Implementation of a Sequence-to-Sequence Stacked Sparse Long Short-Term Memory Autoencoder for Anomaly Detection on Multivariate Timeseries Data of Industrial Blower Ball Bearing Units

The advent of Industry 4.0 introduced new ways for businesses to evolve by implementing maintenance policies leading to advancements in terms of productivity, efficiency, and financial performance. In line with the growing emphasis on sustainability, industries implement predictive techniques based on Artificial Intelligence for the purpose of mitigating machine and equipment failures by predicting anomalies during their production process. In this work, a new dataset that was made publicly available, collected from an industrial blower, is presented, analyzed and modeled using a Sequence-to-Sequence Stacked Sparse Long Short-Term Memory Autoencoder. Specifically the right and left mounted ball bearing units were measured during several months of normal operational condition as well as during an encumbered operational state. An anomaly detection model was developed for the purpose of analyzing the operational behavior of the two bearing units. A stacked sparse Long Short-Term Memory Autoencoder was successfully trained on the data obtained from the left unit under normal operating conditions, learning the underlying patterns and statistical connections of the data. The model was evaluated by means of the Mean Squared Error using data from the unit's encumbered state, as well as using data collected from the right unit. The model performed satisfactorily throughout its evaluation on all collected datasets. Also, the model proved its capability for generalization along with adaptability on assessing the behavior of equipment similar to the one it was trained on.

Nicotinic cholinergic system and COVID-19: In silico identification of interactions between α7 nicotinic acetylcholine receptor and the cryptic epitopes of SARS-Co-V and SARS-CoV-2 Spike glycoproteins

SARS-CoV-2 is the coronavirus that originated in Wuhan in December 2019 and has spread globally. Studies have shown that smokers are less likely to be diagnosed with or be hospitalized for COVID-19 but, once hospitalized, have higher odds for an adverse outcome. We have previously presented the potential interaction between SARS-CoV-2 Spike glycoprotein and nicotinic acetylcholine receptors (nAChRs), due to a "toxin-like" epitope on the Spike glycoprotein, with homology to a sequence of a snake venom toxin. This epitope coincides with the well-described cryptic epitope for the human anti-SARS-CoV antibody CR3022. In this study, we present the molecular complexes of both SARS-CoV and SARS-CoV-2 Spike glycoproteins, at their open or closed conformations, with the model of the human α7 nAChR. We found that all studied protein complexes' interface involves a large part of the "toxin-like" sequences of SARS-CoV and SARS-CoV-2 Spike glycoproteins and toxin binding site of human α7 nAChR. Our findings provide further support to the hypothesis about the protective role of nicotine and other cholinergic agonists. The potential therapeutic role of CR3022 and other similar monoclonal antibodies with increased affinity for SARS-CoV-2 Spike glycoprotein against the clinical effects originating from the dysregulated cholinergic pathway should be further explored.

Nicotinic cholinergic system and COVID-19: In silico evaluation of nicotinic acetylcholine receptor agonists as potential therapeutic interventions

SARS-CoV-2 infection was announced as a pandemic in March 2020. Since then, several scientists have focused on the low prevalence of smokers among hospitalized COVID-19 patients. These findings led to our hypothesis that the Nicotinic Cholinergic System (NCS) plays a crucial role in the manifestation of COVID-19 and its severe symptoms. Molecular modeling revealed that the SARS-CoV-2 Spike glycoprotein might bind to nicotinic acetylcholine receptors (nAChRs) through a cryptic epitope homologous to snake toxins, substrates well documented and known for their affinity to the nAChRs. This binding model could provide logical explanations for the acute inflammatory disorder in patients with COVID-19, which may be linked to severe dysregulation of NCS. In this study, we present a series of complexes with cholinergic agonists that can potentially prevent SARS-CoV-2 Spike glycoprotein from binding to nAChRs, avoiding dysregulation of the NCS and moderating the symptoms and clinical manifestations of COVID-19. If our hypothesis is verified by in vitro and in vivo studies, repurposing agents currently approved for smoking cessation and neurological conditions could provide the scientific community with a therapeutic option in severe COVID-19.

Person Identification from the EEG using Nonlinear Signal Classification

Objectives: This paper focusses on the person identification problem based on features extracted from the ElectroEncephaloGram (EEG). A bilinear rather than a purely linear model is fitted on the EEG signal, prompted by the existence of non-linear components in the EEG signal – a conjecture already investigated in previous research works. The novelty of the present work lies in the comparison between the linear and the bilinear results, obtained from real field EEG data, aiming towards identification of healthy subjects rather than classification of pathological cases for diagnosis.

Methods: The EEG signal of a, in principle, healthy individual is processed via (non)linear (AR, bilinear) methods and classified by an artificial neural network classifier.

Results: Experiments performed on real field data show that utilization of the bilinear model parameters as features improves correct classification scores at the cost of increased complexity and computations. Results are seen to be statistically significant at the 99.5% level of significance, via the χ2 test for contingency.

Conclusions: The results obtained in the present study further corroborate existing research, which shows evidence that the EEG carries individual-specific information, and that it can be successfully exploited for purposes of person identification and authentication.

Combined effects of age and maturation on maximum isometric leg press strength in young basketball players

AIM

Muscular strength of the leg extensor muscles in children can be affected by several factors such as age, sexual maturation, body mass and training status of the subjects. The purpose of the study was to examine maximal isometric strength characteristics of young male basketball players taking into consideration the combined effects of chronological age and sexual maturation.

METHODS

One hundred and twenty male basketball players, aged from 12 to 17 years divided into 6 equivalent age subgroups performed maximum bilateral isometric leg press efforts. The parameters analysed were the maximal voluntary isometric force (MVC), relative strength (MVC/body mass and MVC/fat free mass), starting strength (F50: force exerted during the first 50 ms of the contraction) and speed strength index (the ratio of maximal force to time to attain maximal force).

RESULTS

The results indicated that in almost all absolute force parameters, the 12-and 13-year olds demonstrated significantly (p<0.05) lower values compared with the 15(-1)6-and 17-years old groups. Age differences were also significant (p<0.05) when the effects of sexual maturation were taken into consideration in the statistical analysis but they were reduced when strength was adjusted for body mass. Finally, no significant differences for strength per unit of fat free mass were found (p>0.05).

CONCLUSIONS

Maximum absolute strength of basketball players is significantly increased from 12 to 17 years and as sexual maturation stage increases. It also appears that body mass and fat free mass should be taken into consideration when examining age effects on strength in basketball players.

Comparison between auto-cross-correlation coefficients and coherence methods applied to the EEG for diagnostic purposes

BACKGROUND

Power-spectrum and coherence analysis of the EEG has often been applied in the study of various forms of brain dysfunction. The examination of differences in intra-hemispheric coherence and a novel method based on auto-cross-correlation algorithms between the left and right hemispheres of the same EEG are investigated in this paper.

MATERIAL/METHODS

Both coherence and the proposed method are computed from the same EEG segment across the EEG spectrum (0-64 Hz). In this experiment, five EEGs of different patients with severe brain damage were used. The novelty of this study lies in the fact that the autocorrelation coefficients are extracted in a particular spectrum, in contrast to the traditional coherence method, which is based on a measure of the square of the linear association between the two EEG signals from the entire EEG spectrum.

RESULTS

The results of the application of coherence and the proposed method showed that particular pairs of symmetrical cortex channels differed dramatically in specific damaged areas.

CONCLUSIONS

Statistical analysis of the above results showed that the proposed method is more accurate than coherence because it could recognize the cases of the patients to a significant statistical degree, whereas the coherence method showed weakness in recognizing them.

Person identification from the EEG using nonlinear signal classification

OBJECTIVES

This paper focusses on the person identification problem based on features extracted from the ElectroEncephaloGram (EEG). A bilinear rather than a purely linear model is fitted on the EEG signal, prompted by the existence of non-linear components in the EEG signal--a conjecture already investigated in previous research works. The novelty of the present work lies in the comparison between the linear and the bilinear results, obtained from real field EEG data, aiming towards identification of healthy subjects rather than classification of pathological cases for diagnosis.

METHODS

The EEG signal of a, in principle, healthy individual is processed via (non)linear (AR, bilinear) methods and classified by an artificial neural network classifier.

RESULTS

Experiments performed on real field data show that utilization of the bilinear model parameters as features improves correct classification scores at the cost of increased complexity and computations. Results are seen to be statistically significant at the 99.5% level of significance, via the chi 2 test for contingency.

CONCLUSIONS

The results obtained in the present study further corroborate existing research, which shows evidence that the EEG carries individual-specific information, and that it can be successfully exploited for purposes of person identification and authentication.

On the use of EEG features towards person identification via neural networks

Person identification based on spectral information extracted from the EEG is addressed in this work a problem that has not yet been seen in a signal processing framework. Spectral features are extracted non-parametrically from real EEG data recorded from healthy individuals. Neural network classification is applied on these features using a Learning Vector Quantizer in an attempt to experimentally investigate the connection between a person's EEG and genetically specific information. The proposed method, compared with previously proposed methods, has yielded encouraging correct classification scores in the range of 80% to 100% (case-dependent). These results are in agreement with previous research showing evidence that the EEG carries genetic information.

Participative HIS development: an approach and a case study

User participation in HIS development is considered essential for achieving systems implementation success. Realizing a participative HIS development, where users are full members of the development team, requires not only choosing an appropriate methodology but also organizing the participation process in a way that is tailored to the particular situation in order to achieve the desired results. A general approach to this problem is presented in this paper. An application of the approach to the particular context of a Greek hospital is described.