Deep-learning-based cardiac amyloidosis classification from early acquired pet images

The objective of the present work was to evaluate the potential of deep learning tools for characterizing the presence of cardiac amyloidosis from early acquired PET images, i.e. 15 min after [18F]-Florbetaben tracer injection. 47 subjects were included in the study: 13 patients with transthyretin-related amyloidosis cardiac amyloidosis (ATTR-CA), 15 patients with immunoglobulin light-chain amyloidosis (AL-CA), and 19 control-patients (CTRL). [18F]-Florbetaben PET/CT images were acquired in list mode and data was sorted into a sinogram, covering a time interval of 5 min starting 15 min after the injection. The resulting sinogram was reconstructed using OSEM iterative algorithm. A deep convolutional neural network (CAclassNet) was designed and implemented, consisting of five 2D convolutional layers, three fully connected layers and a final classifier returning AL, ATTR and CTRL scores. A total of 1107 2D images (375 from AL-subtype patients, 312 from ATTR-subtype, and 420 from Controls) have been considered in the study and used to train, validate and test the proposed network. CAclassNet cross-validation resulted with train error mean ± sd of 2.001% ± 0.96%, validation error of 4.5% ± 2.26%, and net accuracy of 95.49% ± 2.26%. Network test error resulted in a mean ± sd values of 10.73% ± 0.76%. Sensitivity, specificity, and accuracy evaluated on the test dataset were respectively for AL-CA sub-type: 1, 0.912, 0.936; for ATTR-CA: 0.935, 0.897, 0.972; for control subjects: 0.809, 0.971, 0.909. In conclusion, the proposed CAclassNet model seems very promising as an aid for the clinician in the diagnosis of CA from cardiac [18F]-Florbetaben PET images acquired a few minutes after the injection.

[18F]-florbetaben PET/CT Imaging in the Alzheimer's Disease Mouse Model APPswe/PS1dE9

BACKGROUND

Positron-emission-tomography (PET) using 18F labeled florbetaben allows noninvasive in vivo-assessment of amyloid-beta (Aβ), a pathological hallmark of Alzheimer's disease (AD). In preclinical research, [¹⁸F]-florbetaben-PET has already been used to test the amyloid-lowering potential of new drugs, both in humans and in transgenic models of cerebral amyloidosis. The aim of this study was to characterize the spatial pattern of cerebral uptake of [¹⁸F]-florbetaben in the APPswe/ PS1dE9 mouse model of AD in comparison to histologically determined number and size of cerebral Aβ plaques.

METHODS

Both, APPswe/PS1dE9 and wild type mice at an age of 12 months were investigated by smallanimal PET/CT after intravenous injection of [¹⁸F]-florbetaben. High-resolution magnetic resonance imaging data were used for quantification of the PET data by volume of interest analysis. The standardized uptake values (SUVs) of [¹⁸F]-florbetaben in vivo as well as post mortem cerebral Aβ plaque load in cortex, hippocampus and cerebellum were analyzed.

RESULTS

Visual inspection and SUVs revealed an increased cerebral uptake of [¹⁸F]-florbetaben in APPswe/ PS1dE9 mice compared with wild type mice especially in the cortex, the hippocampus and the cerebellum. However, SUV ratios (SUVRs) relative to cerebellum revealed only significant differences in the hippocampus between the APPswe/PS1dE9 and wild type mice but not in cortex; this differential effect may reflect the lower plaque area in the cortex than in the hippocampus as found in the histological analysis.

CONCLUSION

The findings suggest that histopathological characteristics of Aβ plaque size and spatial distribution can be depicted in vivo using [¹⁸F]-florbetaben in the APPswe/PS1dE9 mouse model.

PET-CT evaluation of amyloid systemic involvement with [18F]-florbetaben in patient with proved cardiac amyloidosis: a case report

BACKGROUND

One of the most frequent disorders which lead to cardiac amyloidosis is transthyretin-related amyloidosis (ATTR). Some PET radio-pharmaceuticals for the detection of beta-amyloid deposits within the brain have shown to be able to detect also cardiac amyloid deposits. We present a case of a man with ATTR studied with [18F]-florbetaben PET-CT.

RESULTS

Total-body scan showed a moderate uptake in the bone marrow, especially in correspondence of the vertebral column, while no significant myocardial uptake was present. Cardiac-focused scans showed low mean cardiac SUV values confirming the absence of significant myocardial uptake. Brain scan showed a significant cortical brain uptake of the radio-pharmaceutical more evident in correspondence of frontal and temporal lobes.

CONCLUSIONS

Distinct subtypes of amyloidosis show different uptake of the radiotracer. Brain amyloid deposition in the presence of a systemic disease could not be caused by the same amyloid precursor.

Automated Spatial Brain Normalization and Hindbrain White Matter Reference Tissue Give Improved [(18)F]-Florbetaben PET Quantitation in Alzheimer's Model Mice

Preclinical PET studies of β-amyloid (Aβ) accumulation are of growing importance, but comparisons between research sites require standardized and optimized methods for quantitation. Therefore, we aimed to evaluate systematically the (1) impact of an automated algorithm for spatial brain normalization, and (2) intensity scaling methods of different reference regions for Aβ-PET in a large dataset of transgenic mice. PS2APP mice in a 6 week longitudinal setting (N = 37) and another set of PS2APP mice at a histologically assessed narrow range of Aβ burden (N = 40) were investigated by [¹⁸F]-florbetaben PET. Manual spatial normalization by three readers at different training levels was performed prior to application of an automated brain spatial normalization and inter-reader agreement was assessed by Fleiss Kappa (κ). For this method the impact of templates at different pathology stages was investigated. Four different reference regions on brain uptake normalization were used to calculate frontal cortical standardized uptake value ratios (SUVR_CTX∕REF), relative to raw SUV_CTX. Results were compared on the basis of longitudinal stability (Cohen's d), and in reference to gold standard histopathological quantitation (Pearson's R). Application of an automated brain spatial normalization resulted in nearly perfect agreement (all κ≥0.99) between different readers, with constant or improved correlation with histology. Templates based on inappropriate pathology stage resulted in up to 2.9% systematic bias for SUVR_CTX∕REF. All SUVR_CTX∕REF methods performed better than SUV_CTX both with regard to longitudinal stability (d≥1.21 vs. d = 0.23) and histological gold standard agreement (R≥0.66 vs. R≥0.31). Voxel-wise analysis suggested a physiologically implausible longitudinal decrease by global mean scaling. The hindbrain white matter reference (R_mean = 0.75) was slightly superior to the brainstem (R_mean = 0.74) and the cerebellum (R_mean = 0.73). Automated brain normalization with reference region templates presents an excellent method to avoid the inter-reader variability in preclinical Aβ-PET scans. Intracerebral reference regions lacking Aβ pathology serve for precise longitudinal in vivo quantification of [¹⁸F]-florbetaben PET. Hindbrain white matter reference performed best when considering the composite of quality criteria.