Comparison of Total Variation with a Motion Estimation Based Compressed Sensing Approach for Self-Gated Cardiac Cine MRI in Small Animal Studies

Purpose

Compressed sensing (CS) has been widely applied to prospective cardiac cine MRI. The aim of this work is to study the benefits obtained by including motion estimation in the CS framework for small-animal retrospective cardiac cine.

Methods

We propose a novel B-spline-based compressed sensing method (SPLICS) that includes motion estimation and generalizes previous spatiotemporal total variation (ST-TV) methods by taking into account motion between frames. In addition, we assess the effect of an optimum weighting between spatial and temporal sparsity to further improve results. Both methods were implemented using the efficient Split Bregman methodology and were evaluated on rat data comparing animals with myocardial infarction with controls for several acceleration factors.

Results

ST-TV with optimum selection of the weighting sparsity parameter led to results similar to those of SPLICS; ST-TV with large relative temporal sparsity led to temporal blurring effects. However, SPLICS always properly corrected temporal blurring, independently of the weighting parameter. At acceleration factors of 15, SPLICS did not distort temporal intensity information but led to some artefacts and slight over-smoothing. At an acceleration factor of 7, images were reconstructed without significant loss of quality.

Conclusion

We have validated SPLICS for retrospective cardiac cine in small animal, achieving high acceleration factors. In addition, we have shown that motion modelling may not be essential for retrospective cine and that similar results can be obtained by using ST-TV provided that an optimum selection of the spatiotemporal sparsity weighting parameter is performed.     

Estimating leaf area index using unmanned aerial vehicle data: shallow vs. deep machine learning algorithms

Measuring leaf area index (LAI) is essential for evaluating crop growth and estimating yield, thereby facilitating high-throughput phenotyping of maize (Zea mays). LAI estimation models use multi-source data from unmanned aerial vehicles (UAVs), but using multimodal data to estimate maize LAI, and the effect of tassels and soil background, remain understudied. Our research aims to (1) determine how multimodal data contribute to LAI and propose a framework for estimating LAI based on remote-sensing data, (2) evaluate the robustness and adaptability of an LAI estimation model that uses multimodal data fusion and deep neural networks (DNNs) in single- and whole growth stages, and (3) explore how soil background and maize tasseling affect LAI estimation. To construct multimodal datasets, our UAV collected red–green–blue, multispectral, and thermal infrared images. We then developed partial least square regression (PLSR), support vector regression, and random forest regression models to estimate LAI. We also developed a deep learning model with three hidden layers. This multimodal data structure accurately estimated maize LAI. The DNN model provided the best estimate (coefficient of determination [R²] = 0.89, relative root mean square error [rRMSE] = 12.92%) for a single growth period, and the PLSR model provided the best estimate (R² = 0.70, rRMSE = 12.78%) for a whole growth period. Tassels reduced the accuracy of LAI estimation, but the soil background provided additional image feature information, improving accuracy. These results indicate that multimodal data fusion using low-cost UAVs and DNNs can accurately and reliably estimate LAI for crops, which is valuable for high-throughput phenotyping and high-spatial precision farmland management.

Multimodal data fusion (red–green–blue, multispectral, and thermal infrared) using low-cost unmanned aerial vehicles in a deep neural network and machine learning framework estimates maize leaf area index     

Parasite prevalence and the worldwide distribution of cognitive ability

In this study, we hypothesize that the worldwide distribution of cognitive ability is determined in part by variation in the intensity of infectious diseases. From an energetics standpoint, a developing human will have difficulty building a brain and fighting off infectious diseases at the same time, as both are very metabolically costly tasks. Using three measures of average national intelligence quotient (IQ), we found that the zero-order correlation between average IQ and parasite stress ranges from r = −0.76 to r = −0.82 (p < 0.0001). These correlations are robust worldwide, as well as within five of six world regions. Infectious disease remains the most powerful predictor of average national IQ when temperature, distance from Africa, gross domestic product per capita and several measures of education are controlled for. These findings suggest that the Flynn effect may be caused in part by the decrease in the intensity of infectious diseases as nations develop.     

On estimating intraventricular hemodynamic forces from endocardial dynamics: A comparative study with 4D flow MRI

Intraventricular pressure gradients or hemodynamic forces, which are their global measure integrated over the left ventricular volume, have a fundamental importance in ventricular function. They may help revealing a sub-optimal cardiac function that is not evident in terms of tissue motion, which is naturally heterogeneous and variable, and can influence cardiac adaptation. However, hemodynamic forces are not utilized in clinical cardiology due to the unavailability of simple non-invasive measurement tools.Hemodynamic forces depend on the intraventricular flow; nevertheless, most of them are imputable to the dynamics of the endocardial flow boundary and to the exchange of momentum across the mitral and aortic orifices. In this study, we introduce a simplified model based on first principles of fluid dynamics that allows estimating hemodynamic forces without knowing the velocity field inside the LV.The model is validated with 3D phase-contrast MRI (known as 4D flow MRI) in 15 subjects, (5 healthy and 10 patients) using the endocardial surface reconstructed from the three standard long-axis projections. Results demonstrate that the model provides consistent estimates for the base-apex component (mean correlation coefficient r = 0.77 for instantaneous values and r = 0.88 for root mean square) and good estimates of the inferolateral-anteroseptal component (r = 0.50 and 0.84, respectively).The present method represents a potential integration to the existing ones quantifying endocardial deformation in MRI and echocardiography to add a physics-based estimation of the corresponding hemodynamic forces. These could help the clinician to early detect sub-clinical diseases and differentiate between different cardiac dysfunctional states.     

Functional group bridge for simultaneous regression and support estimation

This paper is motivated by studying differential brain activities to multiple experimental condition presentations in intracranial electroencephalography (iEEG) experiments. Contrasting effects of experimental conditions are often zero in most regions and nonzero in some local regions, yielding locally sparse functions. Such studies are essentially a function-on-scalar regression problem, with interest being focused not only on estimating nonparametric functions but also on recovering the function supports. We propose a weighted group bridge approach for simultaneous function estimation and support recovery in function-on-scalar mixed effect models, while accounting for heterogeneity present in functional data. We use B-splines to transform sparsity of functions to its sparse vector counterpart of increasing dimension, and propose a fast nonconvex optimization algorithm using nested alternative direction method of multipliers (ADMM) for estimation. Large sample properties are established. In particular, we show that the estimated coefficient functions are rate optimal in the minimax sense under the L₂ norm and resemble a phase transition phenomenon. For support estimation, we derive a convergence rate under the $L_{\infty }$ norm that leads to a selection consistency property under δ-sparsity, and obtain a result under strict sparsity using a simple sufficient regularity condition. An adjusted extended Bayesian information criterion is proposed for parameter tuning. The developed method is illustrated through simulations and an application to a novel iEEG data set to study multisensory integration.     

Suppressing Multi-Channel Ultra-Low-Field MRI Measurement Noise Using Data Consistency and Image Sparsity

Ultra-low-field (ULF) MRI (B ₀ = 10–100 µT) typically suffers from a low signal-to-noise ratio (SNR). While SNR can be improved by pre-polarization and signal detection using highly sensitive superconducting quantum interference device (SQUID) sensors, we propose to use the inter-dependency of the k-space data from highly parallel detection with up to tens of sensors readily available in the ULF MRI in order to suppress the noise. Furthermore, the prior information that an image can be sparsely represented can be integrated with this data consistency constraint to further improve the SNR. Simulations and experimental data using 47 SQUID sensors demonstrate the effectiveness of this data consistency constraint and sparsity prior in ULF-MRI reconstruction.     

A compound decision approach to covariance matrix estimation

Covariance matrix estimation is a fundamental statistical task in many applications, but the sample covariance matrix is suboptimal when the sample size is comparable to or less than the number of features. Such high-dimensional settings are common in modern genomics, where covariance matrix estimation is frequently employed as a method for inferring gene networks. To achieve estimation accuracy in these settings, existing methods typically either assume that the population covariance matrix has some particular structure, for example, sparsity, or apply shrinkage to better estimate the population eigenvalues. In this paper, we study a new approach to estimating high-dimensional covariance matrices. We first frame covariance matrix estimation as a compound decision problem. This motivates defining a class of decision rules and using a nonparametric empirical Bayes g-modeling approach to estimate the optimal rule in the class. Simulation results and gene network inference in an RNA-seq experiment in mouse show that our approach is comparable to or can outperform a number of state-of-the-art proposals.     

Intelligence May Moderate the Cognitive Profile of Patients with ASD

BackgroundThe intelligence of individuals with Autism Spectrum Disorder (ASD) varies considerably. The pattern of cognitive deficits associated with ASD may differ depending on intelligence. We aimed to study the absolute and relative severity of cognitive deficits in participants with ASD in relation to IQ.MethodsA total of 274 children (M age = 12.1, 68.6% boys) participated: 30 ASD and 22 controls in the below average Intelligence Quotient (IQ) group (IQ<85), 57 ASD and 54 controls in the average IQ group (85<IQ<115) and 41 ASD and 70 controls in the above average IQ group (IQ>115). Matching for age, sex, Full Scale IQ (FSIQ), Verbal IQ (VIQ), Performance IQ (PIQ) and VIQ-PIQ difference was performed. Speed and accuracy of social cognition, executive functioning, visual pattern recognition and basic processing speed were examined per domain and as a composite score.ResultsThe composite score revealed a trend significant IQ by ASD interaction (significant when excluding the average IQ group). In absolute terms, participants with below average IQs performed poorest (regardless of diagnosis). However, in relative terms, above average intelligent participants with ASD showed the most substantial cognitive problems (particularly for social cognition, visual pattern recognition and verbal working memory) since this group differed significantly from the IQ-matched control group (p < .001), whereas this was not the case for below-average intelligence participants with ASD (p = .57).ConclusionsIn relative terms, cognitive deficits appear somewhat more severe in individuals with ASD and above average IQs compared to the below average IQ patients with ASD. Even though high IQ ASD individuals enjoy a certain protection from their higher IQ, they clearly demonstrate cognitive impairments that may be targeted in clinical assessment and treatment. Conversely, even though in absolute terms ASD patients with below average IQs were clearly more impaired than ASD patients with average to above average IQs, the differences in cognitive functioning between participants with and without ASD on the lower end of the IQ spectrum were less pronounced. Clinically this may imply that cognitive assessment and training of cognitive skills in below average intelligent children with ASD may be a less fruitful endeavour. These findings tentatively suggest that intelligence may act as a moderator in the cognitive presentation of ASD, with qualitatively different cognitive processes affected in patients at the high and low end of the IQ spectrum.     

Kinetic sequencing (k-Seq) as a massively parallel assay for ribozyme kinetics: utility and critical parameters

Characterizing genotype-phenotype relationships of biomolecules (e.g. ribozymes) requires accurate ways to measure activity for a large set of molecules. Kinetic measurement using high-throughput sequencing (e.g. k-Seq) is an emerging assay applicable in various domains that potentially scales up measurement throughput to over 10⁶ unique nucleic acid sequences. However, maximizing the return of such assays requires understanding the technical challenges introduced by sequence heterogeneity and DNA sequencing. We characterized the k-Seq method in terms of model identifiability, effects of sequencing error, accuracy and precision using simulated datasets and experimental data from a variant pool constructed from previously identified ribozymes. Relative abundance, kinetic coefficients, and measurement noise were found to affect the measurement of each sequence. We introduced bootstrapping to robustly quantify the uncertainty in estimating model parameters and proposed interpretable metrics to quantify model identifiability. These efforts enabled the rigorous reporting of data quality for individual sequences in k-Seq experiments. Here we present detailed protocols, define critical experimental factors, and identify general guidelines to maximize the number of sequences and their measurement accuracy from k-Seq data. Analogous practices could be applied to improve the rigor of other sequencing-based assays.     

How sensitive are climatic niche inferences to distribution data sampling? A comparison of Biota of North America Program (BONAP) and Global Biodiversity Information Facility (GBIF) datasets

Climatic niche modeling is widely used in modern macroecology and evolutionary biology to model species' distributions and ecological niches. Frequently, Global Biodiversity Information Facility (GBIF) distribution data are used as raw data for such models. Unfortunately, the accuracy of resulting niche estimates is difficult to assess, and GBIF users continue to call for a better understanding of GBIF data precision and uncertainty. Our research evaluates how GBIF data perform in comparison with curated, county-level species distributions from the Biota of North America Program (BONAP; www.bonap.org) to estimate the climatic niche for Carex (Cyperaceae), one of the largest angiosperm genera in the temperate zone. In particular, we investigate the complementary strengths and weaknesses of the two datasets in the context of climatic niche estimation, namely (1) the incomplete sampling coverage of species distributions of GBIF data, and (2) elevation bias in climatic niche estimates resulting from the coarse resolution of the BONAP distribution dataset. To do so, we quantified climatic niches for 388 North American Carex species and calculated the distance of GBIF and BONAP climatic niche estimates in principal component space. We found little to no relationship between differences in climatic niche estimates and sampling coverage metrics, suggesting that sparse sampling coverage of GBIF data may have negligible average effects on mean climate estimates at the species level. However, elevation had a significant effect on differences in niche estimates, suggesting that using county-level distribution data—in our study, represented by BONAP—may introduce a bias in estimated climatic niche. To investigate if any such bias is phylogenetically structured, we examined the relationship between GBIF and BONAP climatic niche tip states using Phylogenetic Generalized Least Squares Regression (PGLS), estimating phylogenetic signal in the residuals of the regression by estimating Pagel's λ simultaneously with other regression parameters. Estimates were tightly correlated, with little to no phylogenetic signal—low λ—in the regression residuals, suggesting that any potential bias is more or less independent of phylogeny for the sedges of North America. Based on our results, we recommend a hybrid GBIF and BONAP data approach for the best understanding of species' climatic niche, relying on BONAP to fill in distributions primarily where collection records may be sparse. Our findings provide needed context and perspective on the implications of alternative distribution datasets for climatic niche estimation, especially in a macroevolutionary context.     

Individual differences in the perception of probability

In recent studies of humans estimating non-stationary probabilities, estimates appear to be unbiased on average, across the full range of probability values to be estimated. This finding is surprising given that experiments measuring probability estimation in other contexts have often identified conservatism: individuals tend to overestimate low probability events and underestimate high probability events. In other contexts, repulsive biases have also been documented, with individuals producing judgments that tend toward extreme values instead. Using extensive data from a probability estimation task that produces unbiased performance on average, we find substantial biases at the individual level; we document the coexistence of both conservative and repulsive biases in the same experimental context. Individual biases persist despite extensive experience with the task, and are also correlated with other behavioral differences, such as individual variation in response speed and adjustment rates. We conclude that the rich computational demands of our task give rise to a variety of behavioral patterns, and that the apparent unbiasedness of the pooled data is an artifact of the aggregation of heterogeneous biases.     

Evaluation of automatic image quality assessment in chest CT – A human cadaver study

PurposeThe evaluation of clinical image quality (IQ) is important to optimize CT protocols and to keep patient doses as low as reasonably achievable. Considering the significant amount of effort needed for human observer studies, automatic IQ tools are a promising alternative. The purpose of this study was to evaluate automatic IQ assessment in chest CT using Thiel embalmed cadavers.MethodsChest CT’s of Thiel embalmed cadavers were acquired at different exposures. Clinical IQ was determined by performing a visual grading analysis. Physical-technical IQ (noise, contrast-to-noise and contrast-detail) was assessed in a Catphan phantom. Soft and sharp reconstructions were made with filtered back projection and two strengths of iterative reconstruction. In addition to the classical IQ metrics, an automatic algorithm was used to calculate image quality scores (IQs). To be able to compare datasets reconstructed with different kernels, the IQs values were normalized.ResultsGood correlations were found between IQs and the measured physical-technical image quality: noise (ρ = −1.00), contrast-to-noise (ρ = 1.00) and contrast-detail (ρ = 0.96). The correlation coefficients between IQs and the observed clinical image quality of soft and sharp reconstructions were 0.88 and 0.93, respectively.ConclusionsThe automatic scoring algorithm is a promising tool for the evaluation of thoracic CT scans in daily clinical practice. It allows monitoring of the image quality of a chest protocol over time, without human intervention. Different reconstruction kernels can be compared after normalization of the IQs.     

Comparison of Dixon Sequences for Estimation of Percent Breast Fibroglandular Tissue

Objectives

To evaluate sources of error in the Magnetic Resonance Imaging (MRI) measurement of percent fibroglandular tissue (%FGT) using two-point Dixon sequences for fat-water separation.

Methods

Ten female volunteers (median age: 31 yrs, range: 23–50 yrs) gave informed consent following Research Ethics Committee approval. Each volunteer was scanned twice following repositioning to enable an estimation of measurement repeatability from high-resolution gradient-echo (GRE) proton-density (PD)-weighted Dixon sequences. Differences in measures of %FGT attributable to resolution, T₁ weighting and sequence type were assessed by comparison of this Dixon sequence with low-resolution GRE PD-weighted Dixon data, and against gradient-echo (GRE) or spin-echo (SE) based T₁-weighted Dixon datasets, respectively.

Results

%FGT measurement from high-resolution PD-weighted Dixon sequences had a coefficient of repeatability of ±4.3%. There was no significant difference in %FGT between high-resolution and low-resolution PD-weighted data. Values of %FGT from GRE and SE T₁-weighted data were strongly correlated with that derived from PD-weighted data (r = 0.995 and 0.96, respectively). However, both sequences exhibited higher mean %FGT by 2.9% (p < 0.0001) and 12.6% (p < 0.0001), respectively, in comparison with PD-weighted data; the increase in %FGT from the SE T₁-weighted sequence was significantly larger at lower breast densities.

Conclusion

Although measurement of %FGT at low resolution is feasible, T₁ weighting and sequence type impact on the accuracy of Dixon-based %FGT measurements; Dixon MRI protocols for %FGT measurement should be carefully considered, particularly for longitudinal or multi-centre studies.     

Validating Excised Rodent Lungs for Functional Hyperpolarized Xenon-129 MRI

Ex vivo rodent lung models are explored for physiological measurements of respiratory function with hyperpolarized (hp) ¹²⁹Xe MRI. It is shown that excised lung models allow for simplification of the technical challenges involved and provide valuable physiological insights that are not feasible using in vivo MRI protocols. A custom designed breathing apparatus enables MR images of gas distribution on increasing ventilation volumes of actively inhaled hp ¹²⁹Xe. Straightforward hp ¹²⁹Xe MRI protocols provide residual lung volume (RV) data and permit for spatially resolved tracking of small hp ¹²⁹Xe probe volumes during the inhalation cycle. Hp ¹²⁹Xe MRI of lung function in the excised organ demonstrates the persistence of post mortem airway responsiveness to intravenous methacholine challenges. The presented methodology enables physiology of lung function in health and disease without additional regulatory approval requirements and reduces the technical and logistical challenges with hp gas MRI experiments. The post mortem lung functional data can augment histological measurements and should be of interest for drug development studies.     

Construction of Predictive Models to Describe Apparent and Complex Viscosity Values of O/W Model System Meat Emulsions Using Adaptive Neuro – Fuzzy Inference System (ANFIS) and Artificial Neural Networks (ANN)

This paper introduces an adaptive neuro ?C fuzzy inference system (ANFIS) and artificial neural networks (ANN) models to predict the apparent and complex viscosity values of model system meat emulsions. Constructed models were compared with multiple linear regression (MLR) modeling based on their estimation performance. The root mean square error (RMSE), mean absolute error (MAE) and determination coefficient (R ²) statistics were performed to evaluate the accuracy of the models tested. Comparison of the models showed that the ANFIS model performed better than the ANN and MLR models to estimate the apparent and complex viscosity values of the model system meat emulsions. Coefficients of determination (R ²) calculated for estimation performance of ANFIS modeling to predict apparent and complex viscosity of the emulsions were 0.996 and 0.992, respectively. Similar R ² values (0.991 and 0.985) were obtained when estimating the performance of the ANN model. In the present study, use of the constructed ANFIS models can be suggested to effectively predict the apparent and complex viscosity values of model system meat emulsions.     

The return of sleep

While most individuals spend a substantial amount of time sleeping, Japan is among the developed countries whose residents register the fewest number of hours of sleep. I thus examine the causal effects of sleep on labor productivity, utilizing panel datasets for Japanese men. The potential endogeneity of deciding how many hours to sleep is addressed by using fixed effects panel data models with an instrumental variable estimation technique. Exploiting the annual variation in the average sunshine duration between cities as an instrument, I then show that a one-hour increase in the weekly number of hours of sleep increases the wage rate by up to 6–8% on average. These results suggest that sleep duration could enhance labor productivity.     

Network-assisted investigation of combined causal signals from genome-wide association studies in schizophrenia

With the recent success of genome-wide association studies (GWAS), a wealth of association data has been accomplished for more than 200 complex diseases/traits, proposing a strong demand for data integration and interpretation. A combinatory analysis of multiple GWAS datasets, or an integrative analysis of GWAS data and other high-throughput data, has been particularly promising. In this study, we proposed an integrative analysis framework of multiple GWAS datasets by overlaying association signals onto the protein-protein interaction network, and demonstrated it using schizophrenia datasets. Building on a dense module search algorithm, we first searched for significantly enriched subnetworks for schizophrenia in each single GWAS dataset and then implemented a discovery-evaluation strategy to identify module genes with consistent association signals. We validated the module genes in an independent dataset, and also examined them through meta-analysis of the related SNPs using multiple GWAS datasets. As a result, we identified 205 module genes with a joint effect significantly associated with schizophrenia; these module genes included a number of well-studied candidate genes such as DISC1, GNA12, GNA13, GNAI1, GPR17, and GRIN2B. Further functional analysis suggested these genes are involved in neuronal related processes. Additionally, meta-analysis found that 18 SNPs in 9 module genes had P _meta<1×10⁻⁴, including the gene HLA-DQA1 located in the MHC region on chromosome 6, which was reported in previous studies using the largest cohort of schizophrenia patients to date. These results demonstrated our bi-directional network-based strategy is efficient for identifying disease-associated genes with modest signals in GWAS datasets. This approach can be applied to any other complex diseases/traits where multiple GWAS datasets are available.