Quantification Errors Research Articles

Comparing three quantification methods on N/Si ratio analysis using electron energy loss spectroscopy (EELS)

Precision elemental analysis by EELS at a nano-meter scale is a desirable goal for industrial application, as well as basic scientific research on advanced material analysis. However, sample thickness variation leads to a systematic error in element quantification due to electron plural scattering events and prevents it to be a useful quantitative analytic method. In this paper, we compared three methods to minimize thickness effect: Fourier-log deconvolution, Jump ratio, and Si K-edge by quantifying the N/Si ratio versus different thickness on a standard Si3N4 crystal sample. The results indicated that Jump ratio method exceeded other two methods owing to its higher accuracy, short data acquisition time as well as minimum requirement for spectroscopy instrument. The jump ratio method was applied to calculate N/Si ratio in storage nitride layer of 3D NAND trench wafer and blanket wafer, both illustrated that the jump ratio method successfully corrected the thickness effect on EELS composition quantification and largely reduced the relative error compared with the non-corrected result.

Ensemble Convolutional Neural Network Classification for Pancreatic Steatosis Assessment in Biopsy Images

Non-alcoholic fatty pancreas disease (NAFPD) is a common and at the same time not extensively examined pathological condition that is significantly associated with obesity, metabolic syndrome, and insulin resistance. These factors can lead to the development of critical pathogens such as type-2 diabetes mellitus (T2DM), atherosclerosis, acute pancreatitis, and pancreatic cancer. Until recently, the diagnosis of NAFPD was based on noninvasive medical imaging methods and visual evaluations of microscopic histological samples. The present study focuses on the quantification of steatosis prevalence in pancreatic biopsy specimens with varying degrees of NAFPD. All quantification results are extracted using a methodology consisting of digital image processing and transfer learning in pretrained convolutional neural networks for the detection of histological fat structures. The proposed method is applied to 20 digitized histological samples, producing an 0.08% mean fat quantification error thanks to an ensemble CNN voting system and 83.3% mean Dice fat segmentation similarity compared to the semi-quantitative estimates of specialist physicians.

Fast, Accurate, and Robust T2 Mapping of Articular Cartilage by Neural Networks.

For T2 mapping, the underlying mono-exponential signal decay is traditionally quantified by non-linear Least-Squares Estimation (LSE) curve fitting, which is prone to outliers and computationally expensive. This study aimed to validate a fully connected neural network (NN) to estimate T2 relaxation times and to assess its performance versus LSE fitting methods. To this end, the NN was trained and tested in silico on a synthetic dataset of 75 million signal decays. Its quantification error was comparatively evaluated against three LSE methods, i.e., traditional methods without any modification, with an offset, and one with noise correction. Following in-situ acquisition of T2 maps in seven human cadaveric knee joint specimens at high and low signal-to-noise ratios, the NN and LSE methods were used to estimate the T2 relaxation times of the manually segmented patellofemoral cartilage. In-silico modeling at low signal-to-noise ratio indicated significantly lower quantification error for the NN (by medians of 6–33%) than for the LSE methods (p < 0.001). These results were confirmed by the in-situ measurements (medians of 10–35%). T2 quantification by the NN took only 4 s, which was faster than the LSE methods (28–43 s). In conclusion, NNs provide fast, accurate, and robust quantification of T2 relaxation times.

Quantify pixel-level detection of dam surface crack using deep learning

An increasing number of detection methods based on computer vision are applied to detect cracks in water conservancy infrastructure. However, most studies directly use existing feature extraction networks to extract crack information, which are proposed for open-source datasets. As the crack distribution and pixel features are different from these data, the extracted crack information is incomplete. In this paper, a deep learning-based network for dam surface crack detection is proposed, which mainly addresses the semantic segmentation of cracks on the dam surface. Particularly, we design a shallow encoding network to extract features of crack images based on the statistical analysis of cracks. Further, to enhance the relevance of contextual information, we introduce an attention module into the decoding network. During the training, we use the sum of cross-entropy and Dice loss as the loss function to overcome data imbalance. The quantitative crack information is extracted by the imaging principle after using morphological algorithms to extract the morphological features of the predicted result. We built a manual annotation dataset containing 1577 images to verify the effectiveness of the proposed method. This method achieves state-of-the-art performance on our dataset. Specifically, the precision, recall, Intersection of Union (IoU), F1_measure, and accuracy are 90.81%, 81.54%, 75.23%, 85.93%, 99.76%, respectively, and the quantification error of cracks is less than 4%.

Development of a Physics-Informed Doubly Fed Cross-Residual Deep Neural Network for High-Precision Magnetic Flux Leakage Defect Size Estimation

Defect depth is an essential indicator in magnetic flux leakage (MFL) detection and estimation. The quantification errors for defect depth are closely related to length and width errors, and this feature has always been used to support the operator's judgment in defect identification. However, the existing defect quantification algorithms based on shallow and deep neural networks only employed simple general network structures inspired by the field of artificial intelligence; consequently, these network structures lack the support of physical concepts and result in large quantification errors regarding defect size, especially depth. In this article, to describe and integrate the above theory into a deep neural network, we propose a physics-informed doubly fed cross-residual network (DfedResNet) suitable for MFL defect detection based on deep learning. Physics-based MFL defect quantification theory is studied and integrated into loss functions during the neural network training. DfedResNet quantifies defects in MFL data and automatically extracts deep features of defects. The experimental results show that it effectively achieves high-precision quantification of defect length, width, and depth simultaneously, especially defect depth. Moreover, it considers data from all three dimensions during network training, and use the originally measured magnetic signal data in place of recognized images to avoid defect information loss and further improve the quantification accuracy. The deep DfedResNet model proposed in this article reduces defect length and width quantification errors to within 0.3 mm and defect depth quantification errors to within 0.4% <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">t</i> . In addition, compared with other network structures and traditional algorithms, DfedResNet improves defect quantification accuracy by 1–2 orders of magnitude and thus achieves a high quantification performance.

A Bayesian approach to blood rheological uncertainties in aortic hemodynamics.

Computational hemodynamics has received increasing attention recently. Patient-specific simulations require questionable model assumptions, for example, for geometry, boundary conditions, and material parameters. Consequently, the credibility of these simulations is much doubted, and rightly so. Yet, the matter may be addressed by a rigorous uncertainty quantification. In this contribution, we investigated the impact of blood rheological models on wall shear stress uncertainties in aortic hemodynamics obtained in numerical simulations. Based on shear-rheometric experiments, we compare the non-Newtonian Carreau model to a simple Newtonian model and a Reynolds number-equivalent Newtonian model. Bayesian Probability Theory treats uncertainties consistently and allows to include elusive assumptions such as the comparability of flow regimes. We overcome the prohibitively high computational cost for the simulation with a surrogate model, and account for the uncertainties of the surrogate model itself, too. We have two main findings: (1) The Newtonian models mostly underestimate the uncertainties as compared to the non-Newtonian model. (2) The wall shear stresses of specific persons cannot be distinguished due to largely overlapping uncertainty bands, implying that a more precise determination of person-specific blood rheological properties is necessary for person-specific simulations. While we refrain from a general recommendation for one rheological model, we have quantified the error of the uncertainty quantification associated with these modeling choices.

Abstract P3-09-16: Validating alternative splicing events between bulk and scRNA sequencing data: A bioinformatic approach

Abstract Background: As bulk and single-cell RNA (scRNA) sequencing studies and data continue to accrue to publicly accessible databases, bioinformatic tools continue being developed to analyze these datasets. The resulting pipelines often focus on either bulk or scRNA sequencing but rarely both. scRNA sequencing allows for in-depth characterization of transcription level events in distinct cell populations while bulk data offers a more global view, and for now, is more widely available in databases. Alternative splicing events in particular have been underexplored in a single-cell genomic architecture. The purpose of this study was to develop a bioinformatic pipeline to employ both bulk and single-cell mammary datasets to identify and validate alternative splicing events.Methods: Bulk fastq’s were utilized to identify AS events in luminal progenitor (LP) and mouse mammary stem cell (MSC) lineages in ovariectomized FVB mice that had been randomized into three treatment groups (SHAM [C], estradiol + progesterone [EP], EP + the selective progesterone receptor inhibitor telapristone acetate, TPA [EPT]). scRNA sequencing data from (Bach et al., 2017) was downloaded from the Gene Expression Omnibus (GEO) and utilized for validation of the AS events identified in the bulk data. To separate the data by progesterone level, the scRNA data was parsed into nulliparous, gestational, lactational, and post-involution groups. Further, to determine cell lineage level effects, the data was parsed into luminal and basal compartments using krt18 and krt5 expression. To accomplish the overall comparison between bulk and scRNA sequencing data, we developed a pipeline that would process and parse the scRNA data, identify the alternative splicing in bulk sequencing using rMATS and in the scRNA data using Outrigger (part of the Expedition suite). The resulting data from rMATS [bulk] LP/MSC: CvEP, CvEPT, &amp; EPvEPT were filtered for events meeting statistical significance (p-value &amp;lt; 0.05, FDR &amp;lt; 0.01). After filtering, the events from each treatment comparison were then filtered so only events unique to LP or MSC CvEP, respectively, remain. The remaining significant and unique AS events are then compared to the Krt18-high and Krt5-high Outrigger results from each development stage based on genomic event coordinates (with a buffer +/- 20 base pairs).Results: 12 alternative splicing (AS) events were identified to be shared between the bulk and scRNA sequencing data in the context of increased progesterone exposure (EP and gestational, respectively) as well as cell lineage (LP and luminal respectively). Among the genes identified as having been alternatively spiced with an exact nucleotide match for the splice were Eif4a2, Pik3c2a, Brd4, Cdh1, and Enah. Conclusions: Alternative splicing events can be identified and validated between genomic sequencing methods, specifically bulk and single-cell RNA seq. We hypothesize that as scRNA-seq becomes more developed and quantification errors caused by short-read lengths fixed by full length scRNA seq methods currently in development, comparing data between single-cell and bulk may be made easier. However, we have shown that, when orthogonal methods for validation may not be feasible, events as specific as alternative splices can be validated using publicly available data and in-silico methods. Citation Format: Gannon Cottone, Benjamin T Spike, Elnaz Mirzaei Mehrabad, Seema A Khan, Susan Clare. Validating alternative splicing events between bulk and scRNA sequencing data: A bioinformatic approach [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P3-09-16.

Development of a cardiovascular magnetic resonance-compatible large animal isolated heart model for direct comparison of beating and arrested hearts.

Cardiac motion results in image artefacts and quantification errors in many cardiovascular magnetic resonance (CMR) techniques, including microstructural assessment using diffusion tensor cardiovascular magnetic resonance (DT‐CMR). Here, we develop a CMR‐compatible isolated perfused porcine heart model that allows comparison of data obtained in beating and arrested states. Ten porcine hearts (8/10 for protocol optimisation) were harvested using a donor heart retrieval protocol and transported to the remote CMR facility. Langendorff perfusion in a 3D‐printed chamber and perfusion circuit re‐established contraction. Hearts were imaged using cine, parametric mapping and STEAM DT‐CMR at cardiac phases with the minimum and maximum wall thickness. High potassium and lithium perfusates were then used to arrest the heart in a slack and contracted state, respectively. Imaging was repeated in both arrested states. After imaging, tissue was removed for subsequent histology in a location matched to the DT‐CMR data using fiducial markers. Regular sustained contraction was successfully established in six out of 10 hearts, including the final five hearts. Imaging was performed in four hearts and one underwent the full protocol, including colocalised histology. The image quality was good and there was good agreement between DT‐CMR data in equivalent beating and arrested states. Despite the use of autologous blood and dextran within the perfusate, T2 mapping results, DT‐CMR measures and an increase in mass were consistent with development of myocardial oedema, resulting in failure to achieve a true diastolic‐like state. A contiguous stack of 313 5‐μm histological sections at and a 100‐μm thick section showing cell morphology on 3D fluorescent confocal microscopy colocalised to DT‐CMR data were obtained. A CMR‐compatible isolated perfused beating heart setup for large animal hearts allows direct comparisons of beating and arrested heart data with subsequent colocalised histology, without the need for onsite preclinical facilities.

Application of Seismic Inversion in Estimating Reservoir Petrophysical Properties: Case Study of Jay field of Niger Delta

The conventional approach in reservoir characterization uses an extrapolation of petrophysical properties calculated from well logs in determining the entire field volume. This technique fails to acknowledge/ factor in the heterogeneity of a reservoir, and consequently leads to errors in reservoir capacity quantification. This research work employs a deterministic seismic inversion approach in determining the petrophysical properties of each reservoir sand. First, the reservoir sand was identified from intervals of low gamma-ray value with corresponding high resistivity log value. A synthetic seismogram was generated to tie the well to the seismic data and identify events that correspond to the reservoir sands. This was done by convolving the generated reflectivity series, derived from approximated Zeoppritz reflectivity with a zero-phase wavelet generated using the seismic. Acoustic impedance from the well and seismic were analyzed and proved to have a strong correlation. The impedance was then constrained with determined petrophysical parameters from the well, and it was used in generating the petrophysics base model used in the inversion. Deterministic seismic inversion was carried out to be able to determine the petrophysical properties such as porosity, the volume of shale, and water saturation that are associated with the fluids from the seismic volume of Jay-field shallow offshore of Niger Delta.

Hepatic dual-contrast CT imaging: slow triple kVp switching CT with CNN-based sinogram completion and material decomposition.

Purpose: Dual-contrast protocols are a promising clinical multienergy computed tomography (CT) application for focal liver lesion detection and characterization. One avenue to enable multienergy CT is the introduction of photon-counting detectors (PCD). Although clinical translation is highly desired because of the diagnostic benefits of PCDs, it will still be a decade or more before they are broadly available. In our work, we investigated an alternative solution that can be implemented on widely used conventional CT systems (single source and integrating detector) to perform multimaterial spectral decomposition for dual-contrast imaging. Approach: We propose to slowly alternate the x-ray tube voltage between 3 kVp levels so each kVp level covers a few degrees of gantry rotation. This leads to the challenge of sparsely sampled projection data in each energy level. Performing the material decomposition (MD) in the sinogram domain is not directly possible as the projection images of the three energy levels are not angularly aligned. In order to overcome this challenge, we developed a convolutional neural network (CNN) framework for sparse sinogram completion (SC) and MD. To evaluate the feasibility of the slow kVp switching scheme, simulation studies of an abdominal phantom, which included liver lesions, were conducted. Results: The line-integral SC network yielded sinograms with a pixel-wise of the line-integrals compared to the ground truth. This provided acceptable image quality up to a switching angle of 9deg per kVp. The MD network we developed allowed us to differentiate iodine and gadolinium in the sinogram domain. The average relative quantification errors for iodine and gadolinium were below 10%. Conclusions: We developed a slow triple kVp switching data acquisition scheme and a CNN-based data processing pipeline. Results from a digital phantom validation illustrate the potential for future applications of dual-contrast agent protocols on practically available single-energy CT systems.

Proton magnetic resonance spectroscopy thermometry: Impact of separately acquired full water or partially suppressed water data on quantification and measurement error

In proton magnetic resonance spectroscopy (1 H MRS) thermometry, separately acquired full water and partially suppressed water are commonly used for measuring temperature. This paper compares these two approaches. Single-voxel 1 H MRS data were collected on a 3-T GE scanner from 26 human subjects. Every subject underwent five continuous MRS sessions, each separated by a 2-min phase. Each MRS session lasted 13 min and consisted of two free induction decays (FIDs) without water suppression (with full water [FW or w]) and 64 FIDs with partial water suppression (with partially suppressed water [PW or w']). Frequency differences between the two FWs, the first two PWs, the second FW and the first PW (FW2 , PW1 ), or between averaged water ( ) and N-acetylaspartate (NAA), were measured. Intrasubject and intersubject variations of the frequency differences were used as a metric for the error in temperature measurement. The intrasubject variations of frequency differences between FW2 and PW1 , calculated from the five MRS sessions for each subject, were larger than those between the two FWs or between the first two PWs (p = 1.54 x 10-4 and p = 1.72 x 10-4 , respectively). The mean values of intrasubject variations of for all subjects were 4.7 and 4.5 times those of and , respectively. The intrasubject variations of the temperatures based on frequency differences, or ( ), were about 2.5 times greater than those based on averaged water and NAA frequencies ). The mean temperature measured from ) (n = 26) was 0.29°C lower than that measured from and was 0.83°C higher than that from ( ). It was concluded that the use of separately acquired unsuppressed or partially suppressed water signals may result in large errors in frequency and, consequently, temperature measurement.

MR-based truncation correction using an advanced HUGE method to improve attenuation correction in PET/MR imaging of obese patients.

Truncation artifacts in the periphery of the magnetic resonance (MR) field-of-view (FOV) and thus, in the MR-based attenuation correction (AC) map, may hamper accurate positron emission tomography (PET) quantification in whole-body PET/MR, which is especially problematic in patients with obesity with overall large body dimensions. Therefore, an advanced truncation correction (TC) method to extend the conventional MR FOV is needed. The extent of MR-based AC-map truncations in obese patients was determined in a dataset including n=10 patients that underwent whole-body PET/MR exams. Patient inclusion criteria were defined as BMI>30kg/m2 and body weight>100kg. Truncations in PET/MR patients with obesity were quantified comparing the MR-based AC-map volume to segmented non-AC PET data, serving as the reference body volume without truncations to demonstrate the need of improved TC. The new method implemented in this study, termed "advanced HUGE," was modified and extended from the original HUGE method by Blumhagen etal. in order to provide improved TC across the entire axial MR FOV and to unlock new clinical applications of PET/MR. Advanced HUGE was then systematically tested in PET/MR NEMA phantom measurements. Relative differences between computed tomography (CT) AC PET data of the phantom setup (reference) and MR-based Dixon AC, respectively Dixon + advanced HUGE AC, were calculated. The applicability of the method for advanced TC was then demonstrated in first MR-based measurements in healthy volunteers. It was found that the MR-based AC maps of obese patients often reveal truncations in the anterior-posterior direction. Especially, the abdominal region could benefit from improved TC, where maximal relative differences in the AC-map volume up to -17% were calculated. Applying advanced HUGE to improve the MR-based AC in PET/MR, PET quantification errors in the large-volume phantom setup could be considerably reduced from average -18.6% (Dixon AC) to 4.6% compared to the CT AC reference. Volunteer measurements demonstrate that formerly missing AC-map volume in the Dixon-VIBE AC-map could be added due to advanced HUGE in the anterior-posterior direction and thus, potentially improves AC in PET/MR. The advanced HUGE method for truncation correction considerably reduces truncations in the anterior-posterior direction demonstrated in phantom measurements and healthy volunteers and thus, further improves MR-based AC in PET/MR imaging.

PtairMS: real-time processing and analysis of PTR-TOF-MS data for biomarker discovery in exhaled breath.

MotivationAnalysis of volatile organic compounds (VOCs) in exhaled breath by proton transfer reaction time-of-flight mass spectrometry (PTR-TOF-MS) is of increasing interest for real-time, non-invasive diagnosis, phenotyping and therapeutic drug monitoring in the clinics. However, there is currently a lack of methods and software tools for the processing of PTR-TOF-MS data from cohorts and suited for biomarker discovery studies.ResultsWe developed a comprehensive suite of algorithms that process raw data from patient acquisitions and generate the table of feature intensities. Notably, we included an innovative two-dimensional peak deconvolution model based on penalized splines signal regression for accurate estimation of the temporal profile and feature quantification, as well as a method to specifically select the VOCs from exhaled breath. The workflow was implemented as the ptairMS software, which contains a graphical interface to facilitate cohort management and data analysis. The approach was validated on both simulated and experimental datasets, and we showed that the sensitivity and specificity of the VOC detection reached 99% and 98.4%, respectively, and that the error of quantification was below 8.1% for concentrations down to 19 ppb.Availability and implementationThe ptairMS software is publicly available as an R package on Bioconductor (doi: 10.18129/B9.bioc.ptairMS), as well as its companion experiment package ptairData (doi: 10.18129/B9.bioc.ptairData).Supplementary information Supplementary data are available at Bioinformatics online.

Quantification correction for free-breathing myocardial T1ρ mapping in mice using a recursively derived description of a T1ρ* relaxation pathway

BackgroundFast and accurate T1ρ mapping in myocardium is still a major challenge, particularly in small animal models. The complex sequence design owing to electrocardiogram and respiratory gating leads to quantification errors in in vivo experiments, due to variations of the T1ρ relaxation pathway. In this study, we present an improved quantification method for T1ρ using a newly derived formalism of a T1ρ* relaxation pathway.MethodsThe new signal equation was derived by solving a recursion problem for spin-lock prepared fast gradient echo readouts. Based on Bloch simulations, we compared quantification errors using the common monoexponential model and our corrected model. The method was validated in phantom experiments and tested in vivo for myocardial T1ρ mapping in mice. Here, the impact of the breath dependent spin recovery time Trec on the quantification results was examined in detail.ResultsSimulations indicate that a correction is necessary, since systematically underestimated values are measured under in vivo conditions. In the phantom study, the mean quantification error could be reduced from − 7.4% to − 0.97%. In vivo, a correlation of uncorrected T1ρ with the respiratory cycle was observed. Using the newly derived correction method, this correlation was significantly reduced from r = 0.708 (p < 0.001) to r = 0.204 and the standard deviation of left ventricular T1ρ values in different animals was reduced by at least 39%.ConclusionThe suggested quantification formalism enables fast and precise myocardial T1ρ quantification for small animals during free breathing and can improve the comparability of study results. Our new technique offers a reasonable tool for assessing myocardial diseases, since pathologies that cause a change in heart or breathing rates do not lead to systematic misinterpretations. Besides, the derived signal equation can be used for sequence optimization or for subsequent correction of prior study results.

Minimizing errors in RT-PCR detection and quantification of SARS-CoV-2 RNA for wastewater surveillance

Wastewater surveillance for pathogens using reverse transcription-polymerase chain reaction (RT-PCR) is an effective and resource-efficient tool for gathering community-level public health information, including the incidence of coronavirus disease-19 (COVID-19). Surveillance of Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2) in wastewater can potentially provide an early warning signal of COVID-19 infections in a community. The capacity of the world's environmental microbiology and virology laboratories for SARS-CoV-2 RNA characterization in wastewater is increasing rapidly. However, there are no standardized protocols or harmonized quality assurance and quality control (QA/QC) procedures for SARS-CoV-2 wastewater surveillance. This paper is a technical review of factors that can cause false-positive and false-negative errors in the surveillance of SARS-CoV-2 RNA in wastewater, culminating in recommended strategies that can be implemented to identify and mitigate some of these errors. Recommendations include stringent QA/QC measures, representative sampling approaches, effective virus concentration and efficient RNA extraction, PCR inhibition assessment, inclusion of sample processing controls, and considerations for RT-PCR assay selection and data interpretation. Clear data interpretation guidelines (e.g., determination of positive and negative samples) are critical, particularly when the incidence of SARS-CoV-2 in wastewater is low. Corrective and confirmatory actions must be in place for inconclusive results or results diverging from current trends (e.g., initial onset or reemergence of COVID-19 in a community). It is also prudent to perform interlaboratory comparisons to ensure results' reliability and interpretability for prospective and retrospective analyses. The strategies that are recommended in this review aim to improve SARS-CoV-2 characterization and detection for wastewater surveillance applications. A silver lining of the COVID-19 pandemic is that the efficacy of wastewater surveillance continues to be demonstrated during this global crisis. In the future, wastewater should also play an important role in the surveillance of a range of other communicable diseases.

Vibration Control for Flexible Air-Breathing Hypersonic Vehicles with Input Quantification

This paper studies vibration control of flexible air-breathing hypersonic vehicles (FAHVs) with quantized input signal. Based on the Lyapunov's direct method, according to the stability of the quantized parameters, boundary feedback control is designed to ensure the ultimately uniformly bounded stable of the closed-loop system. Compared with existing results, we propose a quantizer, based on hysteresis-uniform quantizer and hysteresis-logarithmic quantizer. It can reduce the communication burden, reduce quantification errors, avoid system chattering, and ensure that the system has good performance. In addition, the hyperbolic tangent function is introduced into the quantification input process. Finally, the simulation results demonstrate the effectiveness of the scheme.

Stain-Free Approach for Western Blot Analysis of Zebrafish Embryos.

Western blots are employed to detect and characterize amounts of proteins in biological samples. Quantifications are traditionally carried out through data normalization by housekeeping protein method. This approach does not account for variations not intrinsically dependent on the sample such as different experimental conditions and type of samples. Zebrafish researchers often face the challenge of comparing embryos at different developmental stages or from different strains. Housekeeping protein amount can change in these conditions therefore adding an unwanted quantification error. Here we describe the method to analyze mutant zebrafish embryos at different stages by western blot using the Stain-Free technology for normalization. We present Globin X quantification at 2 and 5days postfertilization in wild type and in the bloodless Vlad Tepes (vlt) zebrafish mutant that lack red blood cells.

Ranking-Based Deep Hashing Network for Image Retrieval

In large-scale image retrieval, the deep learning-based hashing methods have significantly progressed. However, most of the existing deep hashing methods still have the problems of low feature learning efficiency and weak ranking relationship discrimination. To remedy these problems, a novel Ranking-based Deep Hashing Network (RDHN) is proposed for image retrieval in this paper, which integrates the feature learning module and hash learning module into a unified deep hashing network framework to jointly learn a powerful hash function so that the raw images can be mapped to discrete hash codes with significant discrimination. Specifically, a novel difference convolution is designed based on edge detection operators, and then it is uniquely applied to the first convolutional layer of the convolutional neural network (CNN), which can take advantage of the sensitive characteristics of edge detection operators for edge information to extract richer image edge information. Meanwhile, in hash learning, the ranking metric Mean Average Precision (MAP) is optimized using the idea of scaling, and then a ranking loss function based on MAP is carefully designed to enhance the neighborhood ranking capabilities of the hash codes. Furthermore, to reduce the quantification error, a quantization loss function is also designed. Finally, the ranking loss function is combined with the quantization loss function to form the objective function. The proposed method can generate high-quality discrete hash codes while learning to preserve ranking information, effectively improving retrieval performance. Extensive experimental results on three widely used benchmark datasets demonstrate that the proposed method outperforms the state-of-the-art hashing approaches.

Detection Efficiency Modeling and Joint Activity and Attenuation Reconstruction in Non-TOF 3-D PET From Multiple-Energy Window Data

Emission-based attenuation correction (AC) meth-ods offer the possibility of overcoming quantification errors induced by conventional MR-based approaches in PET/MR imaging. However, the joint problem of determining AC and the activity of interest is strongly ill-posed in non-TOF PET. This can be improved by exploiting the extra information arising from low energy window photons, but the feasibility of this approach has only been studied with relatively simplistic analytic simulations so far. This manuscript aims to address some of the remaining challenges needed to handle realistic measurements; in particular, the detection efficiency (“normalisation”) estimation for each energy window is investigated. An energy-dependent detection efficiency model is proposed, accounting for the presence of unscattered events in the lower energy window due to detector scatter. Geometric calibration factors are estimated prior to the reconstruction for both scattered and unscattered events. Different reconstruction methods are also compared. Results show that geometric factors differ markedly between the energy windows and that our analytical model correspond in good approximation to Monte Carlo simulation; the multiple energy window reconstruction appears sensitive to input/model mismatch. Our method applies to Monte Carlo generated data but can be extended to measured data. This study is restricted to single scatter events.

DIRECT-Net: A unified mutual-domain material decomposition network for quantitative dual-energy CT imaging.

The purpose of this paper is to present an end-to-end deep convolutional neural network to improve the dual-energy CT (DECT) material decompositionperformance. In this study, we proposes a unified mutual-domain (sinogram domain and CT domain) material decomposition network (DIRECT-Net) for DECT imaging. By design, the DIRECT-Net has immediate access to mutual-domain data, and utilizes stacked convolution neural network layers for noise reduction and material decomposition. The training data are numerically generated following the fundamental DECT imaging physics. Numerical simulation of the XCAT digital phantom, experiments of a biological specimen, a calcium chloride phantom and an iodine solution phantom are carried out to evaluate the performance of DIRECT-Net. Comparisons are performed with different DECT decompositionalgorithms. Results demonstrate that the proposed DIRECT-Net can generate water and bone basis images with less artifacts compared to the other decomposition methods. Additionally, the quantification errors of the calcium chloride (75-375 mg/cm3 ) and the iodine (2-20 mg/cm3 ) are less than 4%. An end-to-end material decomposition network is proposed for quantitative DECT imaging. The qualitative and quantitative results demonstrate that this new DIRECT-Net has promising benefits in improving the DECT imagequality.