Separate Tests Research Articles

Increasing the capabilities of a test ballistic missile to separate test objects

Increasing the capabilities of a test ballistic missile to separate test objects

AI-BASED AUTOMATED VARUS/VALGUS ALIGNMENT MEASUREMENT IN STANDARD KNEE RADIOGRAPHS

IntroductionAccurate assessment of alignment in pre-operative and post-operative knee radiographs is important for planning and evaluating knee replacement surgery. Existing methods predominantly rely on manual measurements using long-leg radiographs, which are time-consuming to perform and are prone to reliability errors. In this study, we propose a machine-learning-based approach to automatically measure anatomical varus/valgus alignment in pre-operative and post-operative standard AP knee radiographs.MethodWe collected a training dataset of 816 pre-operative and 457 one-year post-operative AP knee radiographs of patients who underwent knee replacement surgery. Further, we have collected a separate distinct test dataset with both pre-operative and one-year post-operative radiographs for 376 patients. We manually outlined the distal femur and the proximal tibia/fibula with points to capture the knee joint (including implants in the post-operative images). This included point positions used to permit calculation of the anatomical tibiofemoral angle. We defined varus/valgus as negative/positive deviations from zero. Ground truth measurements were obtained from the manually placed points. We used the training dataset to develop a machine-learning-based automatic system to locate the point positions and derive the automatic measurements. Agreement between the automatic and manual measurements for the test dataset was assessed by intra-class correlation coefficient (ICC), mean absolute difference (MAD) and Bland-Altman analysis.ResultAnalysing the agreement between the manual and automated measurements, ICC values were excellent pre-/post-operatively (0.96, CI: 0.94-0.96) / (0.95, CI: 0.95-0.96). Pre-/post-operative MAD values were 1.3°±1.4°SD / 0.7°±0.6°SD. The Bland-Altman analysis showed a pre-/post-operative mean difference (bias) of 0.3°±1.9°SD/-0.02°±0.9°SD, with pre-/post-operative 95% limits of agreement of ±3.7°/±1.8°, respectively.ConclusionThe developed machine-learning-based system demonstrates high accuracy and reliability in automatically measuring anatomical varus/valgus alignment in pre-operative and post-operative knee radiographs. It provides a promising approach for automating the measurement of anatomical alignment without the need for long-leg radiographs.AcknowledgementsThis research was funded by the Wellcome Trust [223267/Z/21/Z].

Abstract A074: Early detection of breast implant associated anaplastic large cell lymphoma by multiplex lateral flow assay of CD30 and IL-10

Abstract Introduction: More than 1300 women have developed a unique breast implant associated anaplastic large cell lymphoma (BIA-ALCL) around breast implants. BIA-ALCL presents as an accumulation of fluid (seroma) a median of 9 years after implant placement. When detected early, survival with surgery alone is nearly 100%. Remaining patients present with masses and/or lymph node metastases requiring radiation, chemotherapy and/or immunotherapy and fatalities have occurred. Methods: Preliminary studies identified CD30 and IL-10 as the most sensitive and specific biomarkers for BIA-ALCL in seroma fluids. Therefore, we constructed a multiplex LFA to detect both CD30 and IL-10 in 30 mL seroma fluid within 20 minutes. LFA is a point of care immunoassay using capture and detection antibodies that detect separate epitopes of the target molecule. The detection antibody is linked to a colored particle; in this multiplex assay CD30 antibody was linked to blue and IL-10 to red particles. Separate positive test lines (blue and red) indicate presence of CD30 or IL-10. A control line confirms migration of the seroma fluid past the test lines. To determine the dynamic range of sensitivity we spiked recombinant CD30 and IL-10 into benign seromas. Test and control lines were quantitated with NIH Image J software and ratios of test line to control line calculate for each sample. Forty-eight hour supernatants of supernatants of BIA-ALCL lines TLBR1 and TLBR2 were tested for IL-10 and CD30. Fifteen patients, 8 with BIA-ALCL and 7 with benign seromas provided consent for testing of their seroma fluids. Results: Multiplex CD30/IL10 LFA produced positive test lines for neat and 1:10 dilutions of 48 hour supernatnats of BIA-ALCL lines TLBR1 and TLBR2. Both IL-10 and CD-30 could be detected on the multiplex LFA strip using a 1:3 dilution of seroma fluids. The multiplex LFA shows both IL-10 and CD30 are significantly higher in malignant samples (IL-10 TL/CL: malignant, 0.2±0.16 n=8 vs benign, 0.004±0.007 n=7, p<0.05; CD30 TL/CL: malignant, 0.38±0.2 vs benign, 0.02±0.02, p<0.001, two-way ANOVA). Importantly, when either IL-10 or CD30 yields a faint positive test line, the other is not detected reducing the possibility of false positives. Conclusions: The CD30/IL-10 multiplex LFA had 100% positive and negative predictive value in this pilot study. The assay was able to distinguish lymphoma from recurrent breast cancer and more common benign seromas. The use of this multiplex assay mitigates positive false positive results from more sensitive CD30 LFA which has been detected in some benign seromas. This LFA will potentially allow early pre-operative detection of ALCL allowing surgeons to plan their surgical approach. The LFA can replace more costly time-consuming flow cytometry, immunohistochemistry, cyto- and histopathologic studies which require at least 10 ml of fluid. The LFA can be performed outside of medical centers in rural and underserved areas. A larger prospective multicenter study including intra-operative sampling is planned to validate these results. Citation Format: Peng Xu, Katerina Kourentzi, Richard Willson, Marshall Kadin. Early detection of breast implant associated anaplastic large cell lymphoma by multiplex lateral flow assay of CD30 and IL-10 [abstract]. In: Proceedings of the AACR Special Conference: Liquid Biopsy: From Discovery to Clinical Implementation; 2024 Nov 13-16; San Diego, CA. Philadelphia (PA): AACR; Clin Cancer Res 2024;30(21_Suppl):Abstract nr A074.

M-CLUSTER: multistage clustering for unsupervised train wheel condition monitoring

Wheel out-of-roundness (OOR) is a common wheel defect that raises maintenance costs and increases the risk of failure or damage to track components. This paper proposes a novel multistage clustering framework (M-CLUSTER) for unsupervised condition monitoring of train wheels. The framework initially extracts time-domain features from raw acceleration responses collected by rail-mounted sensors. Sensitive features are then selected through an unsupervised feature selection algorithm called local learning-based clustering (LLC). Next, a detector model is trained using density-based spatial clustering of applications with noise (DBSCAN), a data clustering method effective for clusters with similar density. Since this algorithm does not originally involve separate training and testing phases, a new two-step mechanism is introduced: (1) training on a healthy dataset and (2) testing on an unlabelled dataset. Finally, the severity of train wheel defects is classified by K-means, with cluster validity indices (CVI) automatically determining the number of severity clusters (classes). The framework’s efficiency is demonstrated through the detection of defective wheels using the Alfa Pendular passenger model. Results indicate that M-CLUSTER accurately identifies train wheel flats and polygonal wear without labelled data, achieving 98% accuracy by selecting 10 features from the set.

Abstract 4131622: Opportunistic Screening of Chronic Liver Disease With Deep Learning Enhanced Echocardiography

Introduction: Chronic liver disease affects more than 1.5 billion adults worldwide, but the majority of cases are asymptomatic and undiagnosed. Echocardiography is broadly performed and visualizes the liver; however, this information is not diagnostically leveraged. Hypothesis and Aims: We hypothesized that a deep-learning algorithm can detect chronic liver diseases using subcostal echocardiography images that contains hepatic tissue. To develop and evaluate a deep learning algorithm on subcostal echocardiography videos to enable opportunistic screening for chronic liver disease. Methods: We identified adult patients who received echocardiography and abdominal imaging (either abdominal ultrasound or abdominal magnetic resonance imaging) with ≤30 days between tests. A convolutional neural network pipeline was developed to predict the presence of cirrhosis or steatotic liver disease (SLD) using echocardiogram images. The model performance was evaluated in a held-out test dataset, dataset in which diagnosis was made by magnetic resonance imaging, and external dataset. Results: A total of 2,083,932 echocardiography videos (51,608 studies) from Cedars-Sinai Medical Center (CSMC) were used to develop EchoNet-Liver, an automated pipeline that identifies high quality subcostal images from echocardiogram studies and detects presence of cirrhosis or SLD. In a total of 11,419 quality-controlled subcostal videos from 4,849 patients, a chronic liver disease detection model was able to detect the presence of cirrhosis with an AUC of 0.837 (0.789 - 0.880) and SLD with an AUC of 0.799 (0.758 - 0.837). In a separate test cohort with paired abdominal MRIs, cirrhosis was detected with an AUC of 0.726 (0.659-0.790) compared to MR elastography and SLD was detected with an AUC of 0.704 (0.689-0.718). In the external test cohort of 66 patients (n = 130 videos), the model detected cirrhosis with an AUC of 0.830 (0.738 - 0.909) and SLD with an AUC of 0.768 (0.652 – 0.875). Conclusions: Deep learning assessment of clinically indicated echocardiography enables opportunistic screening of SLD and cirrhosis. Application of this algorithm may identify patients who may benefit from further diagnostic testing and treatment for hepatic disease.

Can computer vision / artificial intelligence locate key reference points and make clinically relevant measurements on axillary radiographs?

Computer vision and artificial intelligence (AI) offer the opportunity to rapidly and accurately interpret standardized x-rays. We trained and validated a machine learning tool that identified key reference points and determined glenoid retroversion and glenohumeral relationships on axillary radiographs. Standardized pre and post arthroplasty axillary radiographs were manually annotated locating six reference points and used to train a computer vision model that could identify these reference points without human guidance. The model then used these reference points to determine humeroglenoid alignment in the anterior to posterior direction and glenoid version. The model's accuracy was tested on a separate test set of axillary images not used in training, comparing its reference point locations, alignment and version to the corresponding values assessed by two surgeons. On the test set of pre- and post-operative images not used in the training process, the model was able to rapidly identify all six reference point locations to within a mean of 2mm of the surgeon-assessed points. The mean variation in alignment and version measurements between the surgeon assessors and the model was similar to the variation between the two surgeon assessors. This article reports on the development and validation of a computer vision/artificial intelligence model that can independently identify key landmarks and determine the glenohumeral relationship and glenoid version on axillary radiographs. This observer-independent approach has the potential to enable efficient human observer independent assessment of shoulder radiographs, lessening the burden of manual x-ray interpretation and enabling scaling of these measurements across large numbers of patients from multiple centers so that pre and postoperative anatomy can be correlated with patient reported clinical outcomes. Level III Study of Diagnostic Test.

Digitalization, Management, and Synthesis of Thermal-Hydraulic Legacy Data Using the Dynamical System Scaling

A unique attribute of the nuclear industry is the extent by which designers, developers, operators, and regulators pay attention to demonstrating the safety case. Nuclear installation resiliency or safety takes overriding priority over function and performance. The preparation of the safety case heavily relies on extensive experimental campaigns that challenge the target design under a spectrum of postulated accident scenarios. For safety reasons, these experiments, typically conducted in subscale test apparatuses intended to reproduce the postulated accident scenarios and event sequences realistically, but subject to proven similarity criteria, are often referred to as scaling analyses. Since the dawn of the nuclear industry, significant resources have been committed to the construction and operation of such test facilities, and a massive amount of data has been generated. Such test data have been and are still used to validate the fundamental assumptions of nuclear power plant phenomena. As we move into the digital world, capturing this multiple decades worth of information in a transparent, easily accessible, and usable manner for the public and industry stakeholders is of paramount importance. Over the last few years, FPoliSolutions has been actively developing an enterprise digital data ontology platform (OGMA) intended to do just that. The goal is to help designers use experimental data to assess their safety case evaluation models. The platform or application programming interface (API) was designed to ontologically organize the experimental data, as well guide the user in the complex analytics associated with the interpretation and use of such test results. The OGMA API supports the user in accessing a library of sophisticated mathematical procedures for performing scaling analyses. For this purpose, the dynamical system scaling (DSS) procedure invented by Dr. Jose’ Reyes was formulated as one of the services in the digital platform. These methods have been demonstrated to provide an excelled mathematical apparatus to identify similarity criteria or quantify distortions between measured data and the target prototypical conditions. Moreover, DSS can provide a level of synthesis across separate effect tests and integral effect tests that has the promise of yielding efficiency in the evaluation code assessment activities, as setting up evaluation models that support the safety case of current and future nuclear power plants is often a daunting and expensive task.

Resolution-dependent MRI-to-CT translation for orthotopic breast cancer models using deep learning.


This study aims to investigate the feasibility of utilizing generative adversarial networks (GANs) to synthesize high-fidelity CT images from lower-resolution MR images. The goal is to reduce patient exposure to ionizing radiation while maintaining treatment accuracy and accelerating MR image acquisition. The primary focus is to determine the extent to which low-resolution MR images can be utilized to generate high-quality CT images through a systematic study of spatial resolution-dependent MRI-to-CT image conversion. 
Approach. 
Paired MRI-CT images were acquired from healthy control and tumor models, generated by injecting MDA-MB-231 and 4T1 tumor cells into the mammary fat pad of nude and BALB/c mice to ensure model diversification. To explore various MRI resolutions, we downscaled the highest-resolution MR image into three lower resolutions. Using a customized U-Net model, we automated region of interest masking for both MRI and CT modalities with precise alignment, achieved through three-dimensional affine paired MRI-CT registrations. Then our customized models, Nested U-Net Generative Adversarial Network (NUGAN) and Attention U-Net Generative Adversarial Network (AUGAN), were employed to translate low-resolution MR images into high-resolution CT images, followed by evaluation with separate testing datasets.
Main Results.
Our approach successfully generated high-quality CT images (0.142 mm²) from both lower-resolution (0.282 mm²) and higher-resolution (0.142 mm²) MR images, with no statistically significant differences between them, effectively doubling the speed of MR image acquisition. Our customized GANs successfully preserved anatomical details, addressing the typical loss issue seen in other MRI-CT translation techniques across all resolutions of MR image inputs.
 Significance. 
This study demonstrates the potential of using low-resolution MR images to generate high-quality CT images, thereby reducing radiation exposure and expediting MRI acquisition while maintaining accuracy for radiotherapy.&#xD.

Cardiac motion correction with a deep learning network for perfusion defect assessment in SPECT myocardial perfusion imaging

BackgroundIn myocardial perfusion imaging (MPI) with SPECT ungated studies are used for evaluation of perfusion defects despite motion blur. We investigate the potential benefit of motion correction using a deep-learning (DL) network for evaluating perfusion defects. MethodsWe employed a DL network for cardiac motion correction in ECG-gated SPECT-MPI images, wherein the image data from different cardiac phases are combined with respect to a reference gate to reduce motion blur. For training the DL network 197 cases were used. Given the variability of gated images during the cardiac cycle, we investigated the detectability of perfusion defects in two distinct reference gates. To assess perfusion defect detection, we performed receiver-operating-characteristic (ROC) analyses on the motion-corrected images using a separate test dataset of clinical 194 subjects, in which studies were created from actual patient data with inserted simulated-lesions as ground truth. The reconstructed images were assessed by the quantitative-perfusion SPECT (QPS) software. We also evaluated the performance on reduced-count studies (by two and four folds). ResultsThe quantitative results, measured by area-under-the-ROC curve (AUC), demonstrated that DL motion correction improves the detectability of perfusion defects significantly on both standard- and reduced-count studies, and that the detectability can vary with reference cardiac phases. A joint assessment from two reference-phases achieved AUC=0.841 on the quarter-count data, higher than with ungated full-count data (AUC=0.795, p-value=0.0054). ConclusionsDL motion correction can benefit assessment of perfusion defects in standard- and reduced-count SPECT-MPI studies. It can also be beneficial to evaluate perfusion images over multiple cardiac phases.

3-D full-field reconstruction of chemically reacting flow towards high-dimension conditions through machine learning

Monitoring chemically reacting flow is crucial for optimizing and controlling the conversion processes in various chemical engineering scenarios. These processes are often influenced by multiple factors, creating a high-dimensional condition space that imposes intensive computational cost to computational fluid dynamics (CFD) methods. We present a machine learning (ML) framework to reconstruct the 3-D scalar field preserving the accuracy and resolution of CFD modeling. We demonstrate the framework’s effectiveness through a case study on biomass combustion in an industrial-scale grate furnace, which exemplifies the challenges of modeling chemically reacting flow in complex geometric domains influenced non-linearly by multiple parameters. The framework implements three core principles: (1) a Reactor-Structure-Resembled Network (RSRNet) reflecting the operational logics of the reactor, (2) an incremental learning strategy based on random Latin Hypercube Sampling, and (3) the incorporation of self-extracted high-level features from data as soft constraints in the loss function. Ground truth data were calculated through a CFD model, with all cases sampled from a 11-variable condition space. Using only 240 randomly sampled cases (around 1 % total cases in the discretized condition space), RSRNet precisely replicated 1-D and 2-D scalar distributions through incremental training. Further reconstruction of 3-D temperature field only used 40 3-D CFD cases, resulting in the training and testing errors decreased to 7.94 × 10-4 and 1.55 × 10-3, respectively, approaching the level of the 2-D reconstruction, with the average error in temperature field predictions not exceeding 50 K. The generalization ability was validated using a separate test dataset with continuously changing excess air ratio and capacity ratio in wide ranges, including some extreme values, and the model successfully captured complex phenomena under unseen conditions.

AUDIT‐C shows reliability when used online in patient portals

The Alcohol Use Disorders Identification Test–Consumption (AUDIT‐C) is a validated screen, but until recently has not been evaluated in routine conditions for adult primary care patients. A new study has found that AUDIT‐C screens have “excellent” test‐retest reliability, which assesses the consistency of results when given at two different times. Excellent test‐retest reliability means that there are similar results across separate test administrations, and the study, supported by the National Institute on Alcohol Abuse and Alcoholism (NIAAA), shows that the AUDIT‐C can be used reliably in adult primary care patients. The study, led by Claire B. Simon, M.D., of the University of Washington in Seattle, used electronic health record (EHR) data from Kaiser Permanente. For the study, more than 18,000 adult primary care patients completed two AUDIT‐C screens at 1 to 21 days apart as part of routine care in 2021. Screen results were reliable whether completed in person or online through patient portals. However, reliability was higher when the patient completed both screens using online patient portals, compared to completing both screens in the clinic or one in the clinic and one online. Therefore, completing two screens may be a better option than just doing one. The study, High test–retest reliability of the Alcohol Use Disorders Identification Test–Consumption (AUDIT‐C) questionnaire completed by primary care patients in routine care, is published in the February 2024 issue of Alcohol, Clinical and Experimental Research. The researchers also found that reliability was lower for American Indian/Alaska Native (AI/AN) patients and multiracial patients.

Hsa_circ_0001492 regulates the hsa-miR-145-5p/ovarian carcinoma immunoreactive antigen domain 2 axis to promote the progression of lung adenocarcinoma.

Circular RNA (circRNA) has been proven to be a key regulator in a range of tumor illnesses, such as lung adenocarcinoma (LUAD); however, the regulatory mechanisms of circRNA remain unclear. In this study, circRNA (hsa_circ_0001492) in LUAD was examined for its regulatory and functional potential. qRT-PCR was used to assess the hsa_circ_0001492 level in LUAD. The RNAse R digestion test was employed to isolate hsa_circ_0001492. The primary location of hsa_circ_0001492 enrichment in LUAD cells was identified through a nucleoplasmic separation test. LUAD cell migration, proliferation, and spherogenicity were examined using wound healing, transwell, EdU, and cell spherogenicity assays. The association between miR-145-5p and hsa_circ_0001492/ovarian carcinoma immunoreactive antigen domain 2 (OCIAD2) was validated using a dual luciferase experiment. The interaction between sh-hsa_circ_0001492 and miR-145-5p was confirmed through an RNA pull-down assay. The effects of hsa_circ_0001492, miR-145-5p, and OCIAD2 on LUAD tumor development were examined using xenograft mouse models and immunohistochemistry tests. Results showed a higher amount of hsa_circ_0001492 in LUAD. The cytoplasm of LUAD cells was observed in the area where hsa_circ_0001492 mainly accumulated; hsa_circ_0001492 enhanced LUAD cell migration, proliferation, and sphere-forming ability. MiR-145-5p and OCIAD2 were identified as targets of hsa_circ_0001492 and miR-145-5p, respectively. The level of OCIAD2 was increased by hsa_circ_0001492 through targeted binding to miR-145-5p. In nude mice, tumor growth was inhibited by silencing hsa_circ_0001492, while knockdown of miR-145-5p and overexpression of OCIAD2 promoted the growth of LUAD tumors. In conclusion, hsa_circ_0001492 regulates the hsa-miR-145-5p/OCIAD2 axis to promote the progression of LUAD, and could be a useful target for the diagnosis and treatment of LUAD.

Individualistic reward-seeking strategies that predict response to nicotine emerge among isogenic male mice living in a micro-society.

Individual animals differ in their traits and preferences, which shape their social interactions, survival, and susceptibility to disease, including addiction. Nicotine use is highly heterogenous and has been linked to the expression of personality traits. Although these relationships are well documented, we have limited understanding of the neurophysiological mechanisms that give rise to distinct behavioral profiles and their connection to nicotine susceptibility. To address this question, we conducted a study using a semi-natural and social environment called "Souris-City" to observe the long-term behavior of individual male mice. Souris-City provided both a communal living area and a separate test area where mice engaged in a reward-seeking task isolated from their peers. Mice developed individualistic reward-seeking strategies when choosing between water and sucrose in the test compartment, which, in turn, predicted how they adapted to the introduction of nicotine as a reinforcer. Moreover, the profiles mice developed while isolated in the test area correlated with their behavior within the social environment, linking decision-making strategies to the expression of behavioral traits. Neurophysiological markers of adaptability within the dopamine system were apparent upon nicotine challenge and were associated with specific profiles. Our findings suggest that environmental adaptations influence behavioral traits and sensitivity to nicotine by acting on dopaminergic reactivity in the face of nicotine exposure, potentially contributing to addiction susceptibility. These results further emphasize the importance of understanding interindividual variability in behavior to gain insight into the mechanisms of decision-making and addiction.

A new approach to investigate secondary hydriding phenomenon on M5Framatome clads under high–temperature LOCA conditions

The focus of this study is about a new experimental approach for a separate effects study of the secondary hydriding phenomenon under LOCA conditions. Many nuclear institutes perform semi–integrals tests to study the cladding behaviour during a LOCA transient. Those tests combined several phenomena and performing a detailed analysis of the secondary hydriding phenomenon using these tests can be challenging. A dedicated experimental protocol aiming at simulating secondary hydriding has been set up. Separate effects tests (SETs) were then carried out using this protocol to study the effects of both oxidation duration and temperature, on the hydrogen absorption during the oxidation stage of the LOCA transient on M5Framatome11M5 and M5Framatome are trademarks or registered trademarks of Framatome or its affiliates, in the USA or other countries. cladding. The effects of gap size were also investigated. Metallographic analysis has been used to characterise the M5Framatome clad metallurgical transformation after the high–temperature (HT) oxidation.

A beautiful face is good when we're judged by others, a moral character is better.

Moral beauty, reflected in one's actions, and facial beauty both affect how we're judged. Here, we investigated how moral and facial beauty interactively affect social judgments and emotional responses, employing event-related brain potentials. Participants (all female) associated positive, neutral, or negative verbal information with faces scoring high or low on attractiveness and performed ratings of the faces as manipulation checks. In a separate test phase, the faces were presented again, and participants made valenced social judgments of the persons. Results show a dominance of moral beauty in valenced social judgments as well as ERPs related to reflexive and evaluative emotional responses (early posterior negativity, EPN, late positive potential, LPP), whereas facial attractiveness mattered little. In contrast, facial attractiveness affected visual processing (N170). Similarly, relatively shallow impressions of attractiveness and likability that require no knowledge about the person were influenced by both facial attractiveness and social-emotional information. This pattern of dominant effects of social-emotional information regardless of attractiveness shows that when it comes to our emotional responses and social judgments, moral beauty is what matters most, even in the face of physical beauty.

Copy Number Analysis of Whole Exome Sequencing Data

Abstract Background Genetic alterations implicated in constitutional disorders range in size from single nucleotide substitutions to losses and gains of millions of bases. Whole exome sequencing (WES) captures many clinically relevant variants, but an analytic pipeline optimized for detection of small substitutions, insertions, and deletions may miss pathogenic copy number variants (CNVs). Historically, our laboratory performed separate cytogenetic tests to evaluate for larger genomic imbalances. Addition of copy number analysis to our WES pipeline could improve the assay’s diagnostic yield. This study describes part of our strategy to validate copy number analysis of our constitutional whole exome sequencing data. Methods We identified 18 CNV specimens that underwent both chromosomal microarray analysis (CMA) and WES between 2017 and 2023. Variants included 13 deletions (size range 14Kb-2Mb) and 5 duplications (size range 325Kb-10Mb). The standard method of constitutional copy number analysis in our clinical laboratory is CMA using the Affymetrix Cytoscan HD Array and analysis software. CNVs identified on CMA were reviewed with reference to the Database of Genomic Variants and multiple academic hospital databases. WES was performed on an Illumina sequencing system following capture-based library enrichment with the Integrated DNA Technologies xGen Exome Research, CNV Backbone, Human mtDNA Research, and Human ID Research Panel probes. CNVs identified on WES were called using the Broad Institute’s Genomic Analysis Toolkit (GATK) with a quality score cutoff of 120. Whole exome sequencing data was visualized on the Integrative Genomics Viewer. Results 14 out of 18 CNVs reported on CMA were also detected on WES with a quality score at or above 120 (78%). This includes 10 out of 13 deletions (77%) and 3 out of 5 duplications (60%). For both deletions and duplications, the size ranges of detected variants at or above the cutoff score and undetected variants overlapped. One duplication reported on CMA was detected on WES with a quality score below 120. Visualization of whole exome sequencing data showed variation in probe density that may have affected CNV calling. Conclusions The majority of CNVs reported after CMA (14/18, 78%) were also detected on WES data using a GATK quality score cutoff of 120. The data does not suggest an association between CNV size and detection by WES, although interpretation is limited by the small number of cases. Data visualization may be required to identify some CNVs. Future steps to characterize the limitations of copy number analysis of WES data include assessments of variance in probe density and read depth across the exome. Copy number analysis of WES data from cases with normal CMA results will also be performed for assay validation.

Machine learning to classify left ventricular hypertrophy using ECG feature extraction by variational autoencoder.

Traditional ECG criteria for left ventricular hypertrophy (LVH) have low diagnostic yield. Machine learning (ML) can improve ECG classification. ECG summary features (rate, intervals, axis), R-wave, S-wave and overall-QRS amplitudes, and QRS/QRST voltage-time integrals (VTIs) were extracted from 12-lead, vectorcardiographic X-Y-Z-lead, and root-mean-square (3D) representative-beat ECGs. Latent features were extracted by variational autoencoder from X-Y-Z and 3D representative-beat ECGs. Logistic regression, random forest, light gradient boosted machine (LGBM), residual network (ResNet) and multilayer perceptron network (MLP) models using ECG features and sex, and a convolutional neural network (CNN) using ECG signals, were trained to predict LVH (left ventricular mass indexed in women >95 g/m², men >115 g/m²) on 225,333 adult ECG-echocardiogram (within 45 days) pairs. AUROCs for LVH classification were obtained in a separate test set for individual ECG variables, traditional criteria and ML models. In the test set (n=25,263), AUROC for LVH classification was higher for ML models using ECG features (LGBM 0.790, MLP 0.789, ResNet 0.788) as compared to the best individual variable (VTI QRS-3D 0.677), the best traditional criterion (Cornell voltage-duration product 0.647) and CNN using ECG signal (0.767). Among patients without LVH who had a follow-up echocardiogram >1 (closest to 5) years later, LGBM false positives, compared to true negatives, had a 2.63 (95% CI 2.01, 3.45)-fold higher risk for developing LVH (p<0.0001). ML models are superior to traditional ECG criteria to classify-and predict future-LVH. Models trained on extracted ECG features, including variational autoencoder latent variables, outperformed CNN directly trained on ECG signal.

Investigation of the potential of in-line particle concentration measurements in gas particle separation processes by using low-cost particulate matter sensors

In addition to the costly laboratory-based particle measurement devices, low-cost particulate matter sensors for ambient air quality measurements can also be employed for the detection of PM in gas cleaning devices. The ability to spatially resolve the actual level of particle number concentration with high temporal resolution within a gas-particle separation process is of significant value for process monitoring and control. The quantitative reliability of the measurement data from low-cost particulate matter sensors remains uncertain when the particle concentrations in the apparatus exceed the upper measurement limit of the sensors, whether temporarily or permanently. This study examines the potential of low-cost particulate matter sensors for in-line particle concentration measurements in gas-particle separation processes. The measurement performance of the low-cost particulate matter sensors is initially examined under idealised conditions in a dust chamber. This is done in order to quantify the deviations of the low-cost particulate matter sensors in comparison to a state-of-the-art precision laboratory high-cost sensor. Subsequently, both sensor types are installed at a wet separator test facility, following the quantification of their respective measurement performances. The low-cost particulate matter sensors are capable of measuring particle number concentration above their specified limit (up to ≈6.1×104cm−3), depending on the measurement conditions. By applying a 3rd polynomial correction function, obtained from the initial quantification measurement, the low-cost particulate matter sensors are able to reproduce the particle number concentration measured by the high-cost sensor. Furthermore, the low-cost particulate matter sensors are capable of reproducing short-term fluctuations in the particle number concentration.

A deep learning approach to predicting UAV sound affect

The most prevalent method by which the impact of an acoustic environment upon people is assessed is the `environmental noise' approach, in which the sounds are quantified based entirely on their recorded weighted sound pressure levels. This does not account, however, for the character of sounds outside of their loudness. Objective metrics such as psychoacoustic annoyance have been developed to attempt to rectify this, however this is still heavily weighted towards loudness as the most important factor and may not provide a good estimate for subjective perceived annoyance for all sound sources, including unconventional aircraft such as unmanned aerial vehicles (UAVs). Here we present a novel method using a convolutional neural network (CNN) to predict perceived annoyance directly from mel-spectrograms derived from recorded audio featuring both traditional and unconventional aircraft sources. The network is trained and tested using data collated from five separate listening tests, and is shown to outperform conventional regression models based on psychoacoustic annoyance as a predictor of perceived annoyance. The time-frequency resolution of the calculated mel-spectrograms is shown to be a key factor in performance.

Information accumulation on the item versus source test of source monitoring: Insights from diffusion modeling.

Source monitoring involves attributing previous experiences (e.g., studied words as items) to their origins (e.g., screen positions as sources). The present study aimed toward a better understanding of temporal aspects of item and source processing. Participants made source decisions for recognized items either in succession (i.e., the standard format) or in separate test blocks providing independent measures of item and source decision speed. Comparable speeds of item and source decision across the test formats would suggest a full separation between item and source processing, whereas different speeds would imply their (partial) temporal overlap. To test these alternatives, we used the drift rate parameter of the diffusion model (Ratcliff, Psychological Review, 85, 59-108, 1978). We examined whether the drift rates, together with the other parameters, assessed separately for the item and source decision varied as a function of the test format. Threshold separation and nondecision time differed between the test formats, but item and source decision speeds represented by drift rates did not change significantly. Thus, despite facilitation on the source decision when the item decision was immediately followed by a test for source memory than when item and source were tested in separate blocks, findings did not suggest that source information already begins accumulating in the item test in the standard format. We discuss the temporal sequence of item and source processing in light of different assumptions about the contribution of familiarity and recollection.