Predicting FDG-PET Images From Multi-Contrast MRI Using Deep Learning in Patients With Brain Neoplasms

BACKGROUND

18 F-fluorodeoxyglucose (FDG) positron emission tomography (PET) is valuable for determining presence of viable tumor, but is limited by geographical restrictions, radiation exposure, and high cost.

PURPOSE

To generate diagnostic-quality PET equivalent imaging for patients with brain neoplasms by deep learning with multi-contrast MRI.

STUDY TYPE

Retrospective.

SUBJECTS

Patients (59 studies from 51 subjects; age 56 ± 13 years; 29 males) who underwent 18 F-FDG PET and MRI for determining recurrent brain tumor.

FIELD STRENGTH/SEQUENCE

3T; 3D GRE T1, 3D GRE T1c, 3D FSE T2-FLAIR, and 3D FSE ASL, 18 F-FDG PET imaging.

ASSESSMENT

Convolutional neural networks were trained using four MRIs as inputs and acquired FDG PET images as output. The agreement between the acquired and synthesized PET was evaluated by quality metrics and Bland-Altman plots for standardized uptake value ratio. Three physicians scored image quality on a 5-point scale, with score ≥3 as high-quality. They assessed the lesions on a 5-point scale, which was binarized to analyze diagnostic consistency of the synthesized PET compared to the acquired PET.

STATISTICAL TESTS

The agreement in ratings between the acquired and synthesized PET were tested with Gwet's AC and exact Bowker test of symmetry. Agreement of the readers was assessed by Gwet's AC. P = 0.05 was used as the cutoff for statistical significance.

RESULTS

The synthesized PET visually resembled the acquired PET and showed significant improvement in quality metrics (+21.7% on PSNR, +22.2% on SSIM, -31.8% on RSME) compared with ASL. A total of 49.7% of the synthesized PET were considered as high-quality compared to 73.4% of the acquired PET which was statistically significant, but with distinct variability between readers. For the positive/negative lesion assessment, the synthesized PET had an accuracy of 87% but had a tendency to overcall.

CONCLUSION

The proposed deep learning model has the potential of synthesizing diagnostic quality FDG PET images without the use of radiotracers.

EVIDENCE LEVEL

3 TECHNICAL EFFICACY: Stage 2.

Feature-based Quality Assessment of Middle Cerebral Artery Occlusion Using 18F-Fluorodeoxyglucose Positron Emission Tomography

In animal experiments, ischemic stroke is usually induced through middle cerebral artery occlusion (MCAO), and quality assessment of this procedure is crucial. However, an accurate assessment method based on 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET) is still lacking. The difficulty lies in the inconsistent preprocessing pipeline, biased intensity normalization, or unclear spatiotemporal uptake of FDG. Here, we propose an image feature-based protocol to assess the quality of the procedure using a 3D scale-invariant feature transform and support vector machine. This feature-based protocol provides a convenient, accurate, and reliable tool to assess the quality of the MCAO procedure in FDG PET studies. Compared with existing approaches, the proposed protocol is fully quantitative, objective, automatic, and bypasses the intensity normalization step. An online interface was constructed to check images and obtain assessment results.

FDG PET imaging in multiple myeloma: implications for response assessments in clinical trials

¹⁸F-fluorodeoxyglucose positron emission tomography integrated with computed tomography (¹⁸F-FDG PET/CT) is an effective modality to assess disease burden, detect extra-medullary disease and monitor minimal residual disease (MRD) for multiple myeloma (MM) patients. This modality of imaging is incorporated in the International Myeloma Working Group (IMWG) response criteria that are widely used in MM clinical trials. Interpretative pitfalls are commonly encountered in ¹⁸F-FDG PET/CT studies and proper interpretation requires knowledge of the normal physiologic distribution of the tracer affecting available ¹⁸F-FDG for tumor tissue uptake. We describe a series of MM patients who exhibited a deep response to treatment, based on clinical features, serum markers and bone marrow (BM) biopsy. However, these patients seemed to have new lesions on post-therapy ¹⁸F-FDG PET/CT images which could be interpreted as progressive disease according to the IMWG criteria. Sequestration phenomenon, which is the disappearance of ¹⁸F-FDG sequestration by myeloma-infiltrated marrow after successful anti-myeloma therapy, could lead to unmasking of new ¹⁸F-FDG-avid lesions on post-therapy PET/CT due to higher ¹⁸F-FDG bioavailability to residual tumor tissue. Clinical correlation, awareness of the ¹⁸F-FDG sequestration in myeloma infiltrated BM and its impact on other ¹⁸F-FDG avid areas in the body are necessary to avoid potential pitfalls in end-of-treatment imaging interpretation. While considering patients for clinical trials, clinicians should be mindful of this sequestration phenomenon in the interpretation of post-therapy PET/CT imaging in MM patients with initially heavily infiltrated marrow.

Evolution of neurodegeneration-imaging biomarkers from clinically normal to dementia in the Alzheimer disease spectrum

The availability of antemortem biomarkers for Alzheimer's disease (AD) enables monitoring the evolution of neurodegenerative processes in real time. Pittsburgh compound B (PIB) positron emission tomography (PET) was used to select participants in the Mayo Clinic Study of Aging and the Mayo Alzheimer's Disease Research Center with elevated β-amyloid, designated as "A+," and hippocampal volume and (18)fluorodeoxyglucose (FDG) positron emission tomography were used to characterize participants as having evidence of neurodegeneration ("N+") at the baseline evaluation. There were 145 clinically normal (CN) A+ individuals, 62 persons with mild cognitive impairment (MCI) who were A+ and 20 with A+ AD dementia. Over a period of 1-6 years, MCI A+N+ individuals showed declines in medial temporal, lateral temporal, lateral parietal, and to a lesser extent, medial parietal regions for both FDG standardized uptake value ratio and gray matter volume that exceeded declines seen in the CN A+N+ group. The AD dementia group showed declines in the same regions on FDG standardized uptake value ratio and gray matter volume with rates that exceeded that in MCI A+N+. Expansion of regional involvement and faster rate of neurodegeneration characterizes progression in the AD pathway.

Effects of a novel phosphodiesterase 10A inhibitor in non-human primates: A therapeutic approach for schizophrenia with improved side effect profile

Schizophrenia symptoms are associated with alterations in basal ganglia-cortical networks that include the cyclic nucleotides (cAMP/cGMP) signaling pathways. Phosphodiesterase 10A (PDE10A) inhibitors have been considered as therapeutic agents for schizophrenia because the regulation of cAMP and cGMP in the striatum by PDE10A plays an important role in the signaling mechanisms of the striatal-cortical network, and thereby in cognitive function. In the present study we assessed in non-human primates (NHPs) the effects of a novel PDE10A inhibitor (FRM-6308) that has demonstrated high potency and selectivity for human recombinant PDE10A in vitro. The behavioral effects of FRM-6308 in a dose range were determined in rhesus monkeys using a standardized motor disability scale for primates, motor tasks, and the "drug effects on the nervous system" (DENS) scale. The neuronal metabolic effects of FRM-6308 were determined with [(18)F]-fluorodeoxyglucose PET imaging. Results showed that FRM-6308 did not have any specific effects on the motor system at s.c. doses up to 0.32 mg/kg in NHPs, which induced a significant increase in the FDG-SUV in striatum (F 16.069, p < 0.05) and cortical (F 15.181, p < 0.05) regions. Higher doses induced sedation and occasional involuntary movements with clear development of tolerance after repeated exposures. These findings suggest that FRM-6308 has the adequate pharmacological profile to advance testing in clinical trials and demonstrate antipsychotic efficacy of PDE10A inhibition for the treatment of schizophrenia patients.

Advances in imaging for diagnosis and management of cardiac sarcoidosis

Sarcoidosis is a multisystem granulomatous disorder with a high prevalence of cardiac involvement. Cardiac sarcoidosis (CS) may be life threatening due to end-stage cardiomyopathy and sudden cardiac death. The frequent absence of specific symptoms and lack of a diagnostic 'gold standard' pose challenges in the diagnosis of CS. Endomyocardial biopsy, although specific, has an unacceptably low sensitivity. Non-invasive cardiac imaging has a huge role in the assessment of patients with known or suspected CS. This comprehensive review compares the diagnostic accuracy, along with advantages and disadvantages, of established and emerging imaging modalities for CS.