Longitudinal changes in metabolic network activity in early Alzheimer's disease

INTRODUCTION

The progression of Alzheimer's disease (AD) has been linked to two metabolic networks, the AD-related pattern (ADRP) and the default mode network (DMN).

METHODS

Converting and clinically stable cognitively normal subjects (n = 47) and individuals with mild cognitive impairment (n = 96) underwent 2-[18 F]fluoro-2-deoxy-D-glucose (FDG) positron emission tomography (PET) three or more times over 6 years (nscans  = 705). Expression levels for ADRP and DMN were measured in each subject and time point, and the resulting changes were correlated with cognitive performance. The role of network expression in predicting conversion to dementia was also evaluated.

RESULTS

Longitudinal increases in ADRP expression were observed in converters, while age-related DMN loss was seen in converters and nonconverters. Cognitive decline correlated with increases in ADRP and declines in DMN, but conversion to dementia was predicted only by baseline ADRP levels.

DISCUSSION

The results point to the potential utility of ADRP as an imaging biomarker of AD progression.

Synthetic Attenuation Correction Maps for SPECT Imaging Using Deep Learning: A Study on Myocardial Perfusion Imaging

(1) Background: The CT-based attenuation correction of SPECT images is essential for obtaining accurate quantitative images in cardiovascular imaging. However, there are still many SPECT cameras without associated CT scanners throughout the world, especially in developing countries. Performing additional CT scans implies troublesome planning logistics and larger radiation doses for patients, making it a suboptimal solution. Deep learning (DL) offers a revolutionary way to generate complementary images for individual patients at a large scale. Hence, we aimed to generate linear attenuation coefficient maps from SPECT emission images reconstructed without attenuation correction using deep learning. (2) Methods: A total of 384 SPECT myocardial perfusion studies that used ^99mTc-sestamibi were included. A DL model based on a 2D U-Net architecture was trained using information from 312 patients. The quality of the generated synthetic attenuation correction maps (ACMs) and reconstructed emission values were evaluated using three metrics and compared to standard-of-care data using Bland-Altman plots. Finally, a quantitative evaluation of myocardial uptake was performed, followed by a semi-quantitative evaluation of myocardial perfusion. (3) Results: In a test set of 66 test patients, the ACM quality metrics were MSSIM = 0.97 ± 0.001 and NMAE = 3.08 ± 1.26 (%), and the reconstructed emission quality metrics were MSSIM = 0.99 ± 0.003 and NMAE = 0.23 ± 0.13 (%). The 95% limits of agreement (LoAs) at the voxel level for reconstructed SPECT images were: [-9.04; 9.00]%, and for the segment level, they were [-11; 10]%. The 95% LoAs for the Summed Stress Score values between the images reconstructed were [-2.8, 3.0]. When global perfusion scores were assessed, only 2 out of 66 patients showed changes in perfusion categories. (4) Conclusion: Deep learning can generate accurate attenuation correction maps from non-attenuation-corrected cardiac SPECT images. These high-quality attenuation maps are suitable for attenuation correction in myocardial perfusion SPECT imaging and could obviate the need for additional imaging in standalone SPECT scanners.

Development of a deep learning network for Alzheimer’s disease classification with evaluation of imaging modality and longitudinal data

Objective. Neuroimaging uncovers important information about disease in the brain. Yet in Alzheimer’s disease (AD), there remains a clear clinical need for reliable tools to extract diagnoses from neuroimages. Significant work has been done to develop deep learning (DL) networks using neuroimaging for AD diagnosis. However, no particular model has emerged as optimal. Due to a lack of direct comparisons and evaluations on independent data, there is no consensus on which modality is best for diagnostic models or whether longitudinal information enhances performance. The purpose of this work was (1) to develop a generalizable DL model to distinguish neuroimaging scans of AD patients from controls and (2) to evaluate the influence of imaging modality and longitudinal data on performance. Approach. We trained a 2-class convolutional neural network (CNN) with and without a cascaded recurrent neural network (RNN). We used datasets of 772 (N AD = 364, N control = 408) 3D 18F-FDG PET scans and 780 (N AD = 280, N control = 500) T1-weighted volumetric-3D MR images (containing 131 and 144 patients with multiple timepoints) from the Alzheimer’s Disease Neuroimaging Initiative, plus an independent set of 104 (N AD = 63, N NC = 41) 18F-FDG PET scans (one per patient) for validation. Main Results. ROC analysis showed that PET-trained models outperformed MRI-trained, achieving maximum AUC with the CNN + RNN model of 0.93 ± 0.08, with accuracy 82.5 ± 8.9%. Adding longitudinal information offered significant improvement to performance on 18F-FDG PET, but not on T1-MRI. CNN model validation with an independent 18F-FDG PET dataset achieved AUC of 0.99. Layer-wise relevance propagation heatmaps added CNN interpretability. Significance. The development of a high-performing tool for AD diagnosis, with the direct evaluation of key influences, reveals the advantage of using 18F-FDG PET and longitudinal data over MRI and single timepoint analysis. This has significant implications for the potential of neuroimaging for future research on AD diagnosis and clinical management of suspected AD patients.

Image intensity histograms as imaging biomarkers: application to immune-related colitis

Purpose.To investigate image intensity histograms as a potential source of useful imaging biomarkers in both a clinical example of detecting immune-related colitis (irColitis) in¹⁸F-FDG PET/CT images of immunotherapy patients and an idealized case of classifying digital reference objects (DRO).Methods.Retrospective analysis of bowel¹⁸F-FDG uptake in N = 40 patients receiving immune checkpoint inhibitors was conducted. A CNN trained to segment the bowel was used to generate the histogram of bowel¹⁸F-FDG uptake, and percentiles of the histogram were considered as potential metrics for detecting inflammation associated with irColitis. A model of the colon was also considered using cylindrical DRO. Classification of DRO with different intensity distributions was undertaken under varying geometry and noise settings.Results.The most predictive biomarker of irColitis was the 95th percentile of the bowel SUV histogram (SUV_95%). Patients later diagnosed with irColitis had a significantly higher increase in SUV_95%from baseline to first on-treatment PET than patients who did not experience irColitis (p = 0.02). An increase in SUV_95%> + 40% separated pre-irColitis change from normal variability with a sensitivity of 75% and specificity of 88%. Furthermore, histogram percentiles were ideal metrics for classifying 'hot center' and 'cold center' DRO, and were robust to varying DRO geometry and noise, and to the presence of spoiler volumes unrelated to the detection task.Conclusions.The 95th percentile of the bowel SUV histogram was the optimal metric for detecting irColitis on¹⁸F-FDG PET/CT. Image intensity histograms are a promising source of imaging biomarkers for clinical tasks.

Optimization of oncological ¹⁸F-FDG PET/CT imaging based on a multiparameter analysis

PURPOSE

This paper describes a method to achieve consistent clinical image quality in (18)F-FDG scans accounting for patient habitus, dose regimen, image acquisition, and processing techniques.

METHODS

Oncological PET/CT scan data for 58 subjects were evaluated retrospectively to derive analytical curves that predict image quality. Patient noise equivalent count rate and coefficient of variation (CV) were used as metrics in their analysis. Optimized acquisition protocols were identified and prospectively applied to 179 subjects.

RESULTS

The adoption of different schemes for three body mass ranges (<60 kg, 60-90 kg, >90 kg) allows improved image quality with both point spread function and ordered-subsets expectation maximization-3D reconstruction methods. The application of this methodology showed that CV improved significantly (p < 0.0001) in clinical practice.

CONCLUSIONS

Consistent oncological PET/CT image quality on a high-performance scanner was achieved from an analysis of the relations existing between dose regimen, patient habitus, acquisition, and processing techniques. The proposed methodology may be used by PET/CT centers to develop protocols to standardize PET/CT imaging procedures and achieve better patient management and cost-effective operations.

Characterization of Elastofibroma Dorsi with (18)FDG PET/CT: a retrospective study

Elastofibroma dorsi has been described in the literature as an unusual tumor or pseudotumor. However, autopsies and imaging studies have revealed that it is a non-negligible finding.

PURPOSE

The aim of this study has been to illustrate and become familiar with this type of lesion in order to prevent misdiagnosis.

MATERIALS AND METHODS

From 3 December 2008 to 5 January 2010, 1,751 patients were evaluated with (18)FDG-PET/CT. Of these, 29 cases of elastofibroma dorsi were recorded as an incidental finding. A retrospective and descriptive analysis was performed on this study series.

RESULTS

The study showed a prevalence of 1.66%. Out of the 29 findings, 22 (75.86%) were females and 7 (24.14%) males. Seventeen (58.62%) cases were bilateral, 12 (41.38%) unilateral and the SUVmax ranged from 1.4 to 3.2. These lesions were reported as soft tissue density images with mild or moderate diffuse metabolic activity.

CONCLUSION

The elastofibroma dorsi is a relatively common finding in PET/CT that should be known in order to avoid making wrong diagnoses.