Contribution of artificial intelligence for understanding animal rabies epidemiology in Morocco: What are the perspectives of an innovative and predictive approaches?

Rabies is a major zoonotic disease legally notifiable in Morocco and elsewhere. Given the burden of rabies and its impact on public health, several national control programs have been implemented since 1986, without achieving their expected objectives. The aim of this study was to design a predictive analysis of rabies in Morocco. The expected outcome was the construction of probabilistic diagrams that can guide actions for the integrated control of this disease, involving all stakeholders, in the country. Such modeling is an essential step in operational epidemiology to optimize expenditure of public funds allocated to the integrated strategy for fighting this disease. The methodology employed combined the use of geospatial analysis tools (kriging) and artificial intelligence models (Machine Learning). In order to investigate the link between the risk of rabies within a territorial municipality (commune) and its socio-economic situation, the following data were analyzed: (1) health data: reported animal cases of rabies between 2004 and 2021 and data obtained through the ArcGIS kriging tool (Geospatial data); (2) demographic and socio-economic data. We compared several Machine Learning models. Of these, the "Imbalanced-Xgboost" model associated with kriging yielded the best results. After optimizing this model, we mapped our results for better visualization. The obtained results complement and consolidate previous study in this field with greater accuracy, showing a strong correlation between a commune's socio-economic status, its geographical location and its risk level of rabies. From this, 399 out of the 1546 communes have been identified as high-risk areas, accounting for 25.8% of the total number of communes. Under this risk-based approach, the results of these analyses make it practical to take targeted decisions for rabies prevention and control, as well as canine population control, in a territorial commune according to its risk level. Such an approach allows obvious optimized distribution of financial resources and adaptation of the control actions to be taken. The study highlights also the importance of using innovative technologies to refine epidemiological approaches and fill gaps in field data. Through this study, we hope to contribute to eradication of rabies in Morocco by providing reliable data and practical recommendations for control actions against rabies.

Machine Learning Approaches for Automated Diagnosis of Cardiovascular Diseases: A Review of Electrocardiogram Data Applications.

Cardiovascular diseases (CVDs) have been identified as the leading cause of mortality worldwide. Electrocardiogram (ECG) is a fundamental diagnostic tool used for the diagnosis and detection of these diseases. The new technological tools can help enhance the effectiveness of ECGs. Machine learning (ML) is widely acknowledged as a highly effective approach in the realm of computer-aided diagnostics. This article presents a review of the effectiveness of ML algorithms and deep-learning algorithms in diagnosing, identifying, and classifying CVDs using ECG data. The review identified relevant studies published in the 5 major databases: PubMed, Web of Science (WoS), Scopus, Springer, and IEEE Xplore; between 2021 and 2023, a total of 30 were chosen for the comprehensive quantitative and qualitative. The study demonstrated that different datasets are available online with data related to CVDs. The various ML techniques are employed for the purpose of classification. Based on our investigation, it has been observed that deep learning-based neural network algorithms, such as convolutional neural networks and deep neural networks, have demonstrated superior performance in the detection of entire record data. Furthermore, deep learning showcases its efficacy even when confronted with a scarcity of data. ML approaches utilizing ECG data exhibit a notable proficiency in the realm of diagnosis, hence holding the potential to mitigate the occurrence of disease-related consequences at advanced stages.

Misclassified: identification of zoonotic transition biomarker candidates for influenza A viruses using deep neural network.

Introduction: Zoonotic transition of Influenza A viruses is the cause of epidemics with high rates of morbidity and mortality. Predicting which viral strains are likely to transition from their genetic sequence could help in the prevention and response against these zoonotic strains. We hypothesized that features predictive of viral hosts could be leveraged to identify biomarkers of zoonotic viral transition. Methods: We trained deep learning models to predict viral hosts based on the virus mRNA or protein sequences. Our multi-host dataset contained 848,630 unique nucleotide sequences obtained from the NCBI Influenza Virus and Influenza Research Databases. Each sequence, representing one gene from one viral strain, was classified into one of the three host categories: Avian, Human, and Swine. Trained models were analyzed using various neural network interpretation methods to identify interesting candidates for zoonotic transition biomarkers. Results: Using mRNA sequences as input led to higher prediction accuracies than amino acids, suggesting that the codon sequence contains information relevant to viral hosts that is lost during protein translation. UMAP visualization of the latent space of our classifiers showed that viral sequences clustered according to their host of origin. Interestingly, sequences from pandemic zoonotic viral strains localized at the margins between hosts, while zoonotic sequences incapable of Human-to-Human transmission localized with non-zoonotic viruses from the same host. In addition, host prediction for pandemic zoonotic sequences had low prediction accuracy, which was not the case for the other zoonotic strains. This supports our hypothesis that ambiguously predicted viral sequences bear features associated with cross-species infectivity. Finally, we compared misclassified sequences to well-classified ones to extract interesting candidates for zoonotic transition biomarkers. While features varied significantly between pairs of species and viral genes, several codons were conserved in Swine-to-Human and Avian-to-Human misclassified sequences, and in particular in the NA, HA, and NP genes, suggesting their importance for zoonosis in Humans. Discussion: Analysis of viral sequences using neural network interpretation approaches revealed important genetic differences between zoonotic viruses with pandemic potential, compared to non-zoonotic viral strains or zoonotic viruses incapable of Human-to-Human transmission.