Approach For Identification Research Articles

Partial causal identification for right censored data with noncompliance

Medical researchers are typically interested in determining the effect of an intervention on the survival time that can be right censored, and the censoring time is usually treated as independent of the survival time. However, the independent censoring assumption is untestable and might bias the inference if taken on faith. Furthermore, in observational studies or randomised experiments with noncompliance when unconfoundedness does not necessarily hold, causal effects might only be partially identified, e.g. using instrumental variables. In this paper, we develop instrumental partial identification approaches for causal effect estimation for right censored observational data, where we aim to estimate the effect of an intervention on survival probabilities under an arbitrary dependence structure between survival and censoring time. We illustrate the proposed interval estimators in a variety of settings. We also demonstrate our method by a data application to evaluate the effect of having more than one living child on elderly parental survival probabilities.

Embryonic reprogramming of the tumor vasculature reveals targets for cancer therapy

A sustained blood supply is critical for tumor growth, as it delivers the nutrients and oxygen required for development. Targeting of blood vessel formation via immunotherapies is an area of great importance. Knowing that certain embryonic genes, such as carcinoembryonic antigens (CEA) and oncofetal fibronectin, become reexpressed in malignant transformation, we hypothesized that a similar phenomenon holds true for tumor endothelial cells (TECs) as well. An approach for identification of highly selective tumor endothelial markers was conducted to develop targeted antiangiogenic immunotherapies. We first queried the transcriptome that is present during embryo development. We then performed a systematic search for genes selectively expressed in the mouse embryo at days E11 and E18, as compared to the transcriptome of the adult mouse. Subsequently, we queried for expression of these embryonic genes in sorted murine TECs. This approach identified among others the tumor endothelial antigens fibrillin-2 (Fbn2), elastin microfibril interface-located protein 2 (Emilin2) as well as the tumor endothelial antigens lysyl oxidase (Lox) and serine/cysteine protease inhibitor, clade E, member 1 (Serpine1; Pai-1). For these selected genes, functional involvement in angiogenesis was confirmed in in vitro bioassays. We subsequently used iBoost conjugate vaccine technology to develop vaccines against the selected targets. For all four targets, vaccination readily induced target-specific antibody responses in mice, resulting in inhibition of tumor growth. Access to highly specific tumor endothelial markers provides opportunities for direct targeting of the tumor vasculature with high specificity, without affecting healthy vasculature.

A vision‐based weigh‐in‐motion approach for vehicle load tracking and identification

AbstractWith the rapid increase in the number of vehicles, accurately identifying vehicle loads is crucial for maintaining and operating transportation infrastructure systems. Existing load identification methods typically rely on collecting vehicle load data from weigh‐in‐motion (WIM) systems when vehicles pass over them. However, cumbersome installation, high costs, and regular maintenance are the main obstacles that prevent WIM from being widely used in practice. This study introduces the visual WIM (V‐WIM) framework, a vision‐based approach for tracking and identifying moving loads. The V‐WIM framework consists of two main components, the vehicle weight estimation and the vehicle tracking and location estimation. Vehicle weight is estimated using tire deformation parameters extracted from tire images through object detection and optical character recognition techniques. A deep learning‐based YOLOv8 algorithm is employed as a vehicle detector, combined with the ByteTrack algorithm for tracking vehicle location. The vehicle weight and its corresponding location are then integrated to enable simultaneous vehicle weight estimation and tracking. The performance of the proposed framework was evaluated through two component validation tests and one on‐site validation test, demonstrating its capability to overcome the limitations of existing methods.

Gene print-based cell subtypes annotation of human disease across heterogeneous datasets with gPRINT.

Identification of disease-specific cell subtypes (DSCSs) has profound implications for understanding disease mechanisms, preoperative diagnosis, and precision therapy. However, achieving unified annotation of DSCSs in heterogeneous single-cell datasets remains a challenge. In this study, we developed the gPRINT algorithm (generalized approach for cell subtype Identification with single cell's voicePRINT). Inspired by the principles of speech recognition in noisy environments, gPRINT transforms gene position and gene expression information into voiceprints based on ordered and clustered gene expression phenomena, obtaining unique "gene print" patterns for each cell. Then, we integrated neural networks to mitigate the impact of background noise on cell identity label mapping. We demonstrated the reproducibility of gPRINT across different donors, single-cell sequencing platforms, and disease subtypes, and its utility for automatic cell subtype annotation across datasets. Moreover, gPRINT achieved higher annotation accuracy of 98.37% when externally validated based on the same tissue, surpassing other algorithms. Furthermore, this approach has been applied to fibrosis-associated diseases in multiple tissues throughout the body, as well as to the annotation of fibroblast subtypes in a single tissue, tendon, where fibrosis is prevalent. We successfully achieved automatic prediction of tendinopathy-specific cell subtypes, key targets, and related drugs. In summary, gPRINT provides an automated and unified approach for identifying DSCSs across datasets, facilitating the elucidation of specific cell subtypes under different disease states and providing a powerful tool for exploring therapeutic targets in diseases.

Techniques of Ensemble Learning for Human Emotional Classification and Detection

The identification of emotions from textual data has attracted attention due to developments in natural language processing (NLP), which provides insightful information about human behaviour in a variety of contexts. This study evaluates the performance of several Using an emotion dataset that is available on Hugging Face Hub, machine learning methods, such as Support Vector Machine (SVM), Logistic Regression, Random Forest, and Multinomial Naive Bayes, are used to classify emotions. The models we tested had different accuracy percentages: multinomial naive Bayes produced 65.90%, logistic regression performed well at 82.40%, random forest produced 85.12%, and support vector machine (SVM) produced 81.56%. Moreover, utilizing ensemble learning methodologies to capitalize on the advantages of many models has improved overall efficiency. Our ensemble learning approach demonstrates the effectiveness of mixing multiple models for improved emotion recognition, achieving an amazing 98% accuracy. Ensemble learning is a promising approach for emotion identification in textual data because it leverages the combined knowledge of separate models, reducing flaws and increasing robustness. Keywords--- Machine learning, Support Vector Machine, Random Forest, logarithmic regression, Multinomial Naive Bayes, Ensemble Voting Classifier, and Emotion Identification.

MCDA Approach for Identification of Rainwater Harvesting Site in Drought-Vulnerable Areas of Genale Dawa River Basin, Southern Ethiopia

MCDA Approach for Identification of Rainwater Harvesting Site in Drought-Vulnerable Areas of Genale Dawa River Basin, Southern Ethiopia

PrOsteoporosis: predicting osteoporosis risk using NHANES data and machine learning approach

ObjectivesOsteoporosis, prevalent among the elderly population, is primarily diagnosed through bone mineral density (BMD) testing, which has limitations in early detection. This study aims to develop and validate a machine learning approach for osteoporosis identification by integrating demographic data, laboratory and questionnaire data, offering a more practical and effective screening alternative.MethodsIn this study, data from the National Health and Nutrition Examination Survey were analyzed to explore factors linked to osteoporosis. After cleaning, 8766 participants with 223 variables were studied. Minimum Redundancy Maximum Relevance and SelectKBest were employed to select the import features. Four Machine learning algorithms (RF, NN, LightGBM and XGBoost.) were applied to examine osteoporosis, with performance comparisons made. Data balancing was done using SMOTE, and metrics like F1 score, and AUC were evaluated for each algorithm.ResultsThe LightGBM model outperformed others with an F1 score of 0.914, an MCC of 0.831, and an AUC of 0.970 on the training set. On the test set, it achieved an F1 score of 0.912, an MCC of 0.826, and an AUC of 0.972. Top predictors for osteoporosis were height, age, and sex.ConclusionsThis study demonstrates the potential of machine learning models in assessing an individual’s risk of developing osteoporosis, a condition that significantly impacts quality of life and imposes substantial healthcare costs. The superior performance of the LightGBM model suggests a promising tool for early detection and personalized prevention strategies. Importantly, identifying height, age, and sex as top predictors offers critical insights into the demographic and physiological factors that clinicians should consider when evaluating patients’ risk profiles.

A Novel Approach to IoT Device Identification via Anti‐Interference Dynamic Integral Neural Network and Multiobjective Fitness‐Dependent Optimizer Algorithm

ABSTRACTThe Internet of Things (IoT) has observed an accelerated development in the quantity of applications due to the rapid development of information technology. It can be difficult to identify IoT devices in heterogeneous, interference‐prone networks. The accuracy, optimization, and robustness of existing techniques are insufficient for dependable classification and application detection. To overcome this complication, a novel approach for IoT device identification using an anti‐interference dynamic integral neural network (AIDINN) and a multiobjective fitness‐dependent optimizer algorithm (MOFDOA) (IoT‐DTI‐AIDINN‐MOFDOA) is proposed. The input data are collected from the Network Traffic Dataset. Then, the input data are given to feature extraction. By using the synchro‐transient‐extracting transform (STET), the features are extracted from the dataset. Then the extracted features are given to AIDINN for IoT device identification, which classifies known IoT devices and unknown IoT devices. In general, AIDINN does not adopt any optimization techniques to determine the ideal parameters for ensuring an accurate IoT device identification. Hence, an MOFDOA is proposed here to optimize the AIDINN, which precisely constructs the IoT application detection. The performance measures like accuracy, precision, recall, specificity, F measure, computational time, and computational complexity are evaluated. The proposed IoT‐DTI‐AIDINN‐MOFDOA method attains a higher accuracy of 25.23%, 16.12%, and 21.27% and a higher precision of 25.26%, 16.22%, and 26.27% when analyzed with the following existing models: IoT device type detection using deep neural network (IoT‐DTI‐DNN), adversarial attacks and IoT for long short‐term memory (AA‐IoT‐LSTM), and IoT device identification depending on fully connected neural network (IoT‐DI‐FCNN), respectively.

A Sparse Representation Classification Framework for Person Identification and Verification Using Neurophysiological Signals

Brain biometrics has received increasing attention from the scientific community due to its unique properties in comparison to traditional biometric methods. Many studies have shown that EEG features are distinct among individuals. SSVEP signals, generated by stationary localized sources and distributed sources in the parietal and occipital regions of the brain, serve as a reliable basis for biometrics. In this study, we present a novel approach that leverages the spatial patterns of brain responses elicited by visual stimulation at specific frequencies. Specifically, we propose integrating common spatial patterns with Sparse Representation Classification (SRC) frameworks for person identification and verification. The use of common spatial patterns enables the design of personalized spatial filters, which play a crucial role in constructing the dictionary used by SRC frameworks. We conducted extensive evaluations of the proposed method, comparing it with several traditional approaches using two SSVEP datasets. Our analysis also explored a broad range of flickering frequencies in the SSVEP experiments. The results from these datasets demonstrated the effectiveness of our approach for person identification and verification, achieving an average correct recognition rate above 90% across various visual stimulus frequencies and short durations of electrophysiological signals.

Machine learning-based approach for identification of new resistance associated mutations from whole genome sequences of Mycobacterium tuberculosis

Abstract Motivation Currently available methods for the prediction of genotypic drug resistance in Mycobacterium tuberculosis utilize information on known markers of drug resistance. Hence, machine learning approaches are needed that can discover new resistance markers. Results Whole genome sequences with known phenotypic drug resistance profiles have been utilized to train XGBoost and ANN classifiers for 5 first-line and 8 second-line tuberculosis drugs. Benchmarking on a completely independent dataset from CRyPTIC database revealed that our method has high sensitivity (90%-95%) and specificity (94%-99%) for five first-line drugs and robust performance for six second-line drugs with a sensitivity of 77%-89% at over 95% specificity. An explainable AI method, SHapley Additive exPlanations, has successfully identified resistance mutations for each drug in a completely automated way. This approach could not only identify known resistance associated mutations in agreement with the WHO mutation catalogue, but also predicted more than 100 other potential resistance associated mutations for 13 antibiotics in new genes outside the known resistance loci. Identification of new resistance markers opens up the opportunity for the discovery of novel mechanisms of drug resistance. Availability and implementation Our prediction method has been implemented as TB-AMRpred webserver and command line tool, available freely at http://www.nii.ac.in/TB-AMRpred.htmlandhttps://github.com/Ankitapal1995/TB-AMRprd

Identification of Staphylococcus warneri from rainbow trout (Oncorhynchus mykiss Walbaum, 1792) using proteomics–based MALDI–TOF MS

Staphylococcus warneri, an opportunistic pathogen, is a causative agent of mortal diseases in rainbow trout farming (Oncorhynchus mykiss), which are of great economic value for Türkiye. In this study, in addition to traditional phenotypic, biochemical, histopathological and genetic methods, a high throughput proteomics based Matrix Assisted Laser Desorption/Ionization Time of Flight Mass Spectrometry (MALDI–TOF MS) method was performed for precise identification of S. warneri. Fourteen isolates obtained from skin, gills, liver, spleen and kidney of a total of fifty diseased fish were phenotypically confirmed as S. warneri using the BBL CrystalTMGP identification system. Only 43% of these isolates showed positive PCR amplification for the 16S rRNA and sodA (superoxide dismutase A) gene, while 100% were identified as S. warneri by MALDI–TOF MS technique with high mass score value (m/z) between 2.35 and 3.05. From the comparative data obtained, it was concluded that MALDI–TOF mass spectrometry analysis can be recommended for the definitive confirmation of S. warneri, which showed indistinguishably close similarities with 16S rRNA gene sequences and sodA PCR results. To the best of knowledge, this is the first report to validate the results of phenotypic, biochemical, genetic and histological methods by the MALDI–TOF MS and shows that this is a successful identification approach, providing a high mass score (m/z) with 100% matching for accurate and faster identification of S. warneri. This promising diagnostic technique can identify many different bacterial fish pathogens, although a larger protein mass database for aquatic organisms is needed.

Proteome‐Wide Mendelian Randomization Identifies Candidate Causal Proteins for Cardiovascular Diseases

AbstractIntegration of human genomics and other omics across different ancestries provides novel, affordable, and systematic approach for target identification. We used Mendelian randomization approaches to unravel causal associations between 2,940 circulating proteins and 19 CVD. We found 218 proteins that impacted risk of one or more CVDs through forward MR (106 and 182 using cis‐pQTLs only and cis‐ + trans‐pQTLs, respectively), among which 107 were previously reported as associated with CVD or CVD‐related traits. There were 102 proteins replicated (FDR < 5%, 53 with cis‐pQTLs only and 88 with cis‐ + trans‐pQTLs) using the FinnGen Olink data. BTN3A2 was highlighted as a novel candidate gene for ischemic stroke, suggesting a crosstalk between immune modulation and stroke pathogenesis. Single cell integration prioritized PAM for stable angina pectoris and ventricular arrhythmia and LPL for peripheral artery disease, whose transcriptional expressions were enriched in cardiomyocytes. Forward and reverse MR found largely non‐overlapping proteins (only 2 overlapped: LGALS4 and MMP12), suggesting distinct proteomic causes and consequences of CVD. Our study provides human genetics‐based evidence of novel candidate genes, a foundational step towards full‐scale causal human biology‐based drug discovery for CVD.

Cable-stayed bridge model updating using a novel approach for deflection influence line identification: Theoretical basis and field validation

Cable-stayed bridge model updating using a novel approach for deflection influence line identification: Theoretical basis and field validation

A machine learning approach for the rapid identification of measurable residual disease in acute myeloid leukemia.

Not available.

Rapid Identification of High-Prevalence Drug-Resistant Mutations in Mycobacterium tuberculosis Using a Portable Colorimetric Array.

The emergence and rise of drug-resistant mutations in Mycobacterium tuberculosis (TB) exacerbates the challenges of TB controlling and treatment. 95% of the TB-related deaths occurred in low- and middle-income countries (LMICs), where there is a lack of access to quality diagnostics for drug-resistant mutations. Here, we present an approach for rapid identification of 24 high-prevalence drug-resistant mutations in M. tuberculosis, in which a portable thermal cycler and a photosensitization colorimetric assay (PCA) were employed for polymerase chain reaction (PCR) and subsequent analysis of the PCR amplicons, respectively. To avoid potential "off-target" effects for mutation detection, we introduced two reverse primers (binding two regions) to ensure the successful amplification of PCR. An analysis of 50 clinical samples revealed that our method achieved a sensitivity of 100%, a specificity of 99.48%, and an accuracy of 99.50%, with the results of the clinical approach serving as the standard reference. Furthermore, a portable colorimetric array (PoCA) was designed to streamline the operations of PCA, thereby increasing the number of detectable mutations per round and enhancing the feasibility of drug-resistant mutation detection in LMICs. To facilitate real-world applications, uracil-DNA glycosylase (UDG) and deoxyuridine triphosphate (dUTP) were integrated into the PCR system, thereby reducing potential cross-contamination. Due to its easy accessibility, high accuracy, and rapid turnaround time, this approach offers a promising option for large-scale identification of drug-resistant mutations in M. tuberculosis.

DHUpredET: A comparative computational approach for identification of dihydrouridine modification sites in RNA sequence.

DHUpredET: A comparative computational approach for identification of dihydrouridine modification sites in RNA sequence.

Concrete strength monitoring and damage detection using piezoelectric-based wireless sensor

Concrete stands as the predominant and extensively employed structural material in civil engineering. Timely evaluation of concrete strength is essential for maintaining structural safety and minimizing construction delays, thereby avoiding possible structural failures. This early check also ensures that concrete structures can bear loads during construction and throughout their service life. A significant challenge in the construction industry is accurately determining the strength of early-age concrete and identifying potential damage without destructive testing. Conventional techniques often require time-consuming procedures and can be impractical for real-time monitoring. To overcome this issue, IoT-based monitoring systems with piezoelectric sensors provide a practical approach for continuous strength monitoring and damage identification in concrete structures. This study utilized a piezoelectric lead zirconate titanate (PZT) sensor, employing a surface-bonding technique to attach the sensor to the cube specimens. The experimental phase spanned four weeks, including intervals of 7th, 14th, and 28th days by standard construction codes to assess necessary strength levels and identify any damage within the structure. This study verified the results obtained by the piezoelectric-based wireless sensor network are practical and reliable. Correlation coefficient values are analyzed to validate the relationship between data from IoT-based testing and compressive strength. All the results are presented in graphical format, confirming that this non-destructive approach can accurately predict concrete strength and identify structural damage. This study uniquely contributed by validating the use of piezoelectric sensors for continuous, in-situ monitoring of concrete, providing a novel approach to early damage detection and evaluating the structural health of the structure.

Genomic and fragmentomic landscapes of cell-free DNA for early cancer detection.

Genomic analyses of cell-free DNA (cfDNA) in plasma are enabling noninvasive blood-based biomarker approaches to cancer detection and disease monitoring. Current approaches for identification of circulating tumour DNA typically use targeted tumour-specific mutations or methylation analyses. An emerging approach is based on the recognition of altered genome-wide cfDNA fragmentation in patients with cancer. Recent studies have revealed a multitude of characteristics that can affect the compendium of cfDNA fragments across the genome, collectively called the 'cfDNA fragmentome'. These changes result from genomic, epigenomic, transcriptomic and chromatin states of an individual and affect the size, position, coverage, mutation, structuraland methylation characteristics of cfDNA. Identifying and monitoring these changes has the potential to improve early detection of cancer, especially using highly sensitive multi-feature machine learning approaches that would be amenable to broad use in populations at increased risk. This Review highlights the rapidly evolving field of genome-wide analyses of cfDNA characteristics, their comparison to existing cfDNA methods, and recent related innovations at the intersection of large-scale sequencing and artificial intelligence. As the breadth of clinical applications of cfDNA fragmentome methods have enormous public health implications for cancer screening and personalized approaches for clinical management of patients with cancer, we outline the challenges and opportunities ahead.

Precise mycobacterial species and subspecies identification using the PEP-TORCH peptidome algorithm.

Mycobacterial infections pose a significant global health concern, requiring precise identification for effective treatment. However, diagnosing them is challenging due to inaccurate identifications and prolonged times. In this study, we aimed to develop a novel peptidome-based method using mycobacterial growth indicator tube (MGIT) cultures for faster and more accurate identification. We created the PEPtide Taxonomy/ORganism CHecking (PEP-TORCH), an algorithm that analyzes tryptic peptides identified by mass spectrometry to diagnose species and subspecies with predominance scores. PEP-TORCH demonstrated 100% accuracy in identifying mycobacterial species, subspecies, and co-infections in 81 individuals suspected of mycobacterial infections, eliminating the need for a sub-solid culture procedure, the gold standard in clinical practice. A notable strength of PEP-TORCH is its ability to provide information on species and subspecies simultaneously, a process conventionally achieved sequentially. This capability significantly expedites pathogen identification. Furthermore, a targeted proteomics method was validated in 63 clinical samples using the taxa-specific peptides selected by PEP-TORCH, making them suitable as biomarkers in more clinically friendly settings. This comprehensive identification approach holds promise for streamlining treatment strategies in clinical practice.

A Risk-Informed Design Framework for Functional Safety System Design of Human–Robot Collaboration Applications

The safety of robotics and automation technologies is a significant concern for stakeholders in Industry 5.0. Ensuring cost-effectiveness and inherent safety requires applying the defense-in-depth principle. This paper introduces a novel risk-informed design framework for functional safety, integrating function-centered hazard identification and risk assessment via fault tree analysis (FTA). Demonstrated in the design of a semi-automated agricultural vehicle, the framework begins with a function-centered hazard identification approach (F-CHIA) based on ISO 12100. It examined design intents, identified hazard zones, and conducted task and function identification. Foreseeable functional hazardous situations are analyzed, leading to functional requirements and the identification of relevant directives, regulations, and standards. The F-CHIA outputs inform the functional safety analysis, assessing the required performance level and deriving specific requirements for software, hardware, and human operators using FTA. The functional requirements derived from F-CHIA are more systematic than traditional methods and serve as effective inputs for functional safety analysis in human–robot collaboration applications. The proposed framework enables design teams to focus on enhancing factors that improve functional safety performance levels, resulting in a more thorough and effective safety design process.