Sharing individualised template MRI data for MEG source reconstruction: a solution for open data while keeping subject confidentiality

The increasing requirements for adoption of FAIR data management and sharing original research data from neuroimaging studies can be at odds with protecting the anonymity of the research participants due to the person-identifiable anatomical features in the data. We propose a solution to this dilemma for anatomical MRIs used in MEG source analysis. In MEG analysis, the channel-level data is reconstructed to the source-level using models derived from anatomical MRIs. Sharing data, therefore, requires sharing the anatomical MRI to replicate the analysis. The suggested solution is to replace the individual anatomical MRIs with individualised warped templates that can be used to carry out the MEG source analysis and that provide sufficient geometrical similarity to the original participants' MRIs. First, we demonstrate how the individualised template warping can be implemented with one of the leading open-source neuroimaging analysis toolboxes. Second, we compare results from four different MEG source reconstruction methods performed with an individualised warped template to those using the participant's original MRI. While the source reconstruction results are not numerically identical, there is a high similarity between the results for single dipole fits, dynamic imaging of coherent sources beamforming, and atlas-based virtual channel beamforming. There is a moderate similarity between minimum-norm estimates, as anticipated due to this method being anatomically constrained and dependent on the exact morphological features of the cortical sheet. We also compared the morphological features of the warped template to those of the original MRI. These showed a high similarity in grey matter volume and surface area, but a low similarity in the average cortical thickness and the mean folding index within cortical parcels. Taken together, this demonstrates that the results obtained by MEG source reconstruction can be preserved with the warped templates, whereas the anatomical and morphological fingerprint is sufficiently altered to protect the anonymity of research participants. In cases where participants consent to sharing anatomical MRI data, it remains preferable to share the original defaced data with an appropriate data use agreement. In cases where participants did not consent to share their MRIs, the individualised warped MRI template offers a good compromise in sharing data for reuse while retaining anonymity for those research participants.

NIMG-16. DEEP LEARNING SUPER-RESOLUTION MR SPECTROSCOPIC IMAGING TO MAP TUMOR METABOLISM IN MUTANT IDH GLIOMA PATIENTS

Metabolic imaging can map spatially abnormal molecular pathways with higher specificity for cancer compared to anatomical imaging. However, acquiring high resolution metabolic maps similar to anatomical MRI is challenging in patients due to low metabolite concentrations, and alternative approaches that increase resolution by post-acquisition image processing can mitigate this limitation. We developed deep learning super-resolution MR spectroscopic imaging (MRSI) to map tumor metabolism in patients with mutant IDH glioma. We used a generative adversarial network (GAN) architecture comprised of a UNet neural network as the generator network and a discriminator network for adversarial training. For initial training we simulated a large data set of 9600 images with realistic quality for acquired MRSI to effectively train the deep learning model to upsample by a factor of four. Two types of training were performed: 1) using only the MRSI data, and 2) using MRSI and prior information from anatomical MRI to further enhance structural details. The performance of super-resolution methods was evaluated by peak SNR (PSNR), structure similarity index (SSIM), and feature similarity index (FSIM). After training on simulations, GAN was evaluated on measured MRSI metabolic maps acquired with resolution 5.2×5.2 mm2 and upsampled to 1.3×1.3 mm2. The GAN trained only on MRSI achieved PSNR = 27.94, SSIM = 0.88, FSIM = 0.89. Using prior anatomical MRI improved GAN performance to PSNR = 30.75, SSIM = 0.90, FSIM = 0.92. In the patient measured data, GAN super-resolution metabolic images provided clearer tumor margins and made apparent the tumor metabolic heterogeneity. Compared to conventional image interpolation such as bicubic or total variation, deep learning methods provided sharper edges and less blurring of structural details. Our results indicate that the proposed deep learning method is effective in enhancing the spatial resolution of metabolite maps which may better guide treatment in mutant IDH glioma patients.

Uncal apex movement and positioning in Alzheimer's Disease

Alzheimer’s Disease (AD) patients have consistently shown declines in declarative memory, consolidation, and in many other cognitive areas. These changes are associated with atrophy and volumetric declines in medial temporal lobe structures, such as the hippocampus. Hippocampal atrophy has been associated with AD. However, the influence of AD atrophy on the position of the uncal apex—an anatomical landmark for the hippocampus—has not been longitudinally examined. The current study’s objective is to investigate changes in the position of the uncal apex of AD patients over the course of two years. The current study draws upon the Alzheimer’s Disease Neuroimaging Initiative (ADNI) data set (adni.loni.usc.edu). For each participant, I obtained demographic data, anatomical MRI images in native space, hippocampal segmentations from the subcortical stream of FreeSurfer (v.5.3.0), and linear transforms to MNI space. Using uncal apex y-positions transformed in MNI space, I found that the uncal apex fell in a more posterior position in AD patients relative to control and that over time, the uncal apex migrates toward a more anterior position in both groups. These results suggest that part of the neuroimaging examinations that are done on AD patients should examine uncal apex positions as a biomarker of early AD progression. Future directions and limitations are discussed

Clinical outcomes of nonoperatively managed degenerative cervical myelopathy: an ambispective longitudinal cohort study in 117 patients

OBJECTIVE Degenerative cervical myelopathy (DCM) is among the most common pathologies affecting the spinal cord but its natural history is poorly characterized. The purpose of this study was to investigate functional outcomes in patients with DCM who were managed nonoperatively as well as the utility of quantitative clinical measures and MRI to detect deterioration. METHODS Patients with newly diagnosed DCM or recurrent myelopathic symptoms after previous surgery who were initially managed nonoperatively were included. Retrospective chart reviews were performed to analyze clinical outcomes and anatomical MRI scans for worsening compression or increased signal change. Quantitative neurological assessments were collected prospectively, including modified Japanese Orthopaedic Association (mJOA) score; Quick-DASH; graded redefined assessment of strength, sensation, and prehension–myelopathy version (GRASSP–M: motor, sensory, and dexterity); grip dynamometer; Berg balance scale score; gait stability ratio; and gait variability index. A deterioration of 10% was considered significant (e.g., a 2-point decrease in mJOA score). RESULTS A total of 117 patients were included (95 newly diagnosed, 22 recurrent myelopathy), including 74 mild, 28 moderate, and 15 severe cases. Over a mean follow-up of 2.5 years, 57% (95% CI 46%–67%) of newly diagnosed patients and 73% (95% CI 50%–88%) of patients with recurrent DCM deteriorated neurologically. Deterioration was best detected with grip strength (60%), GRASSP dexterity (60%), and gait stability ratio (50%), whereas the mJOA score had low sensitivity (33%) in 50 patients. A composite score had a sensitivity of 81% and a specificity of 82%. The sensitivity of anatomical MRI was 28% (83 patients). CONCLUSIONS DCM appears to have a poor natural history; however, prospective studies are needed for validation. Serial assessments should include mJOA score, grip strength, dexterity, balance, and gait analysis. The absence of worsening on anatomical MRI or in mJOA scores is not sufficient to determine clinical stability.

Hypermetabolism and impaired cerebrovascular reactivity beyond the standard MRI-identified tumor border indicate diffuse glioma extended tissue infiltration

Background Diffuse gliomas exhibit diffuse infiltrative growth, often beyond the MRI-detectable tumor lesion. Within this lesion, hypermetabolism and impaired cerebrovascular reactivity are found, but its exact distribution pattern into the peritumoral environment is unknown. Our aim was to better characterize the extent of diffuse glioma tissue infiltration, beyond the visible lesion (i.e. beyond the T1-contrast-enhancing lesion and/or T2/FLAIR defined tumor border), with metabolic PET, and fMRI cerebrovascular reactivity (BOLD-CVR) mapping. Methods From a prospective glioma database 18 subjects (19 datasets) with diffuse glioma (n=2 with anaplastic astrocytoma, n=10 with anaplastic oligodendroglioma, n=7 with glioblastoma) underwent a BOLD-CVR and metabolic PET study between February 2016 and September 2019, 7 of them at primary diagnosis and 12 at tumor recurrence. In addition, 19 matched healthy controls underwent an identical BOLD-CVR study. The tumor lesion was defined using high-resolution anatomical MRI. Volumes of interest (VOIs) starting from the tumor lesion outwards up to 30 mm were created for a detailed peritumoral PET and BOLD-CVR tissue analysis. Student’s t test were used for statistical analysis. Results Patients with diffuse glioma exhibit impaired BOLD-CVR 12 mm beyond the tumor lesion (p=0.02) with normalization of BOLD-CVR values after 24 mm. Metabolic PET shows a difference between the affected and contralateral hemisphere of 6 mm (p=0.05) with PET values normalization after 12 mm. Conclusion We demonstrate hypermetabolism and impaired cerebrovascular reactivity beyond the standard MRI-defined tumor border, suggesting active tumor infiltration in the peritumoral environment.

On the Gestural Origin of Language Lateralisation: Manual Communication reflects Broca’s Asymmetry in Monkeys

Manual gestures and speech recruit a common neural network, involving Broca area in the left hemisphere. Evolutionary questions about this language organization led to a renewed attention for comparative research on gestural communication in our closer primate relatives and its potential language-like features. Here, using in vivo anatomical MRI in 80 baboons, we found that communicative gesturing’s lateralisation – but not handedness for manipulation - is related to Broca homologue’s marker in monkeys, namely contralateral depth hemispheric asymmetry of the ventral portion of the inferior arcuate sulcus. This finding provides strong support for the gestural evolutionary continuities with language-related frontal specialization, dating back not only to Homo sapiens evolution, but rather to a much older common ancestor shared with old-world monkeys, 25-35 million years ago.

The effect of time since stroke, gender, age, and lesion size on thalamus volume in chronic stroke: a pilot study

Recent stroke studies have shown that the ipsi-lesional thalamus longitudinally and significantly decreases after stroke in the acute and subacute stages. However, additional considerations in the chronic stages of stroke require exploration including time since stroke, gender, intracortical volume, aging, and lesion volume to better characterize thalamic differences after cortical infarct. This cross-sectional retrospective study quantified the ipsilesional and contralesional thalamus volume from 69 chronic stroke subjects’ anatomical MRI data (age 35–92) and related the thalamus volume to time since stroke, gender, intracortical volume, age, and lesion volume. The ipsi-lesional thalamus volume was significantly smaller than the contra-lesional thalamus volume (t(68) = 13.89, p < 0.0001). In the ipsilesional thalamus, significant effect for intracortical volume (t(68) = 2.76, p = 0.008), age (t(68) = 2.47, p = 0.02), lesion volume (t(68) = − 3.54, p = 0.0008), and age*time since stroke (t(68) = 2.46, p = 0.02) were identified. In the contralesional thalamus, significant effect for intracortical volume (t(68) = 3.2, p = 0.002) and age (t = − 3.17, p = 0.002) were identified. Clinical factors age and intracortical volume influence both ipsi- and contralesional thalamus volume and lesion volume influences the ipsilesional thalamus. Due to the cross-sectional nature of this study, additional research is warranted to understand differences in the neural circuitry and subsequent influence on volumetrics after stroke.

DIFFERENCE OF ANATOMY INFORMATION MRI KNEE JOINT ON VARIATION OF TIME REPETITION SEQUENCES STIR IN SAGITTAL SLICES

Background: Time Repetition (TR) is one of the main parameters of Inversion Recovery. The purpose of this study to determine differences in anatomical MRI information on the variation of the knee joint TR sequences STIR Sagittal slices. Method: Type of research is experimental. The study was conducted with MRI 1.5 Tesla. Data in the form of 42 image sequences STIR MRI knee joint with TR 3500, 4000, 4500, 5000, 5500, 6000, and 6500 ms. Anatomical assessments on the anterior cruciate ligament, posterior cruciate ligament, articular cartilage, and meniscus were performed by a radiologist. Data analyzed by Friedman and Wilcoxon test. Result: The results showed that there were differences in the MRI anatomical information of the knee joint of the STIR sagitas slice in the TR variation with p-value < 0.001. There is a difference in anatomical information between TR 5000 and 6000 ms (p-value = 0.034), TR 5000 and 6500 ms (p-value = 0.024), TR 5500 and 6500 ms (p-value = 0.038). There is no difference in anatomical information between TR 4500 and 5000 ms (p-value = 0.395), TR 4500 and 5500 ms (p-value = 0.131), TR 4500 and 6000 ms (p-value = 0.078), TR 4500 and 6500 ms (p-value = 0.066), TR 5000 and 5500 ms (p-value = 0.414), TR 5500 and 6000 ms (p-value = 0.102), TR 6000 and 6500 ms (p-value = 0.083). Conclusion: The optimal value to produce anatomical information of the knee joint sagittal MRI sequences STIR is TR 4500 ms.

Anatomical MRI and [18F]FDG PET/CT imaging of Schistosoma mansoni in a NMRI mouse model

Schistosomiasis represents one of the most devastating worm parasitosis in the world. Current diagnostic methods are insufficient to determine the infection grade and the disease related organ damage. We herein investigated whether discrimination of infection grade and its correlation to liver damage could be accurately performed by multimodal imaging in a mouse model of Schistosoma mansoni infection. Therefore, groups of uninfected and infected mice underwent MRI and [18F]FDG PET/CT imaging. Anatomical MRI images were used for liver volumetry and for quantification of hepatic granulomas. For PET/CT images a volume of interest based analyses were employed to calculate the [18F]FDG uptake in liver, portal vein, spleen and abdomen. Herein, we demonstrate that the combined use of [18F]FDG-PET/CT and MRI represents an appropriate diagnostic tool for Schistosoma mansoni infection, but fails to discriminate the infection grade and the linked organ damage. Only the splenic [18F]FDG uptake in the 25 cercariae group (5.68 ± 0.90%ID/cc) and 50 cercariae group (4.98 ± 1.43%ID/cc) was significantly higher compared to the control group (2.13 ± 0.69%ID/cc). Nevertheless, future multimodal imaging studies with new radiopharmaceuticals could build a highly sensitive and specific basis for the diagnosis and evaluation of organ damage of schistosomiasis.