Deriving accurate attenuation maps for PET/MRI remains a challenging problem because MRI voxel intensities are not related to properties of photon attenuation and bone/air interfaces have similarly low signal. This work presents a learning-based method to derive patient-specific computed tomography (CT) maps from routine T1-weighted MRI in their native space for attenuation correction of brain PET. We developed a machine-learning-based method using a sequence of alternating random forests under the framework of an iterative refinement model. Anatomical feature selection is included in both training and predication stages to achieve optimal performance. To evaluate its accuracy, we retrospectively investigated 17 patients, each of which has been scanned by PET/CT and MR for brain. The PET images were corrected for attenuation on CT images as ground truth, as well as on pseudo CT (PCT) images generated from MR images. The PCT images showed mean average error of 66.1 ± 8.5 HU, average correlation coefficient of 0.974 ± 0.018 and average Dice similarity coefficient (DSC) larger than 0.85 for air, bone and soft tissue. The side-by-side image comparisons and joint histograms demonstrated very good agreement of PET images after correction by PCT and CT. The mean differences of voxel values in selected VOIs were less than 4%, the mean absolute difference of all active area is around 2.5%, and the mean linear correlation coefficient is 0.989 ± 0.017 between PET images corrected by CT and PCT. This work demonstrates a novel learning-based approach to automatically generate CT images from routine T1-weighted MR images based on a random forest regression with patch-based anatomical signatures to effectively capture the relationship between the CT and MR images. Reconstructed PET images using the PCT exhibit errors well below accepted test/retest reliability of PET/CT indicating high quantitative equivalence.

Registration and fusion of magnetic resonance imaging (MRI) and transrectal ultrasound (TRUS) of the prostate can provide guidance for prostate brachytherapy. However, accurate registration remains a challenging task due to the lack of ground truth regarding voxel-level spatial correspondence, limited field of view, low contrast-to-noise ratio, and signal-to-noise ratio in TRUS. In this study, we proposed a fully automated deep learning approach based on a weakly supervised method to address these issues. We employed deep learning techniques to combine image segmentation and registration, including affine and nonrigid registration, to perform an automated deformable MRI-TRUS registration. To start with, we trained two separate fully convolutional neural networks (CNNs) to perform a pixel-wise prediction for MRI and TRUS prostate segmentation. Then, to provide the initialization of the registration, a 2D CNN was used to register MRI-TRUS prostate images using an affine registration. After that, a 3D UNET-like network was applied for nonrigid registration. For both the affine and nonrigid registration, pairs of MRI-TRUS labels were concatenated and fed into the neural networks for training. Due to the unavailability of ground-truth voxel-level correspondences and the lack of accurate intensity-based image similarity measures, we propose to use prostate label-derived volume overlaps and surface agreements as an optimization objective function for weakly supervised network training. Specifically, we proposed a hybrid loss function that integrated a Dice loss, a surface-based loss, and a bending energy regularization loss for the nonrigid registration. The Dice and surface-based losses were used to encourage the alignment of the prostate label between the MRI and the TRUS. The bending energy regularization loss was used to achieve a smooth deformation field. Thirty-six sets of patient data were used to test our registration method. The image registration results showed that the deformed MR image aligned well with the TRUS image, as judged by corresponding cysts and calcifications in the prostate. The quantitative results showed that our method produced a mean target registration error (TRE) of 2.53 1.39 mm and a mean Dice loss of 0.91 0.02. The mean surface distance (MSD) and Hausdorff distance (HD) between the registered MR prostate shape and TRUS prostate shape were 0.88 and 4.41 mm, respectively. This work presents a deep learning-based, weakly supervised network for accurate MRI-TRUS image registration. Our proposed method has achieved promising registration performance in terms of Dice loss, TRE, MSD, and HD.

Lowering either the administered activity or scan time is desirable in PET imaging as it decreases the patient's radiation burden or improves patient comfort and reduces motion artifacts. But reducing these parameters lowers overall photon counts and increases noise, adversely impacting image contrast and quantification. To address this low count statistics problem, we propose a cycle-consistent generative adversarial network (Cycle GAN) model to estimate diagnostic quality PET images using low count data. Cycle GAN learns a transformation to synthesize diagnostic PET images using low count data that would be indistinguishable from our standard clinical protocol. The algorithm also learns an inverse transformation such that cycle low count PET data (inverse of synthetic estimate) generated from synthetic full count PET is close to the true low count PET. We introduced residual blocks into the generator to catch the differences between low count and full count PET in the training dataset and better handle noise. The average mean error and normalized mean square error in whole body were-0.14% ± 1.43% and 0.52% ± 0.19% with Cycle GAN model, compared to 5.59% ± 2.11% and 3.51% ± 4.14% on the original low count PET images. Normalized cross-correlation is improved from 0.970 to 0.996, and peak signal-to-noise ratio is increased from 39.4 dB to 46.0 dB with proposed method. We developed a deep learning-based approach to accurately estimate diagnostic quality PET datasets from one eighth of photon counts, and has great potential to improve low count PET image quality to the level of diagnostic PET used in clinical settings.

Attenuation correction (AC) of PET/MRI faces challenges including inter-scan motion, image artifacts such as truncation and distortion, and erroneous transformation of structural voxel-intensities to PET mu-map values. We propose a deep-learning-based method to derive synthetic CT (sCT) images from non-attenuation corrected PET (NAC PET) images for AC on whole-body PET/MRI imaging. A 3D cycle-consistent generative adversarial networks (CycleGAN) framework was employed to synthesize CT images from NAC PET. The method learns a transformation that minimizes the difference between sCT, generated from NAC PET, and true CT. It also learns an inverse transformation such that cycle NAC PET image generated from the sCT is close to true NAC PET image. A self-attention strategy was also utilized to identify the most informative component and mitigate the disturbance of noise. We conducted a retrospective study on a total of 119 sets of whole-body PET/CT, with 80 sets for training and 39 sets for testing and evaluation. The whole-body sCT images generated with proposed method demonstrate great resemblance to true CT images, and show good contrast on soft tissue, lung and bony tissues. The mean absolute error (MAE) of sCT over true CT is less than 110 HU. Using sCT for whole-body PET AC, the mean error of PET quantification is less than 1% and normalized mean square error (NMSE) is less than 1.4%. Average normalized cross correlation on whole body is close to one, and PSNR is larger than 42 dB. We proposed a deep learning-based approach to generate sCT from whole-body NAC PET for PET AC. sCT generated with proposed method shows great similarity to true CT images both qualitatively and quantitatively, and demonstrates great potential for whole-body PET AC in the absence of structural information.

“Dead-end” metabolites in the metabolic reaction network do not have known metabolic functions. In this article, Gao et al. report that γ-6PGL, a byproduct of the oxiPPP, contributes to the Src-PP2A-AMPK signaling pathway by binding to Src and consequently enhancing PP2A recruitment, suggesting that “dead-end” metabolites may function as signaling molecules.

Alzheimer's disease (AD) is a growing global healthcare burden affecting the aging population and society. Given the lack of effective treatment to AD, early detection at the prodromal stage and timely monitoring of changes during progression are considered the best approach to control and intervene in disease progression. "Liquid biopsy" of AD biomarkers amyloid-β peptides (Aβs) and tau proteins in the cerebrospinal fluid (CSF) or blood samples holds great promises for cost-effective, widely accessible, and easy-administrated noninvasive detection and follow-up of AD. However, current in vitro detection methods have not yet demonstrated sufficient sensitivity and specificity using neither Aβs nor tau proteins biomarkers. One major challenge of accurate detection and measurement of biomarker levels in biofluidic samples is the biofouling effect with nonspecific adsorption of unwanted biomolecules, such as various serum proteins, on the surface of targeted detecting agents or devices, causing false-positive and false-negative findings. In this study, antibiofouling polymer polyethylene glycol-block-allyl glycidyl ether (PEG-b-AGE) coated magnetic iron oxide nanoparticles (IONPs) capable of suppressing the nonspecific interactions with biomolecules, especially proteins, were investigated for the immunomagnetic capturing of Aβ40 and Aβ42 peptides and tau protein spiked in CSF- and serum-mimicking samples using corresponding antibodies conjugated as targeting ligands. Antibody-conjugated antibiofouling IONPs demonstrated improved specificity (>90%) and sensitivity (>95%) over those of antibody-conjugated magnetic micron beads (Dynabeads, ∼50% specificity and 30-40% sensitivity) widely used as magnetic separating agents under the same experimental conditions with the presence of nontargeted interfering proteins. The antibody-conjugated IONPs also exhibited significantly higher sensitivities (80-90%) and better performance of capturing Aβs and tau protein from the human whole blood samples than antibody-conjugated Dynabeads (∼20%).

A non-rigid MR-TRUS image registration framework is proposed for prostate interventions. The registration framework consists of a convolutional neural networks (CNN) for MR prostate segmentation, a CNN for TRUS prostate segmentation and a point-cloud based network for rapid 3D point cloud matching. Volumetric prostate point clouds were generated from the segmented prostate masks using tetrahedron meshing. The point cloud matching network was trained using deformation field that was generated by finite element analysis. Therefore, the network implicitly models the underlying biomechanical constraint when performing point cloud matching. A total of 50 patients’ datasets were used for the network training and testing. Alignment of prostate shapes after registration was evaluated using three metrics including Dice similarity coefficient (DSC), mean surface distance (MSD) and Hausdorff distance (HD). Internal point-to-point registration accuracy was assessed using target registration error (TRE). Jacobian determinant and strain tensors of the predicted deformation field were calculated to analyze the physical fidelity of the deformation field. On average, the mean and standard deviation were 0.94±0.02, 0.90±0.23 mm, 2.96±1.00 mm and 1.57±0.77 mm for DSC, MSD, HD and TRE, respectively. Robustness of our method to point cloud noise was evaluated by adding different levels of noise to the query point clouds. Our results demonstrated that the proposed method could rapidly perform MR-TRUS image registration with good registration accuracy and robustness.

Purpose White matter hyperintensities (WMHs) are a risk factor for Alzheimer’s disease (AD). This study investigated the relationship between WMHs and white matter changes in AD using diffusion tensor imaging (DTI) and the sensitivity of each DTI index in distinguishing AD with WMHs. Subjects and Methods Forty-four subjects with WMHs were included. Subjects were classified into three groups based on the Scheltens rating scale: 15 AD patients with mild WMHs, 12 AD patients with severe WMHs, and 17 controls with mild WMHs. Fractional anisotropy (FA), mean diffusivity (MD), radial diffusivity (DR) and axial diffusivity (DA) were analyzed using the region of interest and Tract-Based Spatial Statistics methods. Sensitivity and specificity of DTI indices in distinguishing AD groups from the controls were evaluated. Results AD patients with mild WMHs exhibited differences from control subjects in most DTI indices in the medial temporal and frontal areas; however, differences in DTI indices from AD patients with mild WMHs and AD patients with severe WMHs were found in the parietal and occipital areas. FA and DR were more sensitive measurements than MD and DA in differentiating AD patients from controls, while MD was a more sensitive measurement in distinguishing AD patients with severe WMHs from those with mild WMHs. Conclusions WMHs may contribute to the white matter changes in AD brains, specifically in temporal and frontal areas. Changes in parietal and occipital lobes may be related to the severity of WMHs. DR may serve as an imaging marker of myelin deficits associated with AD.

Object This study investigates the effect of tumor location on alterations of language network by brain tumors at different locations using blood oxygenation level dependent (BOLD) fMRI and group independent component analysis (ICA). Subjects and Methods BOLD fMRI data were obtained from 43 right handed brain tumor patients. Presurgical mapping of language areas was performed on all 43 patients with a picture naming task. All data were retrospectively analyzed using group ICA. Patents were divided into three groups based on tumor locations, i.e., left frontal region, left temporal region or right hemisphere. Laterality index (LI) was used to assess language lateralization in each group. Results The results from BOLD fMRI and ICA revealed the different language activation patterns in patients with brain tumors located in different brain regions. Language areas, such as Broca’s and Wernicke’s areas, were intact in patients with tumors in the right hemisphere. Significant functional changes were observed in patients with tumor in the left frontal and temporal areas. More specifically, the tumors in the left frontal region affect both Broca’s and Wernicke’s areas, while tumors in the left temporal lobe affect mainly Wernicke’s area. The compensated activation increase was observed in the right frontal areas in patients with left hemisphere tumors. Conclusion Group ICA provides a model free alternative approach for mapping functional networks in brain tumor patients. Altered language activation by different tumor locations suggested reorganization of language functions in brain tumor patients and may help better understanding of the language plasticity.

The gene that encodes the epidermal growth factor receptor (EGFR) is frequently overexpressed or mutated in human cancers, including glioblastoma. However, the efficacy of EGFR-targeted small-molecule inhibitors or monoclonal antibodies in glioblastomas that also have mutation or deletion of the gene encoding phosphatase and tensin homolog (PTEN) has been modest. We found that EGFR signaling was blocked by a small molecule (G5-7) that selectively inhibited Janus kinase 2 (JAK2)-mediated phospho-rylation and activation of EGFR and STAT3 (signal transducer and activator of transcription 3) by binding to JAK2, thereby decreasing the activity of downstream signaling by mTOR (mammalian target of rapamycin) and inducing cell cycle arrest. G5-7 inhibited the proliferation of PTEN-deficient glioblastoma cell lines harboring a constitutively active variant of EGFR (U87MG/EGFRvIII) and human glioblastoma explant neu-rosphere cultures, but the drug only weakly inhibited the proliferation of either glioblastoma cell lines that were wild type for EGFR and stably transfected with PTEN (U87MG/PTEN) or normal neural progenitor cells and astrocytes. Additionally, G5-7 reduced vascular endothelial growth factor (VEGF) secretion and endothelial cell migration and induced apoptosis in glioblastoma xenografts, thereby suppressing glioblastoma growth in vivo. Furthermore, G5-7 was more potent than EGFR or JAK2 inhibitors that interfere with either ligand or adenosine 5′-triphosphate (ATP) binding at impeding glioblastoma cell proliferation, demonstrating that this allosteric JAK2 inhibitor may be an effective clinical strategy.