Signal Distribution Research Articles

Intelligent fault diagnosis of ultrasonic motors based on graph-regularized CNN-BiLSTM

The ultrasonic motor (USM) is peculiarly prone to failure due to continuous high-frequency friction-related power transfer, whose failure mechanisms are remarkably different from traditional induction motors. Intelligent fault diagnosis provides a way to alarm and avoid catastrophic losses proactively. However, previous studies using deep learning usually ignore the inherent geometric structure of the signal distribution. This paper proposes an intelligent multi-signal fault diagnosis framework for USMs to restore the linear or nonlinear manifold structure by preserving the internal structure by integrating graph regularization with deep neural networks. Firstly, the one-dimensional CNN to learn spatial correlations and BiLSTM to exploit temporal dependencies are coalesced to build the deep neural network. Then, an improved k-nearest neighbor graph is proposed to protect the geometric structure information and force the latent features to be more concentrated within their classes. Moreover, the layer in the deep architecture to integrate graph regularization is designed to reduce computation cost, and an adaptive decay strategy is considered to adjust the coefficient of graph regularized automatically. A two-stage training algorithm is developed by considering the time to calculate the graph regularization term. Finally, the proposed multi-signal fault diagnosis framework is validated using datasets from the fault injection experiment of similar USMs in China’s Yutu rover of Chang’e lunar probe. Experimental results show that the proposed method can effectively discriminate different fault types.

A lightweight multi-feature fusion vision transformer bearing fault diagnosis method with strong local sensing ability in complex environments

Fault diagnosis of rolling bearings in complex environments is a difficult problem. First, the median filter can remove the noise in the vibration signals, however, it cannot adaptively adjust the filter weights according to the input signals. Second, the popular vision transformer (ViT) cannot extract local feature information under complex conditions and has a large number of parameters, which result in increased computational complexity. To solve these problems, a lightweight multi-feature fusion ViT bearing fault diagnosis method with strong local awareness in complex environments is proposed. Firstly, to learn the features and statistical distributions of the input signals, the gradient descent method is used to continuously and iteratively update the weights and filter the signals. Then, to better extract critical local fault information, a local sensing module is constructed using multi-scale wide convolutional neural network. Finally, an improved lightweight multi-feature fusion ViT is constructed to perform global feature extraction and fault identification. The results show that the proposed method has better noise reduction effect and feature extraction ability, and can accurately identify the fault types under the complex environments.

Binary discrimination through next-to-leading order

Binary discrimination between well-defined signal and background datasets is a problem of fundamental importance in particle physics. With detailed event simulation and the advent of extensive deep learning tools, identification of the likelihood ratio has typically been reserved as a computational problem. However, this approach can obscure overtraining or excessive sensitivity to tuned features of the simulation that may not be well-defined theoretically. Here, we present the first analysis of binary discrimination for signal and background distributions for which their likelihood ratio is infrared and collinear safe, and can therefore be calculated order-by-order in perturbation theory. We present explicit, general formulas for receiver operator characteristic curves and the area under it through next-to-leading order. These results can then establish absolute upper bounds on discrimination performance because any realistic implementation will have measurement errors, undetected particles, or restrictions on fiducial phase space. As a demonstration of this formalism, we apply it to discrimination of highly-boosted Higgs decays from gluon splitting to bottom quarks. Effects at next-to-leading order are first sensitive to the flow of color in the jet and significantly modify discrimination performance at leading-order. In the limit of infinite boost, these events can be perfectly discriminated because only the gluon will radiate at finite angles from the bottom quarks, and we find that large effects persist at energies accessible at the Large Hadron Collider. Next-to-leading order is therefore required to qualitatively understand results using machine-learning methods.

Studying GPR's direct and reflected waves

AbstractAs the transmitter and receiver (Tx and Rx, respectively) are located in close proximity during a typical ground‐penetrating radar (GPR) survey, the powerful signal generated by the Tx and which is then recorded by the Rx at various time delays, can be saturated at early times (i.e., this is the direct wave (DW) signal reaching the Rx). This often causes the masking of shallow targets, complicating data interpretation. In this study, our aim is to examine the spatial distribution of the electromagnetic signals around the Tx, attempting to locate areas where the DW becomes minimum, whereas the signal strength from subsurface targets (i.e., reflected wave – RW) remains ideally unchanged. The position of these local minima in the DW signal could give rise to advantageous Tx–Rx configurations, where clear reflections from subsurface targets lying at shallow depths can be obtained with the least possible involvement of the DW. To perform such a study, we carried out static field measurements over a flat lying reflector as well as numerical simulations in a reflection, common‐offset mode around a transmitting antenna. In the field, we also collected wide‐angle reflection–refraction data to determine the GPR wave velocity in the uppermost layer. GPR signals were recorded by the Rx around the Tx in three concentric circles of various radii (i.e., varying the Tx/Rx separation), using a specific angular step and varying the Tx/Rx polarization each time. The synthetic data were produced using a three‐dimensional finite‐difference time‐domain modelling tool. Field and numerically simulated data were analysed and compared to study the behaviour of both the DW and RW events around the Tx when changing the Tx/Rx distance, their respective angular position, as well as their relative polarization/orientation.

Fault Diagnosis for Wind Turbine Flange Bolts Based on One-Dimensional Depthwise Separable Convolutions

In this paper, a new bolt fault diagnosis method is developed to solve the fault diagnosis problem of wind turbine flange bolts using one-dimensional depthwise separable convolutions. The main idea is to use a one-dimensional convolutional neural network model to classify and identify the acoustic vibration signals of bolts, which represent different bolt damage states. Through the method of knock test and modal simulation, it is concluded that the damage state of wind turbine flange bolt is related to the natural frequency distribution of acoustic vibration signal. It is found that the bolt damage state affects the modal shape of the structure, and then affects the natural frequency distribution of the bolt vibration signal. Therefore, the damage state can be identified by identifying the natural frequency distribution of the bolt acoustic vibration signal. In the present one-dimensional depth-detachable convolutional neural network model, the one-dimensional vector is first convolved into multiple channels, and then each channel is separately learned by depth-detachable convolution, which can effectively improve the feature quality and the effect of data classification. From the perspective of the realization mechanism of convolution operation, the depthwise separable convolution operation has fewer parameters and faster computing speed, making it easier to build lightweight models and deploy them to mobile devices.

Phasor Measurement Unit Change-Point Detection of Frequency Hurst Exponent Anomaly With Time-to-Event

The objective of this paper is real-time detection of a change-point in the baseline distribution of the frequency signal generated by Phasor Measurement Units (PMUs) that could indicate potential for voltage collapse, false data injection, or other security threats. The alarm flags an anomaly event when the cumulative sum (CUSUM) of the log-likelihood departures of the Hurst exponent of the frequency from its baseline statistic exceeds a threshold set up as a compromise between the conflicting objectives of minimum detection delay and acceptable False Alarm Rate. As main theoretical contribution, an extra protection layer is developed that provides an estimate of the time to the alarm event, giving the Transmission System Operator (TSO) or an autonomous agent more time to deploy proactive measures to avoid large catastrophic system states. The proposed method is illustrated by a retrospective analysis of the 2012 Indian blackout that reveals that 10-12 minutes before the voltage collapse, a significant increase in the Hurst exponent of the frequency could have been detected. Conceptually, this shows that PMUs provide significant value towards ensuring autonomous cognition in the power grid by enabling the abnormal events to be fault-detected in an unsupervised fashion.

Can single-use versus standard duodenoscope improve ergonomics in ERCP? A comparative, simulation-based pilot study.

Background and study aims Musculoskeletal disorders (MSDs) and injuries (MSIs) are frequent in gastrointestinal endoscopy. The aim of this study was to assess potential ergonomic advantages of a lighter single-use duodenoscope compared with a standard reusable one for endoscopists performing endoscopic retrograde cholangiopancreatography (ERCP). Methods Three experienced endoscopists performed an ergonomic, preclinical, comparative protocol-guided simulation study of a single-use and a standard reusable duodenoscope using an anatomic bench model. Surface EMG signals from left forearm and arm muscles were recorded. A commercial inertial sensor-based motion capture system was applied to record body posture as well. Results A significant lowering of root mean square amplitude and amplitude distribution of biceps brachii signal (ranging from 13% to 42%) was recorded in all the participants when using a single-use duodenoscope compared with a reusable one. An overall reduction of muscle activation amplitude and duration was also associated with the single-use duodenoscope for forearm muscles, with different behaviors among subjects. Participants spent most of the time in wrist extension (> 80%) and ulnar deviation (> 65%). A consistent pattern of functional range of motion employed for completing all procedures was observed. Conclusions Our study showed that a lighter scope has a promising effect in reducing upper arm muscle activity during ERCP with potential benefit on musculoskeletal health in the ERCP setting.

Expotecnología 2022

Expotecnología 2022 was an event that took place in the city of Cartagena de Indias, from November 2 to 6, 2022, with the slogan “The future of engineering in the face of technological disruption” This eighteenth version was a meeting that brought together the scientific and academic community in spaces for the dissemination and transfer of knowledge, where innovation was the essence and a fundamental part of the event. Expotecnología was created in the Faculty of Engineering Sciences - FACI, of the Antonio de Arévalo University Foundation - UNITECNAR and in this version it maintained as mission, the development of the interaction between science, technology and society, through establishing of knowledge networks and academic and professional cooperation. The main objective was to address the challenges that emerge from the everyday life of the engineering community, the problems immersed in the development of projects and programs of Higher Education Institutions (IES), organizations and research centers at the national and international level. In the development of the day of presentations in a virtual way, it was possible to promote the dissemination of the research work of professionals, teachers and students from all areas of engineering, framed in four (4) thematic axes, as follows:• New technologies: Big data, Artificial Intelligence, Business Intelligence, Software Engineering, Education and new technologies and Mobile Applications.• Petrochemicals and process control: Materials Science, Instrumentation and Control Systems, Robotics and Automation and Industrial Maintenance.• Electricity, electronics and telecommunications: IoT, Fiber optics, Digital Signal Processing, Wireless Networks, Mobile Communications, Power Systems, Energy Storage, Smart Grids, Renewable Energies, Operation and Distribution, Energy Dispatch, Energy Markets• Construction Summit: Modern Building Systems and Computational Modeling.On this occasion, the event was attended by 512 people, representatives from 25 national and 6 international universities. 32 presentations were made, of which 32 articles were selected for this volume of the IOP Conference Cycle: Materials Science and Engineering and 13 articles for the book Engineering, Innovation and the fourth industrial revolution., After a process of Peer Review. The plenary conferences were in charge of 2 national and 8 international speakers .List of Invited speakers, Organizing Committee, Special thanks are available in this pdf.

Cumulative Distribution Function and Spatially Resolved Surface-Enhanced Raman Spectroscopy for the Quantitative Analysis of Emtricitabine

Surface-enhanced Raman spectroscopy (SERS) has exceptional analytical sensitivity and selectivity. However, SERS irreproducibility presents an obstacle when using it for precise quantitative measurements. In this study, colloidal nanoparticles evaporated to dryness are used as a SERS active surface for the detection of the HIV drug emtricitabine (FTC; trade name Emtriva). Despite the irreproducibility of the SERS resulting from the stochastic process of evaporation, using a SERS scanning instrument, the SERS enhancement factors of spatially resolved spectra have a well-defined distribution of signals for a given analyte concentration. This distribution follows a power law function ranging from weak (very abundant signals) to very strong (sparse signals). By definition, a power law distribution cannot have a true mean. Hence a cumulative distribution function was used to model the concentration of emtricitabine, and calibration curves were constructed. For stochastically generated quantitative data sets, the precision of this approach is superior to methods utilizing signal averaging and improves analytical sensitivity.

Research on Magnetic Memory Detection of Weld Fatigue Damage Based on Two-Dimensional Spectral Entropy

Abstract In light of the intrinsic relationship between weld fatigue damage and the characteristics of magnetic leakage signals, fatigue experiments were conducted on butt welds to analyze the evolution of magnetic leakage signal as fatigue cycles increased. The introduction of two-dimensional spectral entropy allowed for the construction of a fatigue stage division model based on the two-dimensional spectral entropy diagram. The results revealed that distinctions between magnetic leakage signal under loaded and unloaded conditions were solely evident in their magnitude, with consistent curve distribution trends observed across various fatigue cycles. Furthermore, the distribution pattern of magnetic leakage signal displayed significant fluctuations, characterized by a sharp gradient increase following the emergence of fatigue cracks. These distribution characteristics of magnetic leakage signal differed markedly from the magnetic field distribution at other fatigue stages, with a higher growth rate observed as fatigue cycles increased. The two-dimensional spectral entropy spectrum analysis of magnetic leakage signals at different fatigue stages indicated the disorder in the magnetic leakage signal distribution within the high fatigue zone, signifying the expansion of fatigue microcracks. To some extent, two-dimensional spectral entropy can serve as an early warning system for detecting fatigue cracks in welds and achieving early identification of fatigue microcracks.

Probing Hyperbolic Shear Polaritons in β-Ga2O3 Nanostructures Using STEM-EELS.

Phonon polaritons, quasiparticles arising from strong coupling between electromagnetic waves and optical phonons, have potential for applications in subdiffraction imaging, sensing, thermal conduction enhancement, and spectroscopy signal enhancement. A new class of phonon polaritons in low-symmetry monoclinic crystals, hyperbolic shear polaritons (HShPs), have been verified recently in β-Ga2O3 by free electron laser (FEL) measurements. However, detailed behaviors of HShPs in β-Ga2O3 nanostructures still remain unknown. Here, by using monochromatic electron energy loss spectroscopy in conjunction with scanning transmission electron microscopy, the experimental observation of multiple HShPs in β-Ga2O3 in the mid-infrared (MIR) and far-infrared (FIR) ranges is reported. HShPs in various β-Ga2O3 nanorods and a β-Ga2O3 nanodisk are excited. The frequency-dependent rotation and shear effect of HShPs reflect on the distribution of EELS signals. The propagation and reflection of HShPs in nanostructures are clarified by simulations of electric field distribution. These findings suggest that, with its tunable broad spectral HShPs, β-Ga2O3 is an excellent candidate for nanophotonic applications.

Effect of fluidized bed shapes on gas-solid fluidization characters and flow regime transition

The technology of gas-solid dense-phase fluidized bed separation is applied in the separation of fine particles in minerals due to its stable and reliable characteristics. The experiment collected time-domain distribution signals of pressure drop from fluidized beds with different shapes, systematically analyzed the spatiotemporal distribution characteristics of bed density, and studied the effects of bed shape and fluidization number N on the fluidization and flow transition of the bed layer. The results indicate significant differences in Umf among fluidized beds with different shapes, and the tapered bed requires the highest energy for fluidization. When N=1.3, the cylindrical bed layer is the most stable, maintaining an average density of 1.8 g/cm3, with a density fluctuation variance of 0.02 g/cm3. Additionally, when N<1.0, almost no bubbles inside the bed layer. With an increase in gas velocity, the time-domain distribution of pressure drop signals shows different degrees of fluctuation. When N=1.1, Conspicuous non-uniform particle flow is evident in the tapered bed, with a bubble main frequency of 8 Hz, an amplitude of 27.1, and the highest pressure drop energy signal, indicating that the tapered bed enters the bubbling state prematurely. Finally, fine coal separation experiments were conducted, and the results showed that the minimum E value for the cylindrical bed is 0.08 g/cm3, while the maximum E value for the tapered bed is 0.17 g/cm3.

The 21-cm bispectrum from neutral hydrogen islands at z &amp;lt; 6

ABSTRACT Spatial variations in the Lyman-α forest opacity at z &amp;lt; 6 seem to require a late end to cosmic reionization. In this picture, the Universe contains neutral hydrogen ‘islands’ of up to 100 cMpc/h in extent down to redshifts as low as z ∼ 5.3. This delayed end to reionization also seems to be corroborated by various other observables. An implication of this scenario is that the power spectrum of the cosmological 21-cm signal at z &amp;lt; 6 is enhanced relative to conventional reionization models by orders of magnitude. However, these neutral hydrogen islands are also predicted to be at the locations of the deepest voids in the cosmological large-scale structure. As a result, the distribution of the 21-cm signal from them is highly non-Gaussian. We derive the 21-cm bispectrum signal from these regions using high-dynamic-range radiative transfer simulations of reionization. We find that relative to conventional models in which reionization is complete at z &amp;gt; 6, our model has a significantly larger value of the 21-cm bispectrum. The neutral islands also imprint a feature in the isosceles bispectrum at a characteristic scale of ∼1 cMpc−1. We also study the 21-cm bispectrum for general triangle configuration by defining a triangle index. It should be possible to detect the 21-cm bispectrum signal at ν ≳ 200 MHz using SKA1-LOW for 1080 h of observation, assuming optimistic foreground removal.

A model-independent measurement of the CKM angle γ in partially reconstructed B± → D*h± decays with D → KS0\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$ {K}_S^0 $$\\end{document}h+h− (h = π, K)

A measurement of CP-violating observables in B± → D*K± and B± → D*π± decays is made where the photon or neutral pion from the D*→ Dγ or D*→ Dπ0 decay is not reconstructed. The D meson is reconstructed in the self-conjugate decay modes, D → KS0\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$ {K}_S^0 $$\\end{document}π+π− or D → KS0\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$ {K}_S^0 $$\\end{document}K+K−. The distribution of signal yields in the D decay phase space is analysed in a model-independent way. The measurement uses a data sample collected in proton-proton collisions at centre-of-mass energies of 7, 8, and 13 TeV, corresponding to a total integrated luminosity of approximately 9 fb−1. The B± → D*K± and B± → D*π±CP-violating observables are interpreted in terms of hadronic parameters and the CKM angle γ, resulting in a measurement of γ = (92−17+21\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$ {92}_{-17}^{+21} $$\\end{document})°. The total uncertainty includes the statistical and systematic uncertainties, and the uncertainty due to external strong-phase inputs.

Effect of natural fractures on mechanical properties and fracture patterns of shale at microscopic scale: an example from the Lower Cambrian Niutitang formation in Qianbei region

In order to explore the influence of natural fractures on the mechanical properties and failure modes of shale at the micro scale, uniaxial compression numerical experiments were conducted on the shale of the Niutang Formation in northern Guizhou with different natural fracture angles using a rock failure process system and digital image processing technology. It is shown that the compressive strength of shale increases with the increase of natural crack inclination, and the growth rate of shale compressive strength also increases. Shale's microscopic fractures can generally be classified into four categories. The first category is to sprout along the natural cracks to the outside of the shale, and eventually form a crack similar to the "X" type (0°); the second category is to sprout along the natural cracks to the middle and outside of the shale, and eventually form an inverted "Y" type crack (15°, 30°); the third category is to sprout along the natural cracks to the middle and outside of the shale, and eventually form an inverted "Y" type crack (15°, 30°); the second type sprouts along the natural fractures toward the middle and outside of the shale, forming inverted "Y"-type fractures (15°, 30°); the third type cracks along the sides of the natural fractures, forming "Y"-type fractures (45°); and the fourth type does not crack along the natural fractures, forming "S"-type fractures (60°, 75°, and 90°). In the low natural fracture dip shale model, tensile damage mainly occurs, accompanied by a small amount of compressive shear damage; in the high natural fracture dip shale model, tensile damage and compressive shear damage account for a larger proportion in the fracture process.This suggests that the presence of natural cracks in shale has a significant impact on stress distribution. There are two main types of acoustic emission signal distribution and evolutionary features, the evolutionary features of acoustic emission signal distribution are of two types, 0°-45° test and 60°-90° test, and the difference is mainly reflected in the damage stage, the damage of shale with high natural fracture inclination is more intense, which is manifested by the decrease in the number of acoustic emission events, but the level of acoustic emission events in the damage stage is higher, which can reach 61788, 46605 and 94315, the shale with high natural fracture inclination is more brittle.

Trends in Obstetrics and Gynecology Residency Applications in the Year After Abortion Access Changes

State-specific abortion restrictions currently affect the training of approximately 44% of obstetrics and gynecology (OBGYN) residents in the US. Examination of where future trainees apply for residency is important. To assess changes in the percentage of applicants to OBGYN residency programs by state based on abortion restrictions in place after the Dobbs v Jackson Women's Health Organization (hereafter, Dobbs v Jackson) US Supreme Court decision and examine whether applicants' preference for programs, as suggested by the distribution of application signals that express higher interest, was associated with abortion bans. This serial cross-sectional study used anonymized data for all applicants to OBGYN residency programs in the US during September and October from 2019 to 2023. Data were obtained from the Association of American Medical Colleges Electronic Residency Application Service. Applications and program preference signals sent to OBGYN residency programs, analyzed by applicants' self-reported demographics. The primary outcome was differences in the percentage of unique applicants to OBGYN residency programs from 2019 to 2023, with programs categorized by state-based abortion restrictions after the Dobbs v Jackson decision. Secondary outcomes included the distribution of program signals by state abortion ban status. A total of 2463 applicants (2104 [85.4%] women) who applied to OBGYN programs for the 2023 residency match cycle were the focal sample of this study. While overall applicant numbers remained stable between 2019 and 2023, the number of applicants differed significantly by state abortion ban status in the 2022 (F2,1087 = 10.82; P < .001) and the 2023 (F2,1087 = 14.31; P < .001) match cycles. There were no differences in the number of signals received by programs in states with bans after controlling for known covariates such as number of applications received and program size, and there were no differences in the percentage of signals sent by out-of-state applicants to programs in states with different abortion laws than their home states (F2,268 = 2.41; P = .09). In this cross-sectional study, there was a small but statistically significant decrease in the number of applicants to OBGYN residency programs in states with abortion bans in 2023 compared with 2022. However, applicant signaling data did not vary by states' abortion ban status. While OBGYN residency programs almost completely filled in 2023, continued monitoring for the potential consequences of state abortion bans for OBGYN training is needed.

Time varying discrete cosine transform based on shaping regularization and its application in seismic data analysis

Abstract The discrete cosine transform is a commonly used technique in the field of signal processing that employs cosine basis functions for signal analysis. Traditionally, the regression coefficients of the cosine basis functions are solely based on frequency information. This paper extends the regression coefficients associated with the cosine basis functions to take into account both frequency and time information, not just frequency information alone. This modification results in an ill-posed linear system, which requires regularization to prevent overfitting. To address this, the paper uses shaping regularization, a technique used to stabilize ill-posed problems. By doing so, the absolute values of these extended coefficients, now exhibiting variations in both frequency and time domains, are defined as the time-frequency distribution of that input signal. The numerical experiments conducted to validate this approach demonstrate that the proposed method yields a commendable time-frequency resolution. Consequently, it proves valuable for interpreting seismic data, showcasing its potential for applications in this field.

Abstract B032: Comprehensive analysis of clonality, segregation errors, chromosomal instability and heterogeneity in cancer cell lines and cancer samples reveals that cancer evolution is often promoted by highly aberrant cellular divisions

Abstract Chromosomal instability (CIN) is the most common form of genomic instability in cancer. CIN leads to structural and numerical chromosomal aberrations and to heterogeneity of such aberrations inside the same tumor. Together with other forms of genomic and epigenomic instability, CIN underlies the most dangerous feature of cancer cells, namely, ability to evolve, allowing cancer cells to evade immunologic surveillance, to metastasize, to become resistant to drug treatments. Studies of chromosomal instability are significantly impeded by the lack of comprehensive approaches for simultaneous detection of different forms of ongoing chromosomal instability and their outcomes. We developed an approach that allows to analyze the clonal composition and chromosomal heterogeneity in cell lines and cancer samples, as well as rates of different forms of ongoing chromosomal instability in cancer cell lines. At the heart our analysis is a multiplex interphase FISH (miFISH) with seven sequential hybridizations of color probe panels resulting in the assessment of a total of 35 probes per single cell. Each probe panel is designed to include the centromeric probe and probes on the p- and q-arm for a given chromosome that allows for the simultaneous assessment of centromeric probes and gene-specific arm probes. This approach was applied to interphase nuclei, metaphase spreads and bi-nucleated/dividing cancer cells in colorectal and cervical cell lines and cancer samples. miFISH allows for automated acquisition of nuclei and metaphase spreads on a BioView spotcounting system. A gallery provides detailed views for semi-automatic as well as manual analysis of the obtained images. A total of 10000-25000 cell images were collected from each slide. The simultaneous evaluation of our strategically chosen centromeric and gene-specific probes within the same cells allows unprecedented insights in the patterns of distribution of genetic material among cancer cells in cancer cell population, and, especially, in dividing cells. We evaluated clonality, clonal and non-clonal chromosomal heterogeneity, ploidy distributions, and occurrence of endoreplication. Assessment of dividing cells based on equal or unequal distribution of miFISH centromeric and gene-specific signals allowed to estimate and compare the rates and outcomes of different forms of ongoing chromosomal instability such as structural rearrangements and amplifications, loss of chromosomes or chromosomal fragments in micronuclei, non-disjunctions, grossly asymmetrical cell divisions (including multipolar mitoses). Conclusions: Gradual changes in chromosomal/genomic content of cancer cells occur due to structural rearrangements, losses of chromosomes and chromosomal fragments in micronuclei, and by relatively rare single chromosome non-disjunctions. Grossly abnormal cell divisions (massive non-disjunctions and structural rearrangements) is the main source for the high level of chromosomal abnormalities and aneuploidy in evolved cancer cell populations. Citation Format: Anna Roschke. Comprehensive analysis of clonality, segregation errors, chromosomal instability and heterogeneity in cancer cell lines and cancer samples reveals that cancer evolution is often promoted by highly aberrant cellular divisions [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Translating Cancer Evolution and Data Science: The Next Frontier; 2023 Dec 3-6; Boston, Massachusetts. Philadelphia (PA): AACR; Cancer Res 2024;84(3 Suppl_2):Abstract nr B032.

Distilling Invariant Representations With Domain Adversarial Learning for Cross-Subject Children Seizure Prediction

Seizure prediction based on electroencephalogram (EEG) has great potential to improve patients’ life quality. Due to the high heterogeneity in distributions of EEG signals among different patients, conventional studies usually show poor generalization ability when transferring the model to new patients, which also leads to difficulties in clinical applications. To alleviate the challenging issue concerning cross-subject domain shift, we propose a transformer-based domain adversarial model. Our model first adopts a pre-trained general neural network to extract common features from the EEG signals of available patients. Then, we design a distiller module and a domain discriminator module to perform domain adaptation training based on a small amount of labeled data from the new-coming patient. During adaptation process, conditional domain adversarial training with the addition of label information is employed to remove patient-related information from the extracted features to learn a common seizure feature space among different patients. Our proposed seizure prediction method is evaluated on the CHB-MIT EEG database. The proposed model achieves a sensitivity of 79.5%, a false alarm rate (FPR) of 0.258/h, and an AUC of 0.814. Experimental results demonstrate that the proposed method can effectively reduce inter-patient domain disparity compared to state-of-the-art methods.

TSoft-Net: A novel transfer soft thresholding network based on self-attention for intelligent fault diagnosis of rotating machinery

In the industrial scenes, the machinery operates under diverse working conditions and generates varying levels of noise, which can hinder the performance of the intelligent fault diagnosis model that is trained in the laboratory. This is because the different working conditions and environmental noise change the distribution of laboratory vibration signal. To tackle this issue, we proposed a novel transfer soft thresholding network (TSoft-Net) based on self-attention for intelligent fault diagnosis of rotating machinery, which has good anti-noise performance and can be transferred to different working conditions with excellent accuracy. This paper constructs a soft residual block for extracting the fault representation and enhancing the residual learning. In this block, we propose a residual factor to learn and enhance domain-invariant fault representation. Furthermore, a representation soft fusion block is built for extracting and fusing the different scale fault representation. In this block, we propose a scale-attention weight to dynamically fuse the different scale fault representation. The experiments show that the TSoft-Net, compared with six existing methods on eight sub-datasets, has better anti-noise ability and achieves an accuracy of up to 100% on the target domain.