Mediri (Germany)

Alzheimer’s disease is a progressive neuro-degenerative disorder that affects global health. The amyloid beta protein (Aβ) is an important biomarker of Alzheimer’s disease, with levels increasing early in the course of the disease. Patients can be categorized into normal and abnormal in terms of amyloid status. Since currently measuring Aβ requires invasive procedures, we present a method to predict amyloid status using multi-modal medical image data with transformer-based machine learning architectures. Data from ADNI included structural T1-weighted MRI and FDG-PET scans. We compared two architectures for classifying the amyloid status and differentiating the three cognitively related phases: Alzheimer’s disease, mild cognitive impairment, and cognitively normal. Our best setup achieved 85.48% (±4.93%) accuracy for predicting amyloid status and 85.42% (±3.44%) accuracy for distinguishing between the three phases. The proposed methods improved upon the best mono-modal setup 73.67% (±4.70%) on T1-weighted MRI and showed competitive performance against a literature reference using FDG-PET. Our results highlight the potential of using non-specific amyloid-related data, like T1-weighted MRI and FDG-PET, for accurate amyloid status prediction.

Reconfigurable metamaterials offer many degrees of freedom and allow to overcome limitations of present technological capabilities in Magnetic Resonance Imaging (MRI). Various metrics such as speed, efficiency, and local signal-to-noise ratio (SNR) can be drastically improved and tailored applications rather than global effects become possible. So far, in the literature there exist a few metamaterial realizations for MRI, and only a small subset includes non-static mechanisms and a very low degree of reconfigurability, if any. At least, resonance frequency fine-tuning must be provided to handle patient-specific and vendor-related effects. Manual control is too far from clinical use. Only a digital and wireless interface, supplemented by an on-board logic, will have a meaningful impact. True reconfigurability does not only include tuning at the unit cell level and dynamic spatial sensitivity profiles w.r.t. the interaction with the incident transmit and/ or receive fields, but also the precise control in the time domain. Thereupon, the integration into some sequence control and development framework is mandatory, ideally in a vendor-independent way. The state of the art, potential benefits, and various use-cases of static, dynamic, reconfigurable, and interactive metamaterials as well as control strategies are presented. Applications range from simple SNR enhancement to spatial encoding patterns for reduced gradient fields. First-generation prototypes of such metamaterials demonstrate the working principle, and on-bench characterization results as well as MRI measurements are shown. For a specific application, a pre-defined target function must be extremized. A prime example is a desired spatio-temporal sensitivity profile for SNR modifications. Thus, suitable optimization strategies have to be implemented, which include conventional as well as AI-driven approaches. Simulations, on-bench measurements, deep learning results, and MRI verification are demonstrated for a proof-of-principle application.

Abstract BackgroundBlood‐brain barrier (BBB) dysfunction is considered a hallmark of Alzheimer’s disease (AD), making non‐invasive imaging of BBB with arterial spin labeling (ASL) MRI a potential imaging biomarker (BBB‐ASL). However, the normal BBB development over the lifespan is unknown. Here, we created population‐based BBB‐ASL permeability reference maps, and aimed to investigate associations between BBB‐ASL, chronological age, and biological cerebrovascular age – expressed as white matter hyperintensities (WMH) volume.MethodCognitively‐unimpaired (CU) participants were included from Center for Lifespan Changes in Brain and Cognition (LCBC) (n = 28, age 58.9±16.1y) and University of Singapore (n = 30, age 55.8±6.1y). The Singapore cohort is known to be vascularly compromised and thus expected to have early cerebrovascular pathology in the form of WMH. Multi‐echo, multi‐post‐labeling‐delay pseudo‐continuous ASL images were acquired at two identical Siemens 3T scanners and analyzed with ExploreASL. Gray matter (GM) BBB water time‐of‐exchange (Tex) was quantified with and without partial‐volume correction (PVC). For Singapore data, WMH were segmented on 3D FLAIR. Linear regression models were used to investigate associations of Tex with age and WMH volume, with and without covarying for sex and site.ResultBoth site population‐averages (Figure 1) show similar Tex patterns, with slightly higher posterior variability in the Singapore data. GM Tex was negatively associated with age (Figure 2A) irrespective of PVC, and after covarying for sex and site (Table 1). GM Tex was also negatively associated with WMH volume (β = ‐0.005, p = 0.012), although the association became non‐significant if both PVC and sex and site corrections were applied (β = ‐0.005, p = 0.068).ConclusionOur findings support the hypothesis of increased BBB permeability with aging and there are signs of association with increased cerebrovascular age assessed with WMH volume. We aim to further increase our sample size and also investigate associations with other AD biomarkers.

Current structural MRI‐based brain age estimates and their difference from chronological age—the brain age gap (BAG)—are limited to late‐stage pathological brain‐tissue changes. The addition of physiological MRI features may detect early‐stage pathological brain alterations and improve brain age prediction. This study investigated the optimal combination of structural and physiological arterial spin labelling (ASL) image features and algorithms. Healthy participants (n = 341, age 59.7 ± 14.8 years) were scanned at baseline and after 1.7 ± 0.5 years follow‐up (n = 248, mean age 62.4 ± 13.3 years). From 3 T MRI, structural (T1w and FLAIR) volumetric ROI and physiological (ASL) cerebral blood flow (CBF) and spatial coefficient of variation ROI features were constructed. Multiple combinations of features and machine learning algorithms were evaluated using the Mean Absolute Error (MAE). From the best model, longitudinal BAG repeatability and feature importance were assessed. The ElasticNetCV algorithm using T1w + FLAIR+ASL performed best (MAE = 5.0 ± 0.3 years), and better compared with using T1w + FLAIR (MAE = 6.0 ± 0.4 years, p < .01). The three most important features were, in descending order, GM CBF, GM/ICV, and WM CBF. Average baseline and follow‐up BAGs were similar (−1.5 ± 6.3 and − 1.1 ± 6.4 years respectively, ICC = 0.85, 95% CI: 0.8–0.9, p = .16). The addition of ASL features to structural brain age, combined with the ElasticNetCV algorithm, improved brain age prediction the most, and performed best in a cross‐sectional and repeatability comparison. These findings encourage future studies to explore the value of ASL in brain age in various pathologies.

A população em geral toma decisões diariamente antes de se deslocar, buscando rotas mais convenientes possíveis. Diversos fatores podem influenciar nessa escolha, desde o tempo de deslocamento até mesmo à paisagem. Porém, nesta pesquisa buscou-se avaliar parâmetros relacionados ao pavimento, que seriam: tipo e condição. Para isso, foram escolhidos três trechos que interligam uma farmácia à uma escola, na cidade de Campo Grande/RN. Foi aplicada pesquisa de opinião à população que utiliza como modo de transporte, principalmente, os veículos pesados, os carros e as motos, a fim de identificar se o tipo e a condição do pavimento são fatores importantes na escolha de rota. Além disso, foram determinados o Levantamento Visual Contínuo (LVC), de acordo com a norma DNIT 008/2003 – PRO e o Índice de Irregularidade Internacional (IRI), por meio do aplicativo Roadlab Pro, a fim de medir a condição do pavimento e correlacionar esses dados com as escolhas dos usuários. Verificou-se que a rota com pavimento em paralelepípedo (trecho 1) foi o trecho menos escolhido e avaliada como condição ruim, independentemente do tipo de transporte. O segundo e o terceiro trecho, ambos em pavimento asfáltico, apesar de ligeira irregularidade demonstrada pelo IRI, ainda assim foram os preferidos por carros e motos. Por apresentar geometria mais propícia ao deslocamento de veículos pesados, o trecho 3 foi o mais selecionado por essa classe. De forma geral, foi possível notar que o pavimento também pode ser um parâmetro relevante e influente na escolha de rotas dos usuários.

BackgroundAdvances in the understanding of the pathophysiology of frontotemporal dementia (FTD) and related disorders, along with the development of novel candidate disease modifying treatments, have stimulated the need for tools to assess the efficacy of new therapies. While perfusion imaging by arterial spin labeling (ASL) is an attractive approach for longitudinal imaging biomarkers of neurodegeneration, sources of variability between sessions including arterial transit times (ATT) and fluctuations in resting perfusion can reduce its sensitivity. Establishing the magnitude of perfusion changes that can be reliably detected is necessary to delineate longitudinal perfusion changes related to disease processes from the effects of these sources of error. PurposeTo assess the feasibility of ASL for longitudinal monitoring of patients with FTD by quantifying between-session variability of perfusion on a voxel-by-voxel basis. Methods and materialsASL data were collected in 13 healthy controls and 8 patients with FTD or progressive supra-nuclear palsy. Variability in cerebral blood flow (CBF) by single delay pseudo-continuous ASL (SD-pCASL) acquired one month apart were quantified by the coefficient of variation (CV) and intraclass correlation coefficient (ICC). Additionally, CBF by SD-pCASL and ATT by low-resolution multiple inversion time ASL (LowRes-pCASL) were compared to Hadamard encoded sequences which are able to simultaneously measure CBF and ATT with improved time-efficiency. ResultsAgreement of grey-matter perfusion between sessions was found for both patients and controls (CV = 10.8% and 8.3% respectively) with good reliability for both groups (ICC > 0.6). Intensity normalization to remove day-to-day fluctuations in resting perfusion reduced the CV by 28%. Less than 5% of voxels had ATTs above the chosen post labelling delay (2 s), indicating that the ATT was not a significant source of error. Hadamard-encoded perfusion imaging yielded systematically higher CBF compared to SD-pCASL, but produced similar transit-time measurements. Power analysis revealed that SD-pCASL has the sensitivity to detect longitudinal changes as low as 10% with as few as 10 patient participants. ConclusionWith the appropriate labeling parameters, SD-pCASL is a promising approach for assessing longitudinal changes in CBF associated with FTD.

&lt;table&gt; &lt;tbody&gt; &lt;tr&gt; &lt;td&gt;This document reports the work conducted in the inter-pilot and collaboration task&lt;br&gt; (T7.6). It includes new use cases tested in each pilot site, describes the main&lt;br&gt; generalization aspects of the PIXEL Platform and models and lessons learnt&lt;br&gt; regarding the PIXEL framework, in the four use-cases, have been included.&lt;/td&gt; &lt;/tr&gt; &lt;/tbody&gt; &lt;/table&gt; &lt;p&gt;&nbsp;&lt;/p&gt;

Although arterial spin labeling (ASL) is appealing for mapping long-term changes in functional activity, inter-sessional variations in basal blood flow, arterial transit times (ATTs), and alignment errors, can result in significant false activation when comparing images from separate sessions. By taking steps to reduce these sources of noise, this study assessed the ability of ASL to detect functional CBF changes between sessions. ASL data were collected in three sessions to image ATT, resting CBF and CBF changes associated with motor activation (7 participants). Activation maps were generated using rest and task images acquired in the same session and from sessions separated by up to a month. Good agreement was found when comparing between-session activation maps to within-session activation maps with only a 16% decrease in precision (within-session: 90 ± 7%) and a 13% decrease in the Dice similarity (within-session: 0.75 ± 0.07) coefficient after a month. In addition, voxel-wise reproducibility (within-session: 4.7 ± 4.5%) and reliability (within-session: 0.89 ± 0.20) of resting grey-matter CBF decreased by less than 18% for the between-session analysis relative to within-session values. ATT variability between sessions (5.0 ± 2.7%) was roughly half the between-subject variability, indicating that its effects on longitudinal CBF were minimal. These results demonstrate that conducting voxel-wise analysis on CBF images acquired on different days is feasible with only modest loss in precision, highlighting the potential of ASL for longitudinal studies.

Automated three-dimensional breast ultrasound (ABUS) is a valuable adjunct to x-ray mammography for breast cancer screening of women with dense breasts. High image quality is essential for proper diagnostics and computer-aided detection. We propose an automated image quality assessment system for ABUS images that detects artifacts at the time of acquisition. Therefore, we study three aspects that can corrupt ABUS images: the nipple position relative to the rest of the breast, the shadow caused by the nipple, and the shape of the breast contour on the image. Image processing and machine learning algorithms are combined to detect these artifacts based on 368 clinical ABUS images that have been rated manually by two experienced clinicians. At a specificity of 0.99, 55% of the images that were rated as low quality are detected by the proposed algorithms. The areas under the ROC curves of the single classifiers are 0.99 for the nipple position, 0.84 for the nipple shadow, and 0.89 for the breast contour shape. The proposed algorithms work fast and reliably, which makes them adequate for online evaluation of image quality during acquisition. The presented concept may be extended to further image modalities and quality aspects.

Magnetic Resonance (MR) guided Focused Ultrasound Surgery (FUS) of abdominal organs is challenging due to breathing motion and limited patient access in the MR environment. In this work, an experimental robotically assisted FUS setup was combined with a MR-based navigator technique to realize motion-compensated sonications and online temperature imaging. Experiments were carried out in a static phantom, during periodic manual motion of the phantom without triggering, and with triggering to evaluate the triggering method. In contrast to the non-triggered sonication, the results of the triggered sonication show a confined symmetric temperature distribution. In conclusion, the velocity sensitive navigator can be employed for triggered FUS to compensate for periodic motion. Combined with the robotic FUS setup, flexible treatment of abdominal targets might be realized.