Reducing instability of inter-subject covariance of FDG uptake networks using structure-weighted sparse estimation approach.

PURPOSE: Sparse inverse covariance estimation (SICE) is increasingly utilized to estimate inter-subject covariance of FDG uptake (FDGcov) as proxy of metabolic brain connectivity. However, this statistical method suffers from the lack of robustness in the connectivity estimation. Patterns of FDGcov were observed to be spatially similar with patterns of structural connectivity as obtained from DTI imaging. Based on this similarity, we propose to regularize the sparse estimation of FDGcov using the structural connectivity. METHODS: We retrospectively analyzed the FDG-PET and DTI data of 26 healthy controls, 41 patients with Alzheimer's disease (AD), and 30 patients with frontotemporal lobar degeneration (FTLD). Structural connectivity matrix derived from DTI data was introduced as a regularization parameter to assign individual penalties to each potential metabolic connectivity. Leave-one-out cross validation experiments were performed to assess the differential diagnosis ability of structure weighted SICE approach. A few approaches of structure weighted were compared with the standard SICE. RESULTS: Compared to the standard SICE, structural weighting has shown more stable performance in the supervised classification, especially in the differentiation AD vs. FTLD (accuracy of 89-90%, while unweighted SICE only 85%). There was a significant positive relationship between the minimum number of metabolic connection and the robustness of the classification accuracy (r = 0.57, P < 0.001). Shuffling experiments showed significant differences between classification score derived with true structural weighting and those obtained by randomized structure (P < 0.05). CONCLUSION: The structure-weighted sparse estimation can enhance the robustness of metabolic connectivity, which may consequently improve the differentiation of pathological phenotypes.

Hypoxia-specific tumor imaging with 18F-fluoroazomycin arabinoside.

UNLABELLED: The study was performed to compare the (18)F-labeled nitroimidazole compound fluoroazomycin arabinoside ((18)F-FAZA) with the standard hypoxia tracer fluoromisonidazole ((18)F-FMISO) in detection of tumor tissue hypoxia and to verify the oxygenation dependency of (18)F-FAZA uptake. METHODS: Biodistribution of (18)F-FAZA was studied at various time points in EMT6 tumor-bearing BALB/c mice and in AR42J and A431 tumor-bearing nude mice and compared with that of (18)F-FMISO. The presence of tumor tissue hypoxia was verified in 5 EMT6 and 5 AR42J tumors using an oxygen-sensing needle electrode system. To evaluate the oxygenation dependency of (18)F-FAZA uptake, using the Munich prototype animal PET scanner, 2 serial PET scans were performed in 13 A431 tumor-bearing nude mice breathing pure oxygen or room air on 1 d and then selecting the other oxygen breathing condition on the following day. In addition, digital autoradiography was performed with EMT6 tumor-bearing (18)F-FAZA-dosed, nude mice breathing either room air (n = 8) or carbogen (n = 9). RESULTS: Tissue partial pressure of oxygen (Po(2)) electrode measurements revealed that tumor hypoxia was present under room air breathing in EMT6 (tissue Po(2) = 2.9 +/- 2.6) and AR42J tumors (tissue Po(2) = 0.4 +/- 0.2), which was significantly lower compared with that of reference tissue (tissue Po(2) = 25.8 +/- 6.7 and tissue Po(2) = 29.0 +/- 3.0 [mean +/- SD], respectively; P < 0.01). In all tumor models, (18)F-FAZA displayed significantly higher tumor-to-muscle and tumor-to-blood ratios compared with (18)F-FMISO, indicating a faster clearance of (18)F-FAZA from normal tissues. In AR42J tumors, (18)F-FAZA tumor-to-normal ratios were found to increase over time. Serial animal (18)F-FAZA PET studies showed that the tumor-to-background ratio was significantly higher in animals breathing room air compared with that of animals breathing pure oxygen (7.3 +/- 2.3 vs. 4.2 +/- 1.2, respectively; P < 0.001). Similarly, autoradiography showed significantly higher tumor-to-muscle ratios in mice breathing room air compared with those of animals breathing carbogen (5.3 +/- 0.8 vs. 2.2 +/- 0.8; respectively; P < 0.02). CONCLUSION: (18)F-FAZA shows superior biokinetics and is, thus, a promising PET tracer for the visualization of tumor hypoxia. This study also verified a hypoxia-specific uptake mechanism for (18)F-FAZA in murine tumor models.

Evaluation of 18F-fluoride PET/MR and PET/CT in patients with foot pain of unclear cause.

UNLABELLED: Our objective was to compare the quality and diagnostic performance of (18)F-fluoride PET/MR imaging with that of (18)F-fluoride PET/CT imaging in patients with foot pain of unclear cause. METHODS: Twenty-two patients (9 men, 13 women; mean age, 48 ± 18 y; range, 20-78 y) were prospectively included in this study and underwent a single-injection dual-imaging protocol with (18)F-fluoride PET/CT and PET/MR. At a minimum, the PET/MR protocol included T1-weighted spin echo and proton-density fat-saturated sequences in 2 planes each with simultaneous acquisition of PET over 20 min. PET/CT included a native isotropic (0.6 mm) diagnostic CT scan (80 kV, 165 mAs) and a subsequent PET scan (2 min per bed position). By consensus, 2 masked interpreters randomly assessed both PET datasets for image quality (3-point scale) and for the presence of focal lesions with increased (18)F-fluoride uptake (maximum of 4 lesions). For each dataset (PET/CT vs. PET/MR), the diagnoses were defined using both PET and a morphologic dataset. Standardized uptake values (SUVs) from the 2 devices were compared using linear correlation and Bland-Altman plots. Moreover, we estimated the potential for dose reduction for PET/MR compared with PET/CT considering the longer acquisition time of PET/MR analyzing count rate statistics. RESULTS: Image quality was rated diagnostic for both PET datasets. However, with a mean rating of 3.0/3 for PET/MR and 2.3/3 for PET/CT, image quality was significantly superior for PET/MR (P < 0.0001). The sensitivity of the PET datasets in PET/MR and PET/CT was equivalent, with the same 42 lesions showing focal (18)F-fluoride uptake. In PET/MR, the mean SUVmean was 10.4 (range, 2.0-67.7) and the mean SUVmax was 15.6 (range, 2.9-94.1). In PET/CT, the corresponding mean SUVmean of PET/CT was 10.2 (range, 1.8-55.6) and the mean SUVmax was 16.3 (range, 2.5-117.5), resulting in a high linear correlation coefficient (r = 0.96, P < 0.0001, for SUVmean and for SUVmax). A final consensus interpretation revealed the most frequent main diagnoses to be osteoarthritis, stress fracture, and bone marrow edema. PET/CT was more precise in visualizing osteoarthritis, whereas PET/MR was more specific in nondegenerative pathologies because of the higher soft-tissue and bone marrow contrast. The longer acquisition time of MR compared with CT would potentially allow (18)F-fluoride dose reduction using hybrid (18)F-fluoride PET/MR imaging of at least 50% according to the counting rate analysis. CONCLUSION: In patients with foot pain of unclear cause, (18)F-fluoride PET/MR is technically feasible and is more robust in terms of image quality and SUV quantification than (18)F-fluoride PET/CT. In most patients, (18)F-fluoride PET/MR provided more diagnostic information at a higher diagnostic certainty than did PET/CT. Thus, PET/MR combines the high sensitivity of (18)F-fluoride PET to pinpoint areas with the dominant disease activity and the specificity of MR imaging for the final diagnosis with the potential for a substantial dose reduction compared with PET/CT.

Evaluation of feasibility and image quality of 68Ga-DOTATOC positron emission tomography/magnetic resonance in comparison with positron emission tomography/computed tomography in patients with neuroendocrine tumors.

OBJECTIVES: The primary aims of this study were to evaluate the feasibility of simultaneous 68(DOTA(0)-Phe(1)-Tyr(3))octreotide positron emission tomography (PET)/magnetic resonance (MR) acquisition on a fully integrated PET/MR scanner in patients and to compare the quality of PET images acquired with a PET/MR device with those acquired with a PET/computed tomography (CT) scanner. PATIENTS AND METHODS: Sequential PET/CT and PET/MR imaging was performed in 24 patients with neuroendocrine tumors using a single-injection/dual-imaging protocol. After intravenous injection of 68Ga-DOTATOC (mean, 120 MBq), PET/CT imaging including low-dose CT was performed at a mean time of 17 minutes post injection, and subsequently, PET/MR imaging including a Dixon sequence for attenuation correction was started at a mean time of 82 minutes post injection. The PET/CT and PET/MR images were analyzed visually using a 4-point scale for quality, coregistration, anatomical correlation, and lesion conspicuity. The standardized uptake value of background organs and focal lesions was measured and compared between the PET/CT and PET/MR acquisitions. RESULTS: 68Ga-DOTATOC PET acquired on the PET/MR delivered images with a good diagnostic quality (average visual rating PET/CT, 2.83; PET/MR, 2.08; P <; 0.01). The standardized uptake value of focal lesions did not differ between the PET/CT and PET/MR acquisitions (P >; 0.3) and correlated in a linear fashion (correlation coefficient ρ = 0.90). Lesion conspicuity was slightly, but significantly, higher on the PET/CT acquisitions (PET/CT, 2.71; PET/MR, 2.62; P = 0.01). Positron emission tomography/MR detected 153 of 157 lesions identified by PET/CT; however, there was no difference in sensitivity on a patient basis or organ system basis. Anatomical correlates for focal PET lesions could significantly more often be delineated using MR Dixon images compared with low-dose CT (average visual rating PET/CT, 1.78; PET/MR, 2.30; P <; 0.01). Coregistration of functional and morphological data was better on PET/MR compared with PET/CT, which, however, did not reach significance (average visual rating PET/CT, 2.17; PET/MR, 2.46; P = 0.10). CONCLUSIONS: 68Ga-DOTATOC PET/MR imaging is feasible in patients, with good image quality, and detectability of focal PET lesions was equivalent to PET/CT on a patient basis and organ system basis. Now, the clinical value of 68Ga-DOTATOC PET/MR with additional diagnostic MR protocols has to be evaluated against PET/CT with multiphase contrast-enhanced CT protocols in future studies.

A case of multimodality multiparametric 11C-choline PET/MR for biopsy targeting in prior biopsy-negative primary prostate cancer.

We introduce the case of a 70-years-old man with an elevated level of prostate-specific antigen and prior negative biopsy. For targeting rebiopsy, the patient underwent C-choline PET/CT and subsequent PET/MR. Both the high uptake in PET and the abnormal findings in MR gave strong evidence for prostate cancer at the ventral periphery of the right apex. This location is sometimes not covered by routine sextant biopsy. The following targeted rebiopsy was positive for prostate cancer. This case indicates the potential role of PET/MR for identifying primary prostate cancer because of its high soft tissue contrast and the possibility of a multimodality approach.