Middle East and North African Health Informatics Association (MENAHIA)

The region of the Middle East and North Africa (MENA) is diverse and retains a superior growth potential. It benefits from a privileged geographical location with big markets, a young and growing educated population, and competitive advantages in several industries. Regardless of their differences, countries face shared concerns, most notably in health. In response to the COVID-19 pandemic, MENA countries enact reforms to create a more robust and inclusive digital health systems to increase growth, development, and integrity. Throughout the coordinated containment and mitigation efforts, most of the countries have integrated digital technologies into the health systems. These procedures include digital government initiatives, the introduction of digital health training courses, live video surgeries and virtual patient monitoring, rural and remote telemedicine programs, and the development of a national electronic health records (EHR) system. Each country took necessary actions to address equity, literacy, and development of resilient health systems. The nine featured countries in this report illustrate the diversity among the MENA region and account for major opportunities and achievements as well as promises and challenges that digital health presents for its populations. © 2022 IMIA and Georg Thieme Verlag KG.

Improving Sentiment Prediction of Textual Tweets Using Feature Fusion and Deep Machine Ensemble Model

Widespread fear and panic has emerged about COVID-19 on social media platforms which are often supported by falsified and altered content. This mass hysteria creates public anxiety due to misinformation, misunderstandings, and ignorance of the impact of COVID-19. To assist health professionals in addressing this epidemic more appropriately at the onset, sentiment analysis can potentially help the authorities for devising appropriate strategies. This study analyzes tweets related to COVID-19 using a machine learning approach and offers a high-accuracy solution. Experiments are performed involving different machine and deep learning models along with various features such as Word2vec, term-frequency, term-frequency document frequency, and feature fusion of both feature-generating approaches. The proposed approach combines the extra tree classifier and convolutional neural network and uses feature fusion to achieve the highest accuracy score of 99%. The proposed approach obtains far better results than existing sentiment analysis approaches. © 2023 by the authors.

COVID-19 dans une population tunisienne : predicteurs de mortalite

Declaration de liens d'interets: Les auteurs declarent ne pas avoir de liens d'interets. Copyright © 2022

Detection of COVID-19 Using Classification of an X-Ray Image Using a Local Binary Pattern and K-Nearest Neighbors

The recently identified coronavirus pneumonia, which was later given the name COVID-19, is a virus that can be fatal and has affected more than 300,000 individuals around the world. Because there is currently no antiviral therapy or vaccine that has been granted approval by the FDA to cure or prevent this sickness, an automatic method for disease identification is required because of the fast global distribution of this exceedingly contagious and lethal virus. A unique machine learning strategy for automatically detecting this ailment was discovered. Machine learning approaches should be applied in essential jobs in infectious illnesses. As a result, our major aim is to use computer vision algorithms to identify COVID-19 without the need for human interaction. This paper suggested using image processing to classify objects and make early detections using X-ray pictures. Features are extracted for this region using a variety of techniques, including (LBP), (HOG), and use K-Nearest Neighbor algorithm (KNN) for classification, with training percentages of 50%, 60%, 70%, 80%, and 90%. Experiments indicated that using the suggested approach to identify X-ray photos of corona patients, it is feasible to diagnose the disease using X-ray images by training the device on the image data set (about 2,400 photos). The results were tested on the average of the samples taken (random 2000 images) each time and the measurement of multiple training ratios (50%, 60%, 70%, 80%, and 90%). The experimental findings revealed remarkable prediction accuracy in all investigated scenarios, ranging from 85% to 99%. © 2022 IEEE.

Forecasting and classification of new cases of COVID 19 before vaccination using decision trees and Gaussian mixture model

Regarding the pandemic taking place in the world from the spread of the Coronavirus pandemic and viral mutations, the need has arisen to analyze the epidemic data in terms of numbers of infected and deaths, different geographical regions, and the dynamics of the spread of the virus. In China, the total number of reported infections is 224,659 on June 11, 2022. In this paper, the Gaussian Mixture Model and the decision tree method were used to classify and predict new cases of the virus. Although we focus mainly on the Chinese case, the model is general and adapted to any context without loss of validity of the qualitative results. The Chi-Squared (χ2) Automatic Interaction Detection (CHAID) was applied in creating the decision tree structure, the data has been classified into five classes, according to the BIC criterion. The best mixture model is the E (Equal variance) with five components. The considered data sets of the world health organization (WHO) were used from January 5, 2020, to 12, November 2021. We provide numerical results based on the Chinese case. © 2022 THE AUTHORS

Covid-19 Patients’ Hospital Occupancy Prediction During the Recent Omicron Wave via some Recurrent Deep Learning Architectures

This paper described a suggested model to predict bed occupancy for Covid-19 patients by country during the rapid spread of the Omicron variant. This model can be used to make decisions on the introduction or alleviation of restrictive measures and on the prediction of oxygen and health human resource requirements. To predict Covid-19 hospital occupancy, we tested some recurrent deep learning architectures. To train the model, we referred to Covid-19 hospital occupancy data from 15 countries whose curves started their regressions during January 2022. The studied period covers the month of December 2021 and the beginning of January 2022, which represents the period of strong contagion of the omicron variant around the world. The evolution sequences of hospital occupancy, vaccination percentages and median ages of populations were used to train our model. The results are very promising which could help to better manage the current pandemic peak. © 2022. by the authors. Licensee Agora University, Oradea, Romania.

Cardiovascular risk factors and clinical outcomes of patients hospitalized with COVID-19 pneumonia in Somalia.

Background: Coronavirus disease-2019 (COVID-19) is a potentially life-threatening illness with no established treatment. Cardiovascular risk factors (CRFs) exacerbate COVID-19 morbidity and mortality. Objective: To determine the prevalence of CRF and clinical outcomes of patients hospitalized with COVID-19 in a tertiary hospital in Somalia. Methods: We reviewed the medical records of patients aged 18 years or older with a real-time polymerase chain reaction (RT-PCR)-confirmed COVID-19 hospitalized at the De Martino Hospital in Mogadishu, Somalia, between March and July 2020. Results: We enrolled 230 participants; 159 (69.1%) males, median age was 56 (41-66) years. In-hospital mortality was 19.6% (n = 45); 77.8% in the intensive care unit (ICU) compared with 22.2%, in the general wards (p < 0.001). Age ⩾ 40 years [odds ratio (OR): 3.6, 95% confidence interval (CI): 1.2-10.6, p = 0.020], chronic heart disease (OR: 9.3, 95% CI: 2.2-38.9, p = 0.002), and diabetes mellitus (OR: 3.2, 95% CI: 1.6-6.2, p < 0.001) were associated with increased odds of mortality. Forty-three (18.7%) participants required ICU admission. Age ⩾ 40 years (OR: 7.5, 95% CI: 1.7-32.1, p = 0.007), diabetes mellitus (OR: 3.2, 95% CI: 1.6-6.3, p < 0.001), and hypertension (OR: 2.5, 95% CI: 1.2-5.2, p = 0.014) were associated with ICU admission. For every additional CRF, the odds of admission into the ICU increased threefold (OR: 2.7, 95% CI: 1.2-5.2, p < 0.001), while the odds of dying increased twofold (OR: 2.1, 95% CI: 1.3-3.2, p < 0.001). Conclusions: We report a very high prevalence of CRF among patients hospitalized with COVID-19 in Somalia. Mortality rates were unacceptably high, particularly among those with advanced age, underlying chronic heart disease, and diabetes.

McGRATH™ video laryngoscope versus Macintosh for intubation in hypoxemic COVID-19 ICU patients

Introduction: Airway management and intubation are challenging in the ICU especially for COVID-19 patients with severe hypoxemia. Although recommended for COVID-19 patients, because of their capacity to reduce transmission to healthcare providers, there is no evidence that video laryngoscopes improve airway management and reduce time for intubation. The purpose of this study was to compare the McGRATH video laryngoscope and the Direct Laryngoscope (DL) in COVID-19 ICU patients with acute respiratory failure. Methods: Forty patients meeting tracheal intubation criteria for respiratory failure were enrolled and equally randomized into 2 groups according to the used device: McGRATH Group and DL group. All patients had pre oxygenation with noninvasive ventilation withFiO2 = 1, Pep and pressure support levels were set to achieve a tidal volume of 6 ml/kg of ideal body weight. Demographic data, difficult intubation criteria were recorded. Our primary outcome was time to intubation defined as the time from the introduction of the blade in patient's mouth until the first efficient breath delivered. Secondary outcomes were the lowest SpO2 recorded during the procedure, the drop in SpO2, the number of attempts, the use of alternative methods for intubation and the experience of the operators. Results: The 2 groups were comparable concerning demographic data, BMI and difficult intubation criteria (p = 0.091). Time to intubation was shorter in the McGRATH group with no significant difference (p = 0.597). The Delta SpO2 and the lowest SpO2 were similar (p = 0.546 and 0.458 respectively). No difference was noticed concerning the number of attempts (p = 0.378), the use of alternative methods (p = 0.276) and the operator's skills (p = 0.076). Conclusions: These results show that the DL is as effective as the recommended McGRATH video laryngoscope for intubation in COVID patients with severe hypoxemia.

Implementation of Virtual Training: The Example of a Faculty of Computer Science during COVID-19 for Sustainable Development in Engineering Education

Research on faculty engagement in computer science and e-learning environments is limited. Students in computer science majors and courses often cite the lack of engagement of their faculty as a reason for their decision to switch majors, drop out or perform poorly. With the shift to e-learning associated with the current global pandemic, reports of faculty engagement across countries and higher education systems converged to indicate a reduced level of interactivity. Using a cross-sectional sample of 39 lecturers and professors from a southern public university in Saudi Arabia, this manuscript documents empirically the low levels of computer science faculty engagement during the 2020 spring semester (March–May). The study found support for the hypotheses linking higher levels of empathetic instruction, an exhibition of exemplary performance traits, utilization of community building strategies and use of storytelling and students’ engagement. The study also found that many faculties need immediate and significant training on making their online instruction more interactive and exciting. Theoretically, the evidence presented confirms the importance of faculty engagement as the main predictor of desirable students’ outcomes across e-learning, as well as computer science learning environments. © 2022 by the authors. Licensee MDPI, Basel, Switzerland.

Évaluation des connaissances du personnel soignant sur l’aérosolthérapie par nébulisation et son impact budgétaire

Introduction L’aérosolthérapie par nébulisation est un moyen thérapeutique connu depuis des siècles qui s’est avéré efficace avec moins d’effets indésirables systémiques. Il est largement prescrit en milieu hospitalier et préconisé dans les affections broncho-pulmonaires et ORL. Méthodes Une enquête a été faite entre janvier et avril 2020 via un auto-questionnaire qui a été préalablement élaboré sur Google Forms destiné au personnel soignant dans un service de pneumologie de l’hôpital Abderrahmene Mami de l’Ariana, comportant neuf questions à choix multiples abordant différent aspect théorique et pratique de la nébulisation en milieu hospitalier. Le but était d’évaluer les connaissances du personnel soignant Résultats Au total, 90 réponses ont été collectées dont 43 (47,8 %) étaient des résidents en pneumologie, 12 (13,3 %) médecins de famille, 7 (7,8 %) infirmiers, 2 (2,2 %) kinésithérapeute, 7(7,8 %) externe en médecine, 2 (2,2 %) Assistant hospitalo-universitaire et un médecin urgentiste. Pour le débit des nébulisations faites à l’oxygène, 5(8,9 %) avaient répondu qu’il doit être maximal 15l flush, 27(48,2 %) ont coché supérieur à 6l/min, seulement 23 (41,1 %) ont coché le débit minimal jusqu’à l’obtention du nuage de particules. La surveillance des patients BPCO en insuffisance respiratoire aiguë hypercapnique était importante pour 48(85,7 %). Le Gaz vecteur est toujours l’oxygène pour 25 (27,8 %). Huit (8,9 %) ont coché que la position couchée est possible pour un meilleur confort. Chez 59 participants, l’interface utilisée devrait être à usage unique. Pour la durée de la séance, 10 (11,1 %) ont répondu supérieure à 30min. L’estimation du budget alloué à l’oxygène médical de l’hôpital était sous-estimée dans 57,4 % des réponses. Conclusion Cette enquête a mis en évidence le manque de connaissance ainsi qu’un mésusage de ce procédé d’administration médicamenteuse. Ceci expose à une déperdition du gaz vecteur le plus utilisé surtout en ce temps de pandémie COVID-19 et à un impact budgétaire non négligeable à l’échelle de l’institution hospitalière. Ainsi, ce travail permet d’insister sur l’importance de la formation continue du personnel médical et paramédical.

Deep Learning and Uniform LBP Histograms for Position Recognition of Elderly People with Privacy Preservation

For the elderly population, falls are a vital health problem especially in the current context of home care for COVID-19 patients. Given the saturation of health structures, patients are quarantined, in order to prevent the spread of the disease. Therefore, it is highly desirable to have a dedicated monitoring system to adequately improve their independent living and significantly reduce assistance costs. A fall event is considered as a specific and brutal change of pose. Thus, human poses should be first identified in order to detect abnormal events. Prompted by the great results achieved by the deep neural networks, we proposed a new architecture for image classification based on local binary pattern (LBP) histograms for feature extraction. These features were then saved, instead of saving the whole image in the series of identified poses. We aimed to preserve privacy, which is highly recommended in health informatics. The novelty of this study lies in the recognition of individuals' positions in video images avoiding the convolution neural networks (CNNs) exorbitant computational cost and Minimizing the number of necessary inputs when learning a recognition model. The obtained numerical results of our approach application are very promising compared to the results of using other complex architectures like the deep CNNs. © 2021. by the authors. All Rights Reserved.

Artificial Intelligence in the Fight Against the COVID-19 Pandemic: Opportunities and Challenges

The world is witnessing unprecedented times as the novel Coronavirus disease (COVID-19) has already conquered and locked down most of the globe. While some indications suggest that the COVID-19 curve is starting to flatten, as of May 2020, we still see constant linear growth in cases and fatalities. Even worse, it is speculated that the situation may further deteriorate with a possible second wave. As governments around the world continue to impose increasingly stringent measures to fight and limit the spread of the pandemic, Artificial Intelligence (AI) tools can play a significant role in the public health surveillance and diagnostics relating to COVID-19. AI is being heavily leveraged in the diagnosis of COVID-19, prediction of its severity for infection, and the discovery of related drugs and vaccines. However, several challenges can impede the exploitation of AI amid the COVID-19 pandemic such as lack of data, privacy, and maturity of AI applications. This chapter discusses the main AI opportunities and challenges in the fight against the COVID-19 pandemic. © 2021, Springer Nature Switzerland AG.

Control the COVID-19 Pandemic: Face Mask Detection Using Transfer Learning

Currently, in the face of the health crisis caused by the Coronavirus COVID-19 which has spread throughout the worldwide. The fight against this pandemic has become an unavoidable reality for many countries. It is now a matter involving many areas of research in the use of new information technologies, particularly those related to artificial intelligence. In this paper, we present a novel contribution to help in the fight against this pandemic. It concerns the detection of people wearing masks because they cannot work or move around as usual without protection against COVID-19. However, there are only a few research studies about face mask detection. In this work, we investigated using different deep Convolutional Neural Networks (CNN) to extract deep features from images of faces. The extracted features are further processed using various machine learning classifiers such as Support Vector Machine (SVM) and K-Nearest Neighbors (K-NN). Were used and examined all different metrics such as accuracy and precision, to compare all model performances. The best classification rate was getting is 97.1%, which was achieved by combining SVM and the MobileNetV2 model. Despite the small dataset used (1376 images), we have obtained very satisfactory results for the detection of masks on the faces. © 2020 IEEE.