Imaging Time Research Articles

SmartEM: machine-learning guided electron microscopy.

Connectomics provides essential nanometer-resolution, synapse-level maps of neural circuits to understand brain activity and behavior. However, few researchers have access to the high-throughput electron microscopes necessary to generate enough data for whole circuit or brain reconstruction. To date, machine-learning methods have been used after the collection of images by electron microscopy (EM) to accelerate and improve neuronal segmentation, synapse reconstruction and other data analysis. With the computational improvements in processing EM images, acquiring EM images has now become the rate-limiting step. Here, in order to speed up EM imaging, we integrate machine-learning into real-time image acquisition in a single-beam scanning electron microscope. This SmartEM approach allows an electron microscope to perform intelligent, data-aware imaging of specimens. SmartEM allocates the proper imaging time for each region of interest - scanning all pixels equally rapidly, then re-scanning small subareas more slowly where a higher quality signal is required to achieve accurate segmentability, in significantly less time. We demonstrate that this pipeline achieves a 7-fold acceleration of image acquisition time for connectomics using a commercial single-beam SEM. We apply SmartEM to reconstruct a portion of mouse cortex with the same accuracy as traditional microscopy but in less time.

Association between metformin use with circumpapillary retinal nerve fibre layer thickness and capillary vessel density in glaucoma

Background/aimsTo investigate the association between use of metformin and circumpapillary retinal nerve fibre layer (cpRNFL) thickness, as well as whole image capillary density (wiCD), in patients with glaucoma.MethodsThis cross-sectional study...

Feasibility and optimization of 19F MRI on a clinical 3T with a large field-of-view torso coil

Objective. The objective of this work is to: (1) demonstrate fluorine-19 (19F) MRI on a 3T clinical system with a large field of view (FOV) multi-channel torso coil (2) demonstrate an example parameter selection optimization for a 19F agent to maximize the signal-to-noise ratio (SNR)-efficiency for spoiled gradient echo (SPGR), balanced steady-state free precession (bSSFP), and phase-cycled bSSFP (bSSFP-C), and (3) validate detection feasibility in ex vivo tissues. Approach. Measurements were conducted on a 3.0T Discovery MR750w MRI (GE Healthcare, USA) with an 8-channel 1H/19F torso coil (MRI Tools, Germany). Numerical simulations were conducted for perfluoropolyether to determine the theoretical parameters to maximize SNR-efficiency for the sequences. Theoretical parameters were experimentally verified, and the sensitivity of the sequences was compared with a 10 min acquisition time with a 3.125 × 3.125 × 3 mm3 in-plane resolution. Feasibility of a bSSFP-C was also demonstrated in phantom and ex vivo tissues. Main Results. Flip angles (FAs) of 12 and 64° maximized the signal for SPGR and bSSFP, and validation of optimal FA and receiver bandwidth showed close agreement with numerical simulations. Sensitivities of 2.47, 5.81, and 4.44 ms−0.5mM−1 and empirical detection limits of 20.3, 1.5, and 6.2 mM were achieved for SPGR, bSSFP, and bSSFP-C, respectively. bSSFP and bSSFP-C achieved 1.8-fold greater sensitivity over SPGR (p < 0.01). Significance. bSSFP-C was able to improve sensitivity relative to simple SPGR and reduce both bSSFP banding effects and imaging time. The sequence was used to demonstrate the feasibility of 19F MRI at clinical FOVs and field strengths within ex-vivo tissues.

PSDCLS: Parallel simultaneous diffusion–confusion image cryptosystem based on Latin square

Today, due to the unparalleled growth of multimedia data sharing, especially digital images, between users over insecure channels in real-time applications, cryptography algorithms have gained increasing attention for the secure and efficient transmission. In classical chaos-based image cryptosystems, the confusion and diffusion operations are often applied as two separate and independent phases, which threatens the cryptosystem security. To address these problems, in this paper, a fast image cryptosystem based on parallel simultaneous diffusion–confusion strategy has been proposed using Latin squares, called PSDCLS. It consists of three main steps. First, the initial parameters of the Hénon-Sine chaotic map are produced from SHA256 of both the plain image content and the user’s secret key. Second, a chaos-based random Latin square is constructed by employing the chaotic sequence produced through the Hénon-Sine chaotic map. Third, a parallel simultaneous diffusion–confusion scheme is proposed by using Latin square and vectorization technique to overcome the problems of computational complexity and high risk of separable and iterative confusion–diffusion operations in the classical chaos-based image cryptosystems. To analyze and evaluate the security and performance of PSDCLS cryptosystem, we conducted extensive simulations and experiments on various benchmark images. Experimental results and analyses show that PSDCLS achieves excellent scores for information entropy (>7.99), correlation coefficients close to 0, key space (2512), NPCR (>99.60%), UACI (>33.46%). The encryption time for test images of size 512 × 512 and 512×512×3 was around 0.026 and 0.081 s, respectively. Therefore, PSDCLS is highly robust against common cryptographic attacks and serves as a swift cryptosystem for real-time encryption applications. The source code of PSDCLS is accessible at: https://github.com/EbrahimZarei64/PSDCLS.

Deep Learning With Physics-Embedded Neural Network for Full Waveform Ultrasonic Brain Imaging.

The convenience, safety, and affordability of ultrasound imaging make it a vital non-invasive diagnostic technique for examining soft tissues. However, significant differences in acoustic impedance between the skull and soft tissues hinder the successful application of traditional ultrasound for brain imaging. In this study, we propose a physics-embedded neural network with deep learning based full waveform inversion (PEN-FWI), which can achieve reliable quantitative imaging of brain tissues. The network consists of two fundamental components: forward convolutional neural network (FCNN) and inversion sub-neural network (ISNN). The FCNN explores the nonlinear mapping relationship between the brain model and the wavefield, replacing the tedious wavefield calculation process based on the finite difference method. The ISNN implements the mapping from the wavefield to the model. PEN-FWI includes three iterative steps, each embedding the F CNN into the ISNN, ultimately achieving tomography from wavefield to brain models. Simulation and laboratory tests indicate that PEN-FWI can produce high-quality imaging of the skull and soft tissues, even starting from a homogeneous water model. PEN-FWI can achieve excellent imaging of clot models with constant uniform distribution of velocity, randomly Gaussian distribution of velocity, and irregularly shaped randomly distributed velocity. Robust differentiation can also be achieved for brain slices of various tissues and skulls, resulting in high-quality imaging. The imaging time for a horizontal cross-sectional imag e of the brain is only 1.13 seconds. This algorithm can effectively promote ultrasound-based brain tomography and provide feasible solutions in other fields.

Mapping drivers of tropical forest loss with satellite image time series and machine learning

The rates of tropical deforestation remain high, resulting in carbon emissions, biodiversity loss, and impacts on local communities. To design effective policies to tackle this, it is necessary to know what the drivers behind deforestation are. Since drivers vary in space and time, producing accurate spatially explicit maps with regular temporal updates is essential. Drivers can be recognized from satellite imagery but the scale of tropical deforestation makes it unfeasible to do so manually. Machine learning opens up possibilities for automating and scaling up this process. In this study, we developed and trained a deep learning model to classify the drivers of any forest loss—including deforestation—from satellite image time series. Our model architecture allows understanding of how the input time series is used to make a prediction, showing the model learns different patterns for recognizing each driver and highlighting the need for temporal data. We used our model to classify over 588′000 sites to produce a map detailing the drivers behind tropical forest loss. The results confirm that the majority of it is driven by agriculture, but also show significant regional differences. Such data is a crucial source of information to enable targeting specific drivers locally and can be updated in the future using free satellite data.

A robust and accurate feature matching method for multi-modal geographic images spatial registration

ABSTRACT While the current research has achieved satisfactory results for the registration of single mode data, there has always been a significant challenge in the registration of multi-modal images due to the obvious nonlinear radiation differences caused by different imaging mechanisms and imaging time. For example, multi-temporal visible, visible-synthetic aperture radar, visible-near infrared, near infrared-short wave infrared, visible-MAP, etc. To address this problem, we propose a Robust and Accurate Feature Matching Method for Multi-modal Geographic Images Spatial Registration (RAMMR) to fully extract common key points between images, weaken the radiation difference between data, and finally accurately match more inliers to realize multi-modal image registration. Considering the influence of noise and edge information on key point extraction, RAMMR first constructs a new scale space by introducing the Side Window Filter (SWF); Then, we improve Harris algorithm to extract key points based on the SWF scale space; After that, we propose an enhanced log-polar descriptor based on the gradient angles and gradient amplitudes of the scale space, which effectively improves the quality of the descriptor and avoids the mismatch of key points; Based on the standard Euclidean distance, we design a re-match strategy to obtain the initial matching results, and Random Sample Consensus (RANSAC) is used to eliminate outliers. Finally, the affine transformation parameters are calculated based on inliers, and multi-modal image registration is realized. RAMMR is evaluated on different multi-modal datasets and compared with some state-of-art methods. The experimental results show that RAMMR accurately registers multi-modal geographic images and obtains comparative results compared with benchmark methods. Our source datasets are publicly available at https://github.com/RSmfmr/multimodal-dataset.

Partial Acquisition of Spectral Projections Accelerates Four-dimensional Spectral-spatial EPR Imaging for Mouse Tumor Models: A Feasibility Study.

Our study aimed to accelerate the acquisition of four-dimensional (4D) spectral-spatial electron paramagnetic resonance (EPR) imaging for mouse tumor models. This advancement in EPR imaging should reduce the acquisition time of spectroscopic mapping while reducing quality degradation for mouse tumor models. EPR spectra under magnetic field gradients, called spectral projections, were partially measured. Additional spectral projections were later computationally synthesized from the measured spectral projections. Four-dimensional spectral-spatial images were reconstructed from the post-processed spectral projections using the algebraic reconstruction technique (ART) and assessed in terms of their image qualities. We applied this approach to a sample solution and a mouse Hs766T xenograft model of human-derived pancreatic ductal adenocarcinoma cells to demonstrate the feasibility of our concept. The nitroxyl radical imaging agent 2H,15N-DCP was exogenously infused into the mouse xenograft model. The computation code of 4D spectral-spatial imaging was tested with numerically generated spectral projections. In the linewidth mapping of the sample solution, we achieved a relative standard uncertainty (standard deviation/| mean |) of 0.76 μT/45.38 μT = 0.017 on the peak-to-peak first-derivative EPR linewidth. The qualities of the linewidth maps and the effect of computational synthesis of spectral projections were examined. Finally, we obtained the three-dimensional linewidth map of 2H,15N-DCP in a Hs766T tumor-bearing leg in vivo. We achieved a 46.7% reduction in the acquisition time of 4D spectral-spatial EPR imaging without significantly degrading the image quality. A combination of ART and partial acquisition in three-dimensional raster magnetic field gradient settings in orthogonal coordinates is a novel approach. Our approach to 4D spectral-spatial EPR imaging can be applied to any subject, especially for samples with less variation in one direction.

Association of Cardiovascular Fibrosis, Remodeling, and Dysfunction With Frailty, Prefrailty, and Functional Performance: The Multi-Ethnic Study of Atherosclerosis.

Cardiovascular disease is associated with higher incidence of frailty. However, the nature of the mechanisms underlying this association remains unclear. The purpose of this study is to identify cardiovascular phenotypes most associated with physical frailty and functional performance in the Multi-Ethnic Study of Atherosclerosis (MESA). As part of the MESA study, 3 045 participants underwent cardiovascular magnetic resonance and computed tomography between 2010 and 2012. Of these, 1 743 completed a Six-Minute Walk test (6MWT) and questionnaires (follow-up exam: 2016-2018) which were used to generate a binary combined frail/prefrail versus robust score according to a modified FRAIL Scale (self-report questionnaire). Multivariable logistic (binary frail outcome) or linear (6MWT) regression assessed the association between frailty and cardiovascular structure and function, aortic stiffness, coronary artery calcium, and myocardial fibrosis (ECV, extracellular volume fraction). Participants were 66 ± 8 years, 52% female at the time of imaging, and 29.4% were classified as frail or prefrail. Older age and female gender were associated with greater odds of being in the frail/prefrail group. Concentric left ventricular remodeling (odds ratio [OR] 1.89, p = .008; Coef. -52.9, p < .001), increased ECV (OR 1.10, p = .002; Coef. -4.0, p = .001), and worsening left atrial strain rate at early diastole (OR 1.56, p ≤ .001; Coef. -22.75, p = .027) were found to be associated with a greater likelihood of being in a frail state and lower 6MWT distance (m). All associations with 6MWT performance were attenuated with adjustments for risk factors whereas ECV and LA strain rate remained independently associated with frailty. These findings suggest a significant overlap in pathways associated with subclinical cardiac dysfunction, cardiovascular fibrosis, and physical frailty.

GR.4 Circadian rhythm influences ischemic core and penumbra volumes in pediatric and young adult populations

Background: Circadian rhythms are implicated in timing of stroke onset and infarct progression in adults, but this has not been studied in pediatric/young adult populations. Methods: We queried the RAPID Insights database from centers in USA for unique patients &lt;25 years with a CTP (10/05/2018-09/29/2023) and a minimum ischemic core volume (defined as relative cerebral blood flow (rCBF) reduction of &lt;30%) of &gt;0 cc and minimum mismatch of &gt;0 cc. Imaging time was subdivided into three epochs: Nigh (23:00 h-06:59 h), Day (07:00 h-14:59 h), and Evening (15:00 h-22:59 h). We analyzed age by pre-defined strata: &lt;2 years, 2-5, 6-11, 12-18 and 19-25. Perfusion parameters (core, perfusion volume, mismatch ratio) were analyzed using descriptive statistics. Results: 836 patients were included; 52.3% were in the 19-25 category. Median ischemic cores were larger during the Night (23.0cc [10.0 – 58.0]) compared to Day (19.0cc [8.0-42.0]) or Evening (15.0cc [7.0-33.0]), p=0.009. There was a trend towards larger perfusion volumes in the Night epoch. In the 19-25 group, perfusion volumes were significantly larger at Night (127.5cc [51.5 – 203.5]) compared to Day (74.0cc [33.0 – 139.0]) or Evening (76.0cc [38.0 – 157.5]), with larger mismatch volumes at Night. Conclusions: This is the first study to demonstrate diurnal fluctuations in perfusion parameters in a predominantly pediatric cohort.

Cardiovascular magnetic resonance 3D LGE imaging for atrial fibrillation ablation

Abstract Objective To assess feasibility and reliability of cardiac magnetic resonance (CMR) imaging in visualizing left atrial (LA) structural changes that patients with paroxysmal atrial fibrillation (PAF) undergo acutely and long-term post ablation. Two different ablation methods i.e pulsed filed ablation (PFA) and cryoablation were studied. Background Pulmonary vein (PV) isolation has become the cornerstone technique for catheter ablation in patients with drug-refractory atrial fibrillation (AF). At present, almost all ablation technologies are mediated by a thermal effect. Recently, pulsed field ablation (PFA) has been introduced as a new energy source. PFA is a non-thermal ablative modality in which high voltage ultra-short pulses are applied to target tissue, while preserving the extracellular matrix architecture, nerves, and microvascular structures. Few data exist showing different structural LA changes between the two methods using Three-Dimensional (3D) Late Gadolinium Enhancement (LGE) Imaging. Methods CMR imaging was performed pre-ablation, acutely (&amp;lt;24 h), and 3 months post-ablation in 10 patients with paroxysmal atrial fibrillation (AF) undergoing pulmonary vein (PV) isolation with PFA (n = 5) or cryoablation (n = 5). 3D LGE and T2-weighted images were analysed. We obtained two data sets of 3D whole heart LGE imaging with high resolution, 20minutes post gadolinium injection. Patients underwent two acquisitions: the conventional respiratory navigated, electrocardiographically-gated 3D fast gradient echo (1.3slice thickness, 256x256 matrix acquisition) and the image-navigated isotropic high resolution 3D LGE with Dixon water-fat separation. 3D LGE data were analysed using a dedicated (ADAS LA) software in order to produce a 3D-LA model visualization of LGE areas. Results In the acute stage, LGE volume was 50% larger after PFA vs. cryoablation (P &amp;lt; 0.001), and oedema on T2 imaging was significantly smaller (visual estimation) on PFA cases. LGE rings surrounding PV ostia were more homogeneous after PFA, with no sign of microvascular damage or intramural hemorrhage as compared to cryoablation cases. In the chronic stage, the majority of acute LGE had disappeared after PFA, whereas most LGE persisted after cryo-ablation. The two 3D LGE sequences used for analysis showed complete agreement in the visualization of atrial scar with superior image quality of the novel sequence. The mean acquisition time for 3D LGE imaging was 11±4minutes for the conventional fast gradient echo sequence, and 5±2minutes for the novel 3D self-image navigated LGE sequence. Conclusions CMR imaging may help in visualization and understanding of LA structural remodeling post different ablation techniques.

Cardiac magnetic resonance - aided ventricular tachycardia substrate ablation

Abstract Objectives To assess feasibility and reliability of Cardiac magnetic resonance (CMR)-derived pixel signal intensity (PSI) maps obtained by Three Dimensional (3D)- Late Gadolinium Enhancement (LGE) images for detecting potential targets for VT substrate ablation procedures. Background Parametric imaging with color-coded PSI maps obtained from pre-ablation 3D LGE -CMR imaging can be used for identification and characterization of arrhythmic abnormal substrate. Recent publications have shown that PSI maps permit visualization of border zone areas inside the scar. Specifically, corridors of viable tissue within the scar connecting healthy myocardium define VT corridors that may be the electrical equivalent of the abnormal conducting channels on the electroanatomical mapping (EAM). Methods 8 patients with scar-dependent monomorphic VTs who underwent 3D-LGE CMR imaging prior to substrate ablation were included in the study. The studies were performed at 1.5T scanner. We obtained two data sets of 3D whole heart LGE imaging with high resolution. Patients underwent two acquisitions: the conventional respiratory navigated, electrocardiographically-gated 3D fast gradient echo (1.3slice thickness, 256x256 matrix acquisition) and the image-navigated isotropic high resolution 3D GE with Dixon water-fat separation. 3D LGE-CMR images were processed with ADAS-VT software to produce a 3-dimensional model of the heart with 10 layers from the endocardium to epicardium and PSI maps color-coded from the LGE data projected to each of the shells (Figure 1). In the LGE-CMR PSI maps, VT corridors were obtained automatically by the ADAS VT software. Subsequently, data were given to the electrophysiologists and merged with the CARTO3 system EAM data obtained for the ablation (Figure2). CMR information was used to facilitate the identification of target ablation sites in the areas identified by CMR as abnormal substrate and radiofrequency was applied in the presence of both a pathological electrogram (EGM) and a VT corridor as identified by CMR. Results There was a 95% agreement between the abnormal conducting channels as detected by abnormal electrograms in the EAM and the VT corridors identified by 3D LGE CMR imaging. In two occasions, there was one VT corridor that did not correlate with abnormal electrograms and was not ablated. The two 3D LGE sequences used for ADAS-VT corridors analysis showed complete agreement in the number and location of VT corridors. Mean acquisition time for 3D LGE imaging was 11±4minutes for the conventional fast gradient echo sequence, and 5±2minutes for the novel 3D self-image navigated LGE sequence. The image quality of 3D LGE was superior with the Dixon fat-water separation. Overall CMR-aided VT ablation reduced the procedural and fluoroscopy time by 40%. Conclusions CMR-aided VT ablation is feasible, reliable and significantly reduces the procedural time.

#1458 Adverse muscle composition predicts all-cause mortality in CKD

Abstract Background and Aims Sarcopenia is common in CKD and associates with morbidity and mortality. A rapid and standardized MRI (Magnetic Resonance Imaging) scan protocol allows an accurate muscle composition analysis by quantification of skeletal muscle volume and muscle fat infiltration (MFI). The combined observation of low muscle volume and high muscle fat infiltration, i.e. adverse muscle composition (AMC), is a prevalent muscle composition phenotype within CKD and associates to poor function, high prevalence of comorbidity, and an increased risk for coronary heart disease (CHD). It has also been shown that AMC is an independent predictor of all-cause mortality in patients with non-alcoholic fatty liver disease and in general population. In this study we aimed to investigate muscle composition as predictor of all-cause mortality within participants with CKD in the UK Biobank imaging study (UKB). Method This is a prospective study including UKB participants with CKD defined as having an eGFR cystatin C &amp;lt; 60 ml/min/1, 73 m2. A control group was created by matching each participant with CKD for sex, age, and BMI to four participants with normal eGFR. Cystatin C was analyzed based on blood samples at time of study enrolment 7-9 years prior to the MRI. Fat-tissue free thigh muscle volume and MFI were quantified using a rapid neck-to-knee water and fat separated MRI protocol and automated image analysis (AMRA® Researcher). For each participant, a personalized muscle volume z-score (sex- and body size-specific) was calculated and combined with MFI to divide the participants into four muscle composition phenotype groups; 1) normal muscle composition, 2) only low muscle volume, 3) only high MFI and 4) AMC, using previously published thresholds based on the whole UKB cohort. The thresholds were the 25th percentile (&amp;lt; -0.68 SD) for muscle volume z-score and the 75th percentile (men &amp;gt;7.69%; women &amp;gt;8.82%) for MFI. The mortality data were obtained through the UKB´s linkage to national death registries. All-cause mortality was investigated using Kaplan-Meier curves and Cox regression. Models within participants with CKD were crude and subsequently adjusted for sex, age, BMI, low hand grip strength (&amp;lt;16/27 kg for females/males), physical activity, smoking, alcohol, and previous diagnosis of cancer, prevalent CHD and type 2 diabetes, all assessed at the time of imaging. Results Conclusion AMC, determined by standardized MRI, was a strong predictor of all-cause mortality in CKD. The study results indicate that CKD patients with poor muscle health, identified as having both low muscle volume z-score and high MFI, have a significantly higher mortality risk. This vulnerable group would potentially benefit from targeted interventions and is also of specific interest when evaluating new treatments for CKD.

Timing of magnetic resonance imaging in pregnancy for outcome prediction in congenital diaphragmatic hernia.

To evaluate the impact of the timing of MRI on the prediction of survival and morbidity in patients with CDH, and whether serial measurements have a beneficial value. This retrospective cohort study was conducted in two perinatal centers, in Germany and Italy. It included 354 patients with isolated CDH having at least one fetal MRI. The severity was assessed with the observed-to-expected total fetal lung volume (o/e TFLV) measured by two experienced double-blinded operators. The cohort was divided into three groups according to the gestational age (GA) at which the MRI was performed (< 27, 27-32, and > 32weeks' gestation [WG]). The accuracy for the prediction of survival at discharge and morbidity was analyzed with receiver operating characteristic (ROC) curves. Multiple logistic regression analyses and propensity score matching examined the population for balance. The effect of repeated MRI was evaluated in ninety-seven cases. There were no significant differences in the prediction of survival when the o/e TFLV was measured before 27, between 27 and 32, and after 32 WG (area under the curve [AUC]: 0.77, 0.79, and 0.77, respectively). After adjustment for confounding factors, it was seen, that GA at MRI was not associated with survival at discharge, but the risk of mortality was higher with an intrathoracic liver position (adjusted odds ratio [aOR]: 0.30, 95% confidence interval [95%CI] 0.12-0.78), lower GA at birth (aOR 1.48, 95%CI 1.24-1.78) and lower o/e TFLV (aOR 1.13, 95%CI 1.06-1.20). ROC curves showed comparable prediction accuracy for the different timepoints in pregnancy for pulmonary hypertension, the need of extracorporeal membrane oxygenation, and feeding aids. Serial measurements revealed no difference in change rate of the o/e TFLV according to survival. The timing of MRI does not affect the prediction of survival rate or morbidity as the o/e TFLV does not change during pregnancy. Clinicians could choose any gestational age starting mid second trimester for the assessment of severity and counseling.

Dispossessed lands and land-use change in the Colombian armed conflict: Exploring a link through a regional case study

Land dispossession in the Colombian armed conflict is a critical form of victimization and a trigger of territorial configurations that shape the rurality of the country. The consequences of the illegal appropriation of land by means of usurpation or forced displacement result in a humanitarian drama and various socio-spatial and socio-economic transformations. This research analyzes the intricate dynamics of land use and cover changes in four municipalities within the Cesar department and its relationship with instances of land abandonment and dispossession over a 29-year period (1991–2020). Landsat satellite image time series of TM, ETM+ and OLI data were used to determine land use and land cover (LULC). The classification process was carried out using the Random Forest model. The dynamics of land dispossession and forced abandonment were identified by analyzing 195 land restitution court rulings. Geo-referenced properties were subsequently overlaid on the created LULC maps. As a result, a pattern indicating a direct relationship between both variables was found. Consequently, the phenomenon of land dispossession in the context of this internal armed conflict has contributed to the transition from the traditional agrarian model to the agro-industrial one.

Synthesizing 4D Magnetic Resonance Angiography From 3D Time-of-Flight Using Deep Learning: A Feasibility Study.

Objective and background This study aimed to develop a deep convolutional neural network (DCNN) model capable of generating synthetic 4D magnetic resonance angiography (MRA) from 3D time-of-flight (TOF) images, allowing estimation of temporal changes in arterial flow. TOF MRA provides static information about arterial structures through maximum intensity projection (MIP) processing, but it does not capture the dynamic information of contrast agent circulation, which is lost during MIP processing. Considering the principles of TOF, it is hypothesized that dynamic information about arterial blood flow is latent within TOF signals. Although arterial spin labeling (ASL) can extract dynamic arterial information, ASL MRA has drawbacks, such as longer imaging times and lower spatial resolution than TOF MRA. This study's primary aim is to extend the utility of TOF MRA by training a machine-learning model on paired TOF and ASL data to extract latent dynamic information from TOF signals. Methods A DCNN combining a modified U-Net and a long-short-term memory (LSTM) network was trained on a dataset of 13 subjects (11 menand two women, aged 42-77 years) using paired 3D TOF MRA and 4D ASL MRA images. Subjects had no history of cerebral vessel occlusion or significant stenosis. The dataset was acquired using a 3T MRI system with a 32-channel head coil. Preprocessing involved resampling and intensity normalization of TOF and ASL images, followed by data augmentation and arterial mask generation. The model learned to extract flow information from TOF images and generate 8-phase 4D MRA images. The precision of flow estimation was evaluated using the coefficient of determination (R²) and Bland-Altman analysis. A board-certified neuroradiologist validated the quality of the images and the absence of significant stenosis in the major cerebral arteries. Results The generated 4D MRA images closely resembled the ground-truth ASL MRA data, with R² values of 0.92, 0.85, and 0.84 for the internal carotid artery (ICA), proximal middle cerebral artery (MCA), and distal MCA, respectively. Bland-Altman analysis revealed a systematic error of -0.06, with 95% agreement limits ranging from -0.18 to 0.12. Additionally, the model successfully identified flow abnormalities in a subject with left MCA stenosis, displaying a delayed peak and subsequent flattening distal to the stenosis, indicative of reduced blood flow. Visualization of the predicted arterial flow overlaid on the original TOF MRA images highlighted the spatial progression and dynamics of the flow. Conclusions The DCNN model effectively generated synthetic 4D MRA images from TOF images, demonstrating its potential to estimate temporal changes in arterial flow accurately. This non-invasive technique offers a promising alternative to conventional methods for visualizing and evaluating healthy and pathological flow dynamics. It has significant potential to improve the diagnosis and treatment of cerebrovascular diseases by providing detailed temporal flow information without the need for contrast agents or invasive procedures. The practical implementation of this model could enable the extraction of dynamic cerebral blood flow information from routine brain MRI examinations, contributing to the early diagnosis and management of cerebrovascular disorders.

Rapid Ultrasonography for Shock and Hypotension Protocol Performed using Handheld Ultrasound Devices by Paramedics in a Moving Ambulance: Evaluation of Image Accuracy and Time in Motion.

Handheld ultrasound (US) devices have become increasingly popular since the early 2000s due to their portability and affordability compared to conventional devices. The Rapid Ultrasonography for Shock and Hypotension (RUSH) protocol, introduced in 2009, has shown promising accuracy rates when performed with handheld devices. However, there are limited data on the accuracy of such examinations performed in a moving ambulance. This study aimed to assess the feasibility and accuracy of the RUSH protocol performed by paramedics using handheld US devices in a moving ambulance. The study aimed to examine the performability of the RUSH protocol with handheld US devices in a moving ambulance and to evaluate the accuracy of diagnostic views obtained within an appropriate time frame. A prospective study was conducted with paramedics who underwent theoretical and practical training in the RUSH protocol. The participants performed the protocol using a handheld US device in both stationary and moving ambulances. Various cardiac and abdominal views were obtained and evaluated for accuracy. The duration of the protocol performance was recorded for each participant. Nine paramedics completed the study, with 18 performances each in both stationary and moving ambulance groups. The accuracy of diagnostic views obtained during the RUSH protocol did not significantly differ between the stationary and moving groups. However, the duration of protocol performance was significantly shorter in the moving group compared to the stationary group. Paramedics demonstrated the ability to perform the RUSH protocol effectively using handheld US devices in both stationary and moving ambulances following standard theoretical and practical training. The findings suggest that ambulance movement does not significantly affect the accuracy of diagnostic views obtained during the protocol. Further studies with larger sample sizes are warranted to validate these findings and explore the potential benefits of prehospital US in dynamic environments.

A multi-institutional meningioma MRI dataset for automated multi-sequence image segmentation

Meningiomas are the most common primary intracranial tumors and can be associated with significant morbidity and mortality. Radiologists, neurosurgeons, neuro-oncologists, and radiation oncologists rely on brain MRI for diagnosis, treatment planning, and longitudinal treatment monitoring. However, automated, objective, and quantitative tools for non-invasive assessment of meningiomas on multi-sequence MR images are not available. Here we present the BraTS Pre-operative Meningioma Dataset, as the largest multi-institutional expert annotated multilabel meningioma multi-sequence MR image dataset to date. This dataset includes 1,141 multi-sequence MR images from six sites, each with four structural MRI sequences (T2-, T2/FLAIR-, pre-contrast T1-, and post-contrast T1-weighted) accompanied by expert manually refined segmentations of three distinct meningioma sub-compartments: enhancing tumor, non-enhancing tumor, and surrounding non-enhancing T2/FLAIR hyperintensity. Basic demographic data are provided including age at time of initial imaging, sex, and CNS WHO grade. The goal of releasing this dataset is to facilitate the development of automated computational methods for meningioma segmentation and expedite their incorporation into clinical practice, ultimately targeting improvement in the care of meningioma patients.

Enhanced Detection of Artisanal Small-Scale Mining with Spectral and Textural Segmentation of Landsat Time Series

Artisanal small-scale mines (ASMs) in the Amazon Rainforest are an important cause of deforestation, forest degradation, biodiversity loss, sedimentation in rivers, and mercury emissions. Satellite image data are widely used in environmental decision-making to monitor changes in the land surface, but ASMs are difficult to map from space. ASMs are small, irregularly shaped, unevenly distributed, and confused (spectrally) with other land clearance types. To address this issue, we developed a reliable and efficient ASM detection method for the Tapajós River Basin of Brazil—an important gold mining region of the Amazon Rainforest. We enhanced detection in three key ways. First, we used the time-series segmentation (LandTrendr) Google Earth Engine (GEE) Application Programming Interface to map the pixel-wise trajectory of natural vegetation disturbance and recovery on an annual basis with a 2000 to 2019 Landsat image time series. Second, we segmented 26 textural features in addition to 5 spectral features to account for the high spatial heterogeneity in ASM pixels. Third, we trained and tested a Random Forest model to detect ASMs after eliminating irrelevant and redundant features with the Variable Selection Using Random Forests “ensemble of ensembles” technique. The out-of-bag error and overall accuracy of the final Random Forest was 3.73 and 92.6%, which are comparable to studies mapping large industrial mines with the normalized difference vegetation index (NDVI) and LandTrendr. The most important feature in our study was NDVI, followed by textural features in the near and shortwave infrared. Our work paves the way for future ASM regulation through large area monitoring from space with free and open-source GEE and operational satellites. Studies with sufficient computational resources can improve ASM monitoring with advanced sensors consisting of spectral narrow bands (Sentinel-2, Environmental Mapping and Analysis Program, PRecursore IperSpettrale della Missione Applicativa) and deep learning.

High-speed light-sheet fluorescent microscopy with an adaptive three-dimensional region-of-signal identification method

Light-sheet fluorescent microscopy (LSFM) is a crucial 3D imaging technique for tissues with advantages in high 3D spatial resolution, effective optical sectioning, and quick imaging. Prior to imaging, the sample's imaging depth, imaging region, and region of interest (ROI) division should all be established in accordance with the sample's outermost dimensions. However, because biological samples are inherently irregular a traditional imaging flow that images ROI by ROI, slice by slice, would waste a great deal of time and data storage space by additionally recording non-signal regions. In order to solve this problem, we create a high-speed LSFM with an adaptive three-dimensional region-of-signal identification method (RSI). We immediately perform crude imaging for each ROI before the software analyzes the longitudinal distribution and, ultimately, uses dimensionality reduction, binarization, filtering, and redundancy expansion to acquire the signal region result. By imaging two mouse spinal cord segments with a total of 170 ROIs and 180 ROIs on two channels, respectively, the high-speed imaging capability is validated. Our method reduced the overall imaging time and raw data size by more than treble and 14 times, respectively, without losing signal. With a programmable imaging range based on the real signal region, this method offers an efficient tool for flexible light-sheet microscopy. It can also be applied to other microscopy techniques to speed up imaging.