-
Scientific Reports Jun 2024According to the literature, seizure prediction models should be developed following a patient-specific approach. However, seizures are usually very rare events, meaning...
According to the literature, seizure prediction models should be developed following a patient-specific approach. However, seizures are usually very rare events, meaning the number of events that may be used to optimise seizure prediction approaches is limited. To overcome such constraint, we analysed the possibility of using data from patients from an external database to improve patient-specific seizure prediction models. We present seizure prediction models trained using a transfer learning procedure. We trained a deep convolutional autoencoder using electroencephalogram data from 41 patients collected from the EPILEPSIAE database. Then, a bidirectional long short-term memory and a classifier layers were added on the top of the encoder part and were optimised for 24 patients from the Universitätsklinikum Freiburg individually. The encoder was used as a feature extraction module. Therefore, its weights were not changed during the patient-specific training. Experimental results showed that seizure prediction models optimised using pretrained weights present about four times fewer false alarms while maintaining the same ability to predict seizures and achieved more 13% validated patients. Therefore, results evidenced that the optimisation using transfer learning was more stable and faster, saving computational resources. In summary, adopting transfer learning for seizure prediction models represents a significant advancement. It addresses the data limitation seen in the seizure prediction field and offers more efficient and stable training, conserving computational resources. Additionally, despite the compact size, transfer learning allows to easily share data knowledge due to fewer ethical restrictions and lower storage requirements. The convolutional autoencoder developed in this study will be shared with the scientific community, promoting further research.
Topics: Humans; Seizures; Electroencephalography; Databases, Factual; Machine Learning; Female; Male; Neural Networks, Computer; Adult
PubMed: 38898066
DOI: 10.1038/s41598-024-64802-1 -
F1000Research 2024Researchers are focusing their emphasis on quick and real-time healthcare and monitoring systems because of the contemporary modern world's rapid technological...
BACKGROUND
Researchers are focusing their emphasis on quick and real-time healthcare and monitoring systems because of the contemporary modern world's rapid technological improvements. One of the best options is smart healthcare, which uses a variety of on-body and off-body sensors and gadgets to monitor patients' health and exchange data with hospitals and healthcare professionals in real time. Utilizing the primary user (PU) spectrum, cognitive radio (CR) can be highly useful for efficient and intelligent healthcare systems to send and receive patient health data.
METHODS
In this work, we propose a method that combines energy detection (ED) and cyclostationary (CS) spectrum sensing (SS) algorithms. This method was used to test spectrum sensing in CR-based smart healthcare systems. The proposed ED-CS in cognitive radio systems improves the precision of the spectrum sensing. Owing to its straightforward implementation, ED is initially used to identify the idle spectrum. If the ED cannot find the idle spectrum, the signals are found using CS-SS, which uses the cyclic statistical properties of the signals to separate the main users from the interference.
RESULTS
In the simulation analysis, the probability of detection (Pd), probability of a false alarm (Pfa), power spectral density (PSD), and bit error rate (BER) of the proposed ED-CS is compared to those of the traditional Matched Filter (MF), ED, and CS.
CONCLUSIONS
The results indicate that the suggested strategy improves the performance of the framework, making it more appropriate for smart healthcare applications.
Topics: Humans; Algorithms; Delivery of Health Care; Spectrum Analysis
PubMed: 38895702
DOI: 10.12688/f1000research.144624.2 -
BioRxiv : the Preprint Server For... Jun 2024Resident memory T cells (T ) have been described in barrier tissues as having a 'sensing and alarm' function where, upon sensing cognate antigen, they alarm the...
Resident memory T cells (T ) have been described in barrier tissues as having a 'sensing and alarm' function where, upon sensing cognate antigen, they alarm the surrounding tissue and orchestrate local recruitment and activation of immune cells. In the immunologically unique and tightly restricted CNS, it remains unclear if and how brain T , which express the inhibitory receptor PD-1, alarm the surrounding tissue during antigen re-encounter. Here, we reveal that T are sufficient to drive the rapid remodeling of the brain immune landscape through activation of microglia, DCs, NK cells, and B cells, expansion of Tregs, and recruitment of macrophages and monocytic dendritic cells. Moreover, we report that while PD-1 restrains granzyme B expression by reactivated brain T , it has no effect on cytotoxicity or downstream alarm responses. We conclude that T are sufficient to trigger rapid immune activation and recruitment in the CNS and may have an unappreciated role in driving neuroinflammation.
PubMed: 38895249
DOI: 10.1101/2024.06.06.597370 -
Heliyon May 2024Most accidents in a chemical process are caused by abnormal or deviations of the process parameters, and the existing research is focused on short-term prediction. When...
Most accidents in a chemical process are caused by abnormal or deviations of the process parameters, and the existing research is focused on short-term prediction. When the early warning time is advanced, many false and missing alarms will occur in the system, which will cause certain problems for on-site personnel; how to ensure the accuracy of early warning as much as possible while the early warning time is a technical problem requiring an urgent solution. In the present work, a bidirectional long short-term memory network (BiLSTM) model was established according to the temporal variation characteristics of process parameters, and the Whale optimization algorithm (WOA) was used to optimize the model's hyperparameters automatically. The predicted value was further constructed as a Modified Inverted Normal Loss Function (MINLF), and the probability of abnormal fluctuations of process parameters was calculated using the residual time theory. Finally, the WOA-BiLSTM-MINLF process parameter prediction model with inherent risk and trend risk was established, and the fluctuation process of the process parameters was transformed into dynamic risk values. The results show that the prediction model alarms 16 min ahead of distributed control systems (DCS), which can reserve enough time for operators to take safety protection measures in advance and prevent accidents.
PubMed: 38894726
DOI: 10.1016/j.heliyon.2024.e30821 -
Sensors (Basel, Switzerland) Jun 2024The detection of seismic activity precursors as part of an alarm system will provide opportunities for minimization of the social and economic impact caused by...
The detection of seismic activity precursors as part of an alarm system will provide opportunities for minimization of the social and economic impact caused by earthquakes. It has long been envisaged, and a growing body of empirical evidence suggests that the Earth's electromagnetic field could contain precursors to seismic events. The ability to capture and monitor electromagnetic field activity has increased in the past years as more sensors and methodologies emerge. Missions such as Swarm have enabled researchers to access near-continuous observations of electromagnetic activity at second intervals, allowing for more detailed studies on weather and earthquakes. In this paper, we present an approach designed to detect anomalies in electromagnetic field data from Swarm satellites. This works towards developing a continuous and effective monitoring system of seismic activities based on SWARM measurements. We develop an enhanced form of a probabilistic model based on the Martingale theories that allow for testing the null hypothesis to indicate abnormal changes in electromagnetic field activity. We evaluate this enhanced approach in two experiments. Firstly, we perform a quantitative comparison on well-understood and popular benchmark datasets alongside the conventional approach. We find that the enhanced version produces more accurate anomaly detection overall. Secondly, we use three case studies of seismic activity (namely, earthquakes in Mexico, Greece, and Croatia) to assess our approach and the results show that our method can detect anomalous phenomena in the electromagnetic data.
PubMed: 38894445
DOI: 10.3390/s24113654 -
Sensors (Basel, Switzerland) Jun 2024Reliable testing of aviation components depends on the quality and configuration flexibility of measurement systems. In a typical approach to test instrumentation, there...
Reliable testing of aviation components depends on the quality and configuration flexibility of measurement systems. In a typical approach to test instrumentation, there are tens or hundreds of sensors on the test head and test facility, which are connected by wires to measurement cards in control cabinets. The preparation of wiring and the setup of measurement systems are laborious tasks requiring diligence. The use of smart wireless transducers allows for a new approach to test preparation by reducing the number of wires. Moreover, additional functionalities like data processing, alarm-level monitoring, compensation, or self-diagnosis could improve the functionality and accuracy of measurement systems. A combination of low power consumption, wireless communication, and wireless power transfer could speed up the test-rig instrumentation process and bring new test possibilities, e.g., long-term testing of moving or rotating components. This paper presents the design of a wireless smart transducer dedicated for use with sensors typical of aviation laboratories such as thermocouples, RTDs (Resistance Temperature Detectors), strain gauges, and voltage output integrated sensors. The following sections present various design requirements, proposed technical solutions, a study of battery and wireless power supply possibilities, assembly, and test results. All presented tests were carried out in the Components Test Laboratory located at the Łukasiewicz Research Network-Institute of Aviation.
PubMed: 38894377
DOI: 10.3390/s24113585 -
Sensors (Basel, Switzerland) May 2024In this article, the authors focus on the introduction of a hybrid method for risk-based fault detection (FD) using dynamic principal component analysis (DPCA) and...
In this article, the authors focus on the introduction of a hybrid method for risk-based fault detection (FD) using dynamic principal component analysis (DPCA) and failure method and effect analysis (FMEA) based Bayesian networks (BNs). The FD problem has garnered great interest in industrial application, yet methods for integrating process risk into the detection procedure are still scarce. It is, however, critical to assess the risk each possible process fault holds to differentiate between non-safety-critical and safety-critical abnormalities and thus minimize alarm rates. The proposed method utilizes a BN established through FMEA analysis of the supervised process and the results of dynamical principal component analysis to estimate a modified risk priority number () of different process states. The is used parallel to the FD procedure, incorporating the results of both to differentiate between process abnormalities and highlight critical issues. The method is showcased using an industrial benchmark problem as well as the model of a reactor utilized in the emerging liquid organic hydrogen carrier (LOHC) technology.
PubMed: 38894302
DOI: 10.3390/s24113511 -
Diagnostics (Basel, Switzerland) Jun 2024The increased prevalence of obesity worldwide has been implicated in the alarming rise of the incidence of gestational diabetes and preeclampsia, which are both...
IL-6 Polymorphism as a Predisposing Genetic Factor for Gestational Diabetes or Preeclampsia Development in Pregnancy with Obesity in Relation to VEGF and VEGFF Receptor Gene Expression Modalities.
The increased prevalence of obesity worldwide has been implicated in the alarming rise of the incidence of gestational diabetes and preeclampsia, which are both considered threatening conditions for both mother and fetus. We studied gene polymorphisms of the proinflammatory cytokine Interleukin 6 (IL-6) and the gene expression levels of VEGF (vascular endothelial growth factor) and VEGF-R (endothelial growth factor receptor), all known to be involved in pregnancy complications, aiming to identify possible predisposing risk factors in pregnancies with obesity. The G allele of IL-6 was found to correspond with an increased risk for gestational diabetes and preeclampsia occurrence. Furthermore, in obese pregnant mothers with either gestational diabetes or pre-existing type 2 diabetes and those who developed preeclampsia, it was confirmed that gene expression levels of VEGF were reduced while they were increased for VEGF receptors. We conclude that the genetic profile of an obese pregnant woman shares a common background with that of a patient with pre-existing type 2 diabetes mellitus, and therefore predisposes them to complications in pregnancy.
PubMed: 38893732
DOI: 10.3390/diagnostics14111206 -
Diagnostics (Basel, Switzerland) May 2024The study aimed to assess the prevalence of COVID-19 and spp. coinfection across continents. Conducted following PRISMA guidelines, a systematic review utilized PubMed,... (Review)
Review
The study aimed to assess the prevalence of COVID-19 and spp. coinfection across continents. Conducted following PRISMA guidelines, a systematic review utilized PubMed, Embase, SCOPUS, ScienceDirect, and Web of Science databases, searching for literature in English published from December 2019 to December 2022, using specific Health Sciences descriptors. A total of 408 records were identified, but only 50 were eligible, and of these, only 33 were included. Thirty-three references were analyzed to evaluate the correlation between COVID-19 and spp. infections. The tabulated data represented a sample group of 8741 coinfected patients. The findings revealed notable disparities in co-infection rates across continents. In Asia, 23% of individuals were infected with , while in Europe, the proportion of co-infected patients stood at 15%. Strikingly, on the African continent, 43% were found to be infected with , highlighting significant regional variations. Overall, the proportion of co-infections among COVID-positive individuals were determined to be 19%. Particularly concerning was the observation that 1 in 6 ICU coinfections was attributed to , indicating its substantial impact on patient outcomes and healthcare burden. The study underscores the alarming prevalence of co-infection between COVID-19 and , potentially exacerbating the clinical severity of patients and posing challenges to treatment strategies. These findings emphasize the importance of vigilant surveillance and targeted interventions to mitigate the adverse effects of bacterial coinfections in the context of the COVID-19 pandemic.
PubMed: 38893674
DOI: 10.3390/diagnostics14111149 -
Animals : An Open Access Journal From... May 2024Automated activity monitoring (AAM) systems are critical in the dairy industry for detecting estrus and optimizing the timing of artificial insemination (AI), thus...
Automated activity monitoring (AAM) systems are critical in the dairy industry for detecting estrus and optimizing the timing of artificial insemination (AI), thus enhancing pregnancy success rates in cows. This study developed a predictive model to improve pregnancy success by integrating AAM data with cow-specific and environmental factors. Utilizing data from 1,054 cows, this study compared the pregnancy outcomes between two AI timings-8 or 10 h post-AAM alarm. Variables such as age, parity, body condition, locomotion, and vaginal discharge scores, peripartum diseases, the breeding program, the bull used for AI, milk production at the time of AI, and environmental conditions (season, relative humidity, and temperature-humidity index) were considered alongside the AAM data on rumination, activity, and estrus intensity. Six predictive models were assessed to determine their efficacy in predicting pregnancy success: logistic regression, Bagged AdaBoost algorithm, linear discriminant, random forest, support vector machine, and Bagged Classification Tree. Integrating the on-farm data with AAM significantly enhanced the pregnancy prediction accuracy at AI compared to using AAM data alone. The random forest models showed a superior performance, with the highest Kappa statistic and lowest false positive rates. The linear discriminant and logistic regression models demonstrated the best accuracy, minimal false negatives, and the highest area under the curve. These findings suggest that combining on-farm and AAM data can significantly improve reproductive management in the dairy industry.
PubMed: 38891614
DOI: 10.3390/ani14111567