High False Alarm Rate Research Articles

Early fire detection technology based on improved transformers in aircraft cargo compartments

The implementation of early and accurate detection of aircraft cargo compartment fire is of great significance to ensure flight safety. The current airborne fire detection technology mostly relies on single-parameter smoke detection using infrared light. This often results in a high false alarm rate in complex air transportation environments. The traditional deep learning model struggles to effectively address the issue of long-term dependency in multivariate fire information. This paper proposes a multi-technology collaborative fire detection method based on an improved transformers model. Dual-wavelength optical sensors, flue gas analyzers, and other equipment are used to carry out multi-technology collaborative detection methods and characterize various feature dimensions of fire to improve detection accuracy. The improved Transformer model which integrates the self-attention mechanism and position encoding mechanism is applied to the problem of long-time series modeling of fire information from a global perspective, which effectively solves the problem of gradient disappearance and gradient explosion in traditional RNN (recurrent neural network) and CNN (convolutional neural network). Two different multi-head self-attention mechanisms are used to classify and model multivariate fire information, respectively, which solves the problem of confusing time series modeling and classification modeling in dealing with multivariate classification tasks by a single attention mechanism. Finally, the output results of the two models are fused through the gate mechanism. The research results show that, compared with the traditional single-feature detection technology, the multi-technology collaborative fire detection method can better capture fire information. Compared with the traditional deep learning model, the multivariate fire prediction model constructed by the improved Transformer can better detect fires, and the accuracy rate is 0.995.

СПОСОБ ОБЪЕКТНО-ОРИЕНТИРОВАННОГО РАСПОЗНАВАНИЯ ОБЛАЧНОСТИ ГЛУБОКОЙ КОНВЕКЦИИ НА ОСНОВЕ ДАННЫХ ГЕОСТАЦИОНАРНОГО МЕТЕОРОЛОГИЧЕСКОГО ИСКУССТВЕННОГО CПУТНИКА ЗЕМЛИ С ПРИМЕНЕНИЕМ МАШИННОГО ОБУЧЕНИЯ

Due to high spatial and temporal resolution, geostationary meteorological satellite imagery is a valuable source of information on the development of deep convective clouds and related severe weather events. Some methods for automatic deep convection detection from satellite data provide a satisfactory probability of detection for independent datasets, but are characterized by a high false alarm rate. The paper gives a description of an algorithm for automatic detection of deep convective clouds with satellite imagery using gradient boosting, logistic regression, and artificial neural network models. The results of validation of the proposed method using dependent and independent data of ground-based observations for the period 2013-2020 are presented. A low false alarm rate and high probability of detection suggest that the algorithm can be used in the operational mode.

Performance of T2-based PCA mix control chart with KDE control limit for monitoring variable and attribute characteristics

In this work, the mixed multivariate T2 control chart’s detailed performance evaluation based on PCA mix is explored. The control limit of the proposed control chart is calculated using the kernel density approach. Through simulation studies, the proposed chart’s performance is assessed in terms of its capacity to identify outliers and process shifts. When 30% more outliers are included in the data, the proposed chart provides a consistent accuracy rate for identifying mixed outliers. For the balanced percentage of attribute qualities, misdetection happens because of the high false alarm rate. For unbalanced attribute qualities and excessive proportions, the masking effect is the key issue. The proposed chart shows the improved performance for the shift in identifying the shift in the process.

YOLOv8-BYTE: Ship tracking algorithm using short-time sequence SAR images for disaster response leveraging GeoAI

Ship tracking technology is crucial for emergency rescue in the event of a disaster. Quickly identifying the position and status of vessels is vital for rescue teams to be able to deploy efficiently in disaster areas. When responding to emergencies or natural disasters, ship tracking technology plays a critical role in supporting emergency rescue operations and resource allocation, improving the overall resilience of the maritime transportation system. However, the research on multi-object tracking (MOT) algorithms has primarily focused on optical image datasets. In contrast, image data from synthetic aperture radar (SAR) presents unique challenges, such as defocus interference, a high false alarm rate, and a lack of prior samples. To overcome these particular challenges, we propose a robust MOT algorithm developed for SAR images to achieve effective multi-vessel tracking under difficult imaging conditions. In particular, we optimize the YOLOv8 detection network by introducing a diffusion model-based training method for data augmentation. This method improves the robustness of the network to scaling, rotational and translational deformations. Moreover, an enhanced swin transformer is proposed as a feature extraction network, which strengthens the representation capability of the detection network. Furthermore, the state parameters within the KF technique are enhanced by directly capturing the details of the height and width of the tracking rectangle box. This refinement of the ByteTrack algorithm aims to achieve a more precise and accurate fit of the tracking rectangle to the ship, further improving the overall tracking performance. The experimental results from the ship detection and multiple objects tracking datasets show the impressive performance of the proposed model. With a precision of 97.60%, a recall of 96.36%, and an average precision of 96.72%, the model achieves exceptional detection accuracy with an 18% reduction in model parameters. Furthermore, significant improvements can be observed in key tracking metrics such as HOTA, MOTA and IDF1, with improvements of 4.8%, 8.5% and 6.8% respectively compared to the baseline algorithm, alongside a remarkable 37.5% reduction in IDS. It is noteworthy that the tracker works in real time, achieving an average analysis speed of 47 frames per second. The proposed MOT algorithm achieves state-of-art tracking performance on a SAR image dataset with short time sequences. Therefore, the proposed approach is a compelling solution for ship tracking in SAR imagery.

A Dual-Hierarchy Synchronization Method for Signal Preambles with High Detection Rates for Satellite-Based ADS-B Receivers with Different Sensitivities

Existing methods are unable to achieve high detection rates and low false alarm rates of satellite-based Automatic Dependent Surveillance-Broadcast (ADS-B) signal preambles at extremely low signal-to-noise ratios (SNRs) using limited on-star resources. In this paper, a dual-hierarchy synchronization method is proposed, including a first-level coarse synchronization and a second-level fine synchronization. The coarse synchronization process involves three steps: (1) detection of unknown signals, (2) soft decision, and (3) adaptive interval output. The first step introduces the threshold (TMSED) of the minimum signal energy to be detected to guarantee a high detection rate. In the soft decision step, a value (SV) designed to improve the robustness of the system curbs false detection caused by noise interference. In the last step, the coarse synchronization interval radius (r) is mapped out according to the SNR to reduce resource consumption. The fine synchronization process is based on the coarse synchronization output, and the correlation peak is calculated to complete the synchronization of the signal preambles. The results show that the proposed method achieves a high detection rate of 96% at an extremely low SNR using a low sampling frequency of 10 MHz. Furthermore, the adjustment of TMSED allows this method to be applied to ADS-B receivers with different sensitivities. The comprehensive performance of this method to achieve high detection rates and acceptable false alarm rates at extremely low SNRs with limited on-star resources is verified by final simulations to be superior to other methods.

Precise and low-power closed-loop neuromodulation through algorithm-integrated circuit co-design.

Implantable neuromodulation devices have significantly advanced treatments for neurological disorders such as Parkinson's disease, epilepsy, and depression. Traditional open-loop devices like deep brain stimulation (DBS) and spinal cord stimulators (SCS) often lead to overstimulation and lack adaptive precision, raising safety and side-effect concerns. Next-generation closed-loop systems offer real-time monitoring and on-device diagnostics for responsive stimulation, presenting a significant advancement for treating a range of brain diseases. However, the high false alarm rates of current closed-loop technologies limit their efficacy and increase energy consumption due to unnecessary stimulations. In this study, we introduce an artificial intelligence-integrated circuit co-design that targets these issues and using an online demonstration system for closed-loop seizure prediction to showcase its effectiveness. Firstly, two neural network models are obtained with neural-network search and quantization strategies. A binary neural network is optimized for minimal computation with high sensitivity and a convolutional neural network with a false alarm rate as low as 0.1/h for false alarm rejection. Then, a dedicated low-power processor is fabricated in 55 nm technology to implement the two models. With reconfigurable design and event-driven processing feature the resulting application-specific integrated circuit (ASIC) occupies only 5mm2 silicon area and the average power consumption is 142 μW. The proposed solution achieves a significant reduction in both false alarm rates and power consumption when benchmarked against state-of-the-art counterparts.

Spectral-Spatial Feature Fusion for Hyperspectral Anomaly Detection.

Hyperspectral anomaly detection is used to recognize unusual patterns or anomalies in hyperspectral data. Currently, many spectral-spatial detection methods have been proposed with a cascaded manner; however, they often neglect the complementary characteristics between the spectral and spatial dimensions, which easily leads to yield high false alarm rate. To alleviate this issue, a spectral-spatial information fusion (SSIF) method is designed for hyperspectral anomaly detection. First, an isolation forest is exploited to obtain spectral anomaly map, in which the object-level feature is constructed with an entropy rate segmentation algorithm. Then, a local spatial saliency detection scheme is proposed to produce the spatial anomaly result. Finally, the spectral and spatial anomaly scores are integrated together followed by a domain transform recursive filtering to generate the final detection result. Experiments on five hyperspectral datasets covering ocean and airport scenes prove that the proposed SSIF produces superior detection results over other state-of-the-art detection techniques.

STNet: A Time-Frequency Analysis-Based Intrusion Detection Network for Distributed Optical Fiber Acoustic Sensing Systems.

Distributed optical fiber acoustic sensing (DAS) is promising for long-distance intrusion-anomaly detection tasks. However, realistic settings suffer from high-intensity interference noise, compromising the detection performance of DAS systems. To address this issue, we propose STNet, an intrusion detection network based on the Stockwell transform (S-transform), for DAS systems, considering the advantages of the S-transform in terms of noise resistance and ability to detect disturbances. Specifically, the signal detected by a DAS system is divided into space-time data matrices using a sliding window. Subsequently, the S-transform extracts the time-frequency features channel by channel. The extracted features are combined into a multi-channel time-frequency feature matrix and presented to STNet. Finally, a non-maximum suppression algorithm (NMS), suitable for locating intrusions, is used for the post-processing of the detection results. To evaluate the effectiveness of the proposed method, experiments were conducted using a realistic high-speed railway environment with high-intensity noise. The experimental results validated the satisfactory performance of the proposed method. Thus, the proposed method offers an effective solution for achieving high intrusion detection rates and low false alarm rates in complex environments.

Many Models, Little Adoption-What Accounts for Low Uptake of Machine Learning Models for Atrial Fibrillation Prediction and Detection?

(1) Background: Atrial fibrillation (AF) is a major risk factor for stroke and is often underdiagnosed, despite being present in 13-26% of ischemic stroke patients. Recently, a significant number of machine learning (ML)-based models have been proposed for AF prediction and detection for primary and secondary stroke prevention. However, clinical translation of these technological innovations to close the AF care gap has been scant. Herein, we sought to systematically examine studies, employing ML models to predict incident AF in a population without prior AF or to detect paroxysmal AF in stroke cohorts to identify key reasons for the lack of translation into the clinical workflow. We conclude with a set of recommendations to improve the clinical translatability of ML-based models for AF. (2) Methods: MEDLINE, Embase, Web of Science, Clinicaltrials.gov, and ICTRP databases were searched for relevant articles from the inception of the databases up to September 2022 to identify peer-reviewed articles in English that used ML methods to predict incident AF or detect AF after stroke and reported adequate performance metrics. The search yielded 2815 articles, of which 16 studies using ML models to predict incident AF and three studies focusing on ML models to detect AF post-stroke were included. (3) Conclusions: This study highlights that (1) many models utilized only a limited subset of variables available from patients' health records; (2) only 37% of models were externally validated, and stratified analysis was often lacking; (3) 0% of models and 53% of datasets were explicitly made available, limiting reproducibility and transparency; and (4) data pre-processing did not include bias mitigation and sufficient details, leading to potential selection bias. Low generalizability, high false alarm rate, and lack of interpretability were identified as additional factors to be addressed before ML models can be widely deployed in the clinical care setting. Given these limitations, our recommendations to improve the uptake of ML models for better AF outcomes include improving generalizability, reducing potential systemic biases, and investing in external validation studies whilst developing a transparent modeling pipeline to ensure reproducibility.

Performance comparison of 6 in-hospital patient monitoring systems in the detection and alarm of ventricular cardiac arrhythmias

BackgroundPatient monitoring devices are critical in alerting potential cardiac arrhythmias during hospitalization; however, there are concerns of alarm fatigue due to high false alarm rates. ObjectiveTo evaluate sensitivity and false alarm rate of hospital-based continuous electrocardiographic (ECG) monitoring technologies. MethodsSix commonly-used multiparameter bedside monitoring systems available in the United States were evaluated: B125M (GE HealthCare), ePM10 and iPM12 (Mindray), Efficia and IntelliVue (Philips), and Life Scope (Nihon Kohden). Sensitivity was tested using ECG recordings containing 57 true ventricular tachycardia (VT) events. False positive rate testing used 205 patient-hours of ECG recordings containing no cardiac arrhythmias. Signals from ECG recordings were fed to devices simultaneously; high-severity arrhythmia alarms were tracked. Sensitivity to true VT events and false positive rates were determined. Differences were assessed using Fisher’s exact tests (sensitivity) and Z-tests (false positive rates). ResultsB125M raised 56 total alarms for 57 annotated VT events and had the highest sensitivity (98%, P<0.05), followed by iPM12 (84%), Life Scope (81%), Efficia (79%), ePM10 (77%), and IntelliVue (75%). B125M raised 20 false alarms, significantly lower (P<0.0001) than iPM12 (284), Life Scope (292), IntelliVue (304), ePM10 (324), and Efficia (493). The most common false alarm was VT, followed by non-sustained VT. ConclusionWe found significant performance differences among multiparameter bedside ECG monitoring systems using previously-collected recordings. B125M had the highest sensitivity in detecting true VT events and lowest false alarm rate. These results can assist in minimizing alarm fatigue and optimizing patient safety by careful selection of in-hospital continuous monitoring technology.

Ensemble learning classifiers hybrid feature selection for enhancing performance of intrusion detection system

Feature selection (FS) plays an important role in the construction of efficient ensemble classifiers; particularly for intrusion detection system (IDS). An IDS is a utilized in a network architecture to protect the availability of sensitive information. However, existing IDSs suffer from redundancy, high dimensionality, and high false alarm rate (FAR). Also, lots of models are constructed for outdated datasets, which makes them less flexible to deal with new assaults. Therefore, this paper proposes a new IDS relies on hybrid FS and ensemble classifiers. A hybrid FS approach consists of two techniques, hard-voting and mean. In contrast to recent papers, we use three different FS approaches: extra tree classifier importance as an embedded FS, recursive feature elimination (RFE) as a wrapper FS, and mutual information (MI) as a filter FS. Then, a hard-voting technique has been used to fuse output of these approaches and obtain a reduced subset of features. Since each feature has three weights, a mean technique has been utilized to assign one weight to each feature and obtain an optimal subset of features. The experimental outcomes, utilizing the modern InSDN dataset, confirm that the proposed hybrid FS with ensemble soft voting classifier achieves better results than other ensemble and individual classifiers due to several measures.

AE‐Integrated: Real‐time network intrusion detection with Apache Kafka and autoencoder

SummaryUnknown cyber‐attack detection in network traffic streams is challenging but crucial to ensure network security. It is observed that new security threats occur on a daily basis and make cyberspace vulnerable. In the literature, machine learning and deep learning‐based network intrusion detection systems have gained a lot of success but still face many challenges in detecting new security threats and unknown cyber‐attacks in real‐time. Additionally, high false alarm rates and real‐time detection in constantly evolving high‐dimensional network data streams are open issues for the research community. To address this issue, a DL‐based solution is developed to detect real‐time network anomalies in streaming data with high detection accuracy, precision, recall and low false negative and positive scenarios. The proposed novel algorithm, AE‐Integrated, is developed and evaluated on the latest CICIDS‐2017 dataset. The AE‐Integrated is updated with the newest network traffic data stream by the human administrator after a certain period to maintain its prediction accuracy for future inference. The simulation study is conducted with the Apache Kafka and Slack API to get real‐time anomaly alerts. Finally, we compared the result with recent state‐of‐the‐art research to evaluate the significance of the proposed algorithm. It is concluded that combining multiple lightweight autoencoders into a single large architecture provides optimal results. The accuracy, recall, and AUC of AE‐Integrated obtained are 99.54%, 99.53%, and 0.998, respectively.

Embedded Spatial–Temporal Convolutional Neural Network Based on Scattered Light Signals for Fire and Interferential Aerosol Classification

Photoelectric smoke detectors are the most cost-effective devices for very early warning fire alarms. However, due to the different light intensity response values of different kinds of fire smoke and interference from interferential aerosols, they have a high false-alarm rate, which limits their popularity in Chinese homes. To address these issues, an embedded spatial–temporal convolutional neural network (EST-CNN) model is proposed for real fire smoke identification and aerosol (fire smoke and interferential aerosols) classification. The EST-CNN consists of three modules, including information fusion, scattering feature extraction, and aerosol classification. Moreover, a two-dimensional spatial–temporal scattering (2D-TS) matrix is designed to fuse the scattered light intensities in different channels and adjacent time slices, which is the output of the information fusion module and the input for the scattering feature extraction module. The EST-CNN is trained and tested with experimental data measured on an established fire test platform using the developed dual-wavelength dual-angle photoelectric smoke detector. The optimal network parameters were selected through extensive experiments, resulting in an average classification accuracy of 98.96% for different aerosols, with only 67 kB network parameters. The experimental results demonstrate the feasibility of installing the designed EST-CNN model directly in existing commercial photoelectric smoke detectors to realize aerosol classification.

Improved target detection method for space-based optoelectronic systems

The detection of faint and small targets by space-based surveillance systems is difficult owing to the long distances, low energies, high speeds, high false alarm rates, and low algorithmic efficiencies involved in the process. To improve space object detection and help prevent collisions with critical facilities such as satellites, this study proposes an improved method for the detection of faint and small space-based targets. The proposed method consists of two components: star atlas preprocessing and space-based target detection. The star atlas preprocessing step applies multi-exposure image pyramidal weighted fusion to the original image containing the faint and small space-based target. After obtaining the image pyramidal weighted fusion result atlas, the algorithm employs threshold segmentation to improve the overall image clarity, highlight image details, and provide additional information for target detection. The detection of targets partially relies on the local symmetry of the image. Accordingly, a diffusion function describing the local symmetry is established to precisely locate stars by measuring the symmetry factor in a small area surrounding each pixel in the star atlas. This effectively removes the background stars while retaining high-definition and high-contrast images. The efficacy of the algorithm is validated using simulated datasets consisting of space-based and real images. The results demonstrate that the proposed technique improves the applicability of the multistage hypothesis testing (MHT) method in the context of a complex space environment, thus improving the performance of the space-based electro-optical detection system to better catalogue, identify, and track space targets.

Validation of Machine Learning-Aided and Power Line Communication-Based Cable Monitoring Using Measurement Data.

The implementation of power line communications (PLC) in smart electricity grids provides us with exciting opportunities for real-time cable monitoring. In particular, effective fault classification and estimation methods employing machine learning (ML) models have been proposed in the recent past. Often, the research works presenting PLC for ML-aided cable diagnostics are based on the study of synthetically generated channel data. In this work, we validate ML-aided diagnostics by integrating measured channels. Specifically, we consider the concatenation of clustering as a data pre-processing procedure and principal component analysis (PCA)-based dimension reduction for cable anomaly detection. Clustering and PCA are trained with measurement data when the PLC network is working under healthy conditions. A possible cable anomaly is then identified from the analysis of the PCA reconstruction error for a test sample. For the numerical evaluation of our scheme, we apply an experimental setup in which we introduce degradations to power cables. Our results show that the proposed anomaly detector is able to identify a cable degradation with high detection accuracy and low false alarm rate.

An Angular Acceleration Based Looming Detector for Moving UAVs.

Visual perception equips unmanned aerial vehicles (UAVs) with increasingly comprehensive and instant environmental perception, rendering it a crucial technology in intelligent UAV obstacle avoidance. However, the rapid movements of UAVs cause significant changes in the field of view, affecting the algorithms' ability to extract the visual features of collisions accurately. As a result, algorithms suffer from a high rate of false alarms and a delay in warning time. During the study of visual field angle curves of different orders, it was found that the peak times of the curves of higher-order information on the angular size of looming objects are linearly related to the time to collision (TTC) and occur before collisions. This discovery implies that encoding higher-order information on the angular size could resolve the issue of response lag. Furthermore, the fact that the image of a looming object adjusts to meet several looming visual cues compared to the background interference implies that integrating various field-of-view characteristics will likely enhance the model's resistance to motion interference. Therefore, this paper presents a concise A-LGMD model for detecting looming objects. The model is based on image angular acceleration and addresses problems related to imprecise feature extraction and insufficient time series modeling to enhance the model's ability to rapidly and precisely detect looming objects during the rapid self-motion of UAVs. The model draws inspiration from the lobula giant movement detector (LGMD), which shows high sensitivity to acceleration information. In the proposed model, higher-order information on the angular size is abstracted by the network and fused with multiple visual field angle characteristics to promote the selective response to looming objects. Experiments carried out on synthetic and real-world datasets reveal that the model can efficiently detect the angular acceleration of an image, filter out insignificant background motion, and provide early warnings. These findings indicate that the model could have significant potential in embedded collision detection systems of micro or small UAVs.

Use of ZDR columns for early detection of severe convection within the operational radar network of the United Kingdom

AbstractDifferential reflectivity () columns were observed using a Met Office three‐dimensional radar composite. An algorithm for automatic detection of columns was developed, based on dB and dBZ. Across three case days, detected columns were found to precede severe convection in tracked convective cells with a range of lead times from 0 to 20 min depending on the case day. Requiring maxima above 1.4 dB and 30 dBZ of and respectively was an appropriate second condition for all three cases although the skill in the early detection of severe convection varied across case days. Despite the high probability of detections, the high false alarm rate accompanied by low critical success index and data latency limit performance based on the three cases considered in this study. Nevertheless, the ability to detect columns in operational radar data with a useful lead time prior to severe convection in certain conditions is a promising development towards advancing nowcasting of severe convection in the United Kingdom.

Time-slice dynamic prediction and multiway serial PCA for batch industrial process monitoring

This paper introduces an innovative method for batch industrial process monitoring and fault detection. By combining time-slice dynamic prediction with an enhanced adaptive multiway principal component analysis (PCA) technique, it aims to balance the trade-off between high fault detection rates and low false alarm rates in monitoring batch industrial data. The method extracts dynamic features from 3D batch data time slices and utilizes a serial PCA framework with variable control limits for real-time monitoring. Comparative evaluations against traditional adaptive multiway PCA and dynamic time-slice canonical correlation analysis approaches demonstrate its superior performance. A case study using a novel cold rolling mill simulation model underscores its effectiveness for batch industrial process monitoring and fault detection.

Explainable Artificial Intelligence Warning Model Using an Ensemble Approach for In-Hospital Cardiac Arrest Prediction: Retrospective Cohort Study.

Cardiac arrest (CA) is the leading cause of death in critically ill patients. Clinical research has shown that early identification of CA reduces mortality. Algorithms capable of predicting CA with high sensitivity have been developed using multivariate time series data. However, these algorithms suffer from a high rate of false alarms, and their results are not clinically interpretable. We propose an ensemble approach using multiresolution statistical features and cosine similarity-based features for the timely prediction of CA. Furthermore, this approach provides clinically interpretable results that can be adopted by clinicians. Patients were retrospectively analyzed using data from the Medical Information Mart for Intensive Care-IV database and the eICU Collaborative Research Database. Based on the multivariate vital signs of a 24-hour time window for adults diagnosed with heart failure, we extracted multiresolution statistical and cosine similarity-based features. These features were used to construct and develop gradient boosting decision trees. Therefore, we adopted cost-sensitive learning as a solution. Then, 10-fold cross-validation was performed to check the consistency of the model performance, and the Shapley additive explanation algorithm was used to capture the overall interpretability of the proposed model. Next, external validation using the eICU Collaborative Research Database was performed to check the generalization ability. The proposed method yielded an overall area under the receiver operating characteristic curve (AUROC) of 0.86 and area under the precision-recall curve (AUPRC) of 0.58. In terms of the timely prediction of CA, the proposed model achieved an AUROC above 0.80 for predicting CA events up to 6 hours in advance. The proposed method simultaneously improved precision and sensitivity to increase the AUPRC, which reduced the number of false alarms while maintaining high sensitivity. This result indicates that the predictive performance of the proposed model is superior to the performances of the models reported in previous studies. Next, we demonstrated the effect of feature importance on the clinical interpretability of the proposed method and inferred the effect between the non-CA and CA groups. Finally, external validation was performed using the eICU Collaborative Research Database, and an AUROC of 0.74 and AUPRC of 0.44 were obtained in a general intensive care unit population. The proposed framework can provide clinicians with more accurate CA prediction results and reduce false alarm rates through internal and external validation. In addition, clinically interpretable prediction results can facilitate clinician understanding. Furthermore, the similarity of vital sign changes can provide insights into temporal pattern changes in CA prediction in patients with heart failure-related diagnoses. Therefore, our system is sufficiently feasible for routine clinical use. In addition, regarding the proposed CA prediction system, a clinically mature application has been developed and verified in the future digital health field.

Fault detection of high-speed train axle bearings based on a hybridized physical and data-driven temperature model

As a key component connecting axle and frame, axle bearings of the high-speed train are essential for the safe operation of modern high-speed trains. The current on-board fault detection system, which compares the axle box temperature measured from the sensor to a predefined threshold value, faces problems such as poor adaptability to varied service conditions, undesired high rate of false alarms and missing alarms, and lack of potential to be optimized. To solve these problems, in the present study, a novel hybrid model is proposed which predicts the temperature of a healthy bearing under the given service condition, and based on that detects the bearing fault. The hybrid model includes a physical model reflecting the thermal behavior of the axle box and two data-driven models adopting the same inputs but representing the relationship by means of back-propagation (BP) neural network and long-short term memory neural network (LSTM). After calibrating these models by real-life operation data collected from 2.25 million kilometers and using all results as inputs, a fault discriminant mechanism is established. The performance of this new mechanism is superior than the current on-board system and all stand-alone models in terms of indicators such as precision, recall and F1 score. This way, the study shows that similar hybrid models can be a viable solution for the future on-board fault detection system.