that utilize PET imaging.

SUBMITTER: Liu H

PROVIDER: S-EPMC8113180 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:ObjectivesThe purpose of this study was to determine phenotypes characterizing cardiac involvement in AL amyloidosis by using direct (fluorine-18-labeled florbetapir {[18F]florbetapir} positron emission tomography [PET]/computed tomography) and indirect (echocardiography and cardiac magnetic resonance [CMR]) imaging biomarkers of AL amyloidosis.BackgroundCardiac involvement in systemic light chain amyloidosis (AL) is the main determinant of prognosis and, therefore, guides management. The hypothesis of this study was that myocardial AL deposits and expansion of extracellular volume (ECV) could be identified before increases in N-terminal pro-B-type natriuretic peptide or wall thickness.MethodsA total of 45 subjects were prospectively enrolled in 3 groups: 25 with active AL amyloidosis with cardiac involvement (active-CA), 10 with active AL amyloidosis without cardiac involvement by conventional criteria (active-non-CA), and 10 with AL amyloidosis with cardiac involvement in remission for at least 1 year (remission-CA). All subjects underwent echocardiography, CMR, and [18F]florbetapir PET/CT to evaluate cardiac amyloid burden.ResultsThe active-CA group demonstrated the largest myocardial AL amyloid burden, quantified by [18F]florbetapir retention index (RI) 0.110 (interquartile range [IQR]: 0.078 to 0.139) min-1, and the lowest cardiac function by global longitudinal strain (GLS), median GLS -11% (IQR: -8% to -13%). The remission-CA group had expanded extracellular volume (ECV) and [18F]florbetapir RI of 0.097 (IQR: 0.070 to 0.124 min-1), and abnormal GLS despite hematologic remission for >1 year. The active-non-CA cohort had evidence of cardiac amyloid deposition by advanced imaging metrics in 50% of the subjects; cardiac involvement was identified by late gadolinium enhancement in 20%, elevated ECV in 20%, and elevated [18F]florbetapir RI in 50%.ConclusionsEvidence of cardiac amyloid infiltration was found based on direct and indirect imaging biomarkers in subjects without CA by conventional criteria. The findings from [18F]florbetapir PET imaging provided insight into the preclinical disease process and on the basis of interpretation of expanded ECV on CMR and have important implications for future research and clinical management of AL amyloidosis. (Molecular Imaging of Primary Amyloid Cardiomyopathy [MICA]; NCT02641145).

Project description:BackgroundQuantification of myocardium scarring in late gadolinium enhanced (LGE) cardiac magnetic resonance imaging can be challenging due to low scar-to-background contrast and low image quality. To resolve ambiguous LGE regions, experienced readers often use conventional cine sequences to accurately identify the myocardium borders.PurposeTo develop a deep learning model for combining LGE and cine images to improve the robustness and accuracy of LGE scar quantification.Study typeRetrospective.PopulationA total of 191 hypertrophic cardiomyopathy patients: 1) 162 patients from two sites randomly split into training (50%; 81 patients), validation (25%, 40 patients), and testing (25%; 41 patients); and 2) an external testing dataset (29 patients) from a third site.Field strength/sequence1.5T, inversion-recovery segmented gradient-echo LGE and balanced steady-state free-precession cine sequences ASSESSMENT: Two convolutional neural networks (CNN) were trained for myocardium and scar segmentation, one with and one without LGE-Cine fusion. For CNN with fusion, the input was two aligned LGE and cine images at matched cardiac phase and anatomical location. For CNN without fusion, only LGE images were used as input. Manual segmentation of the datasets was used as reference standard.Statistical testsManual and CNN-based quantifications of LGE scar burden and of myocardial volume were assessed using Pearson linear correlation coefficients (r) and Bland-Altman analysis.ResultsBoth CNN models showed strong agreement with manual quantification of LGE scar burden and myocardium volume. CNN with LGE-Cine fusion was more robust than CNN without LGE-Cine fusion, allowing for successful segmentation of significantly more slices (603 [95%] vs. 562 (89%) of 635 slices; P < 0.001). Also, CNN with LGE-Cine fusion showed better agreement with manual quantification of LGE scar burden than CNN without LGE-Cine fusion (%ScarLGE-cine = 0.82 × %Scarmanual , r = 0.84 vs. %ScarLGE = 0.47 × %Scarmanual , r = 0.81) and myocardium volume (VolumeLGE-cine = 1.03 × Volumemanual , r = 0.96 vs. VolumeLGE = 0.91 × Volumemanual , r = 0.91).Data conclusionCNN based LGE-Cine fusion can improve the robustness and accuracy of automated scar quantification.Level of evidence3 TECHNICAL EFFICACY: 1.

Project description:PurposeWe aimed to evaluate the performance of deep learning-based generalization of ultra-low-count amyloid PET/MRI enhancement when applied to studies acquired with different scanning hardware and protocols.MethodsEighty simultaneous [18F]florbetaben PET/MRI studies were acquired, split equally between two sites (site 1: Signa PET/MRI, GE Healthcare, 39 participants, 67 ± 8 years, 23 females; site 2: mMR, Siemens Healthineers, 64 ± 11 years, 23 females) with different MRI protocols. Twenty minutes of list-mode PET data (90-110 min post-injection) were reconstructed as ground-truth. Ultra-low-count data obtained from undersampling by a factor of 100 (site 1) or the first minute of PET acquisition (site 2) were reconstructed for ultra-low-dose/ultra-short-time (1% dose and 5% time, respectively) PET images. A deep convolution neural network was pre-trained with site 1 data and either (A) directly applied or (B) trained further on site 2 data using transfer learning. Networks were also trained from scratch based on (C) site 2 data or (D) all data. Certified physicians determined amyloid uptake (+/-) status for accuracy and scored the image quality. The peak signal-to-noise ratio, structural similarity, and root-mean-squared error were calculated between images and their ground-truth counterparts. Mean regional standardized uptake value ratios (SUVR, reference region: cerebellar cortex) from 37 successful site 2 FreeSurfer segmentations were analyzed.ResultsAll network-synthesized images had reduced noise than their ultra-low-count reconstructions. Quantitatively, image metrics improved the most using method B, where SUVRs had the least variability from the ground-truth and the highest effect size to differentiate between positive and negative images. Method A images had lower accuracy and image quality than other methods; images synthesized from methods B-D scored similarly or better than the ground-truth images.ConclusionsDeep learning can successfully produce diagnostic amyloid PET images from short frame reconstructions. Data bias should be considered when applying pre-trained deep ultra-low-count amyloid PET/MRI networks for generalization.