Applications and Performance of Machine Learning Algorithms in Emergency Medical Services: A Scoping Review.

OBJECTIVE: The aim of this study was to summarize the literature on the applications of machine learning (ML) and their performance in Emergency Medical Services (EMS). METHODS: Four relevant electronic databases were searched (from inception through January 2024) for all original studies that employed EMS-guided ML algorithms to enhance the clinical and operational performance of EMS. Two reviewers screened the retrieved studies and extracted relevant data from the included studies. The characteristics of included studies, employed ML algorithms, and their performance were quantitively described across primary domains and subdomains. RESULTS: This review included a total of 164 studies published from 2005 through 2024. Of those, 125 were clinical domain focused and 39 were operational. The characteristics of ML algorithms such as sample size, number and type of input features, and performance varied between and within domains and subdomains of applications. Clinical applications of ML algorithms involved triage or diagnosis classification (n = 62), treatment prediction (n = 12), or clinical outcome prediction (n = 50), mainly for out-of-hospital cardiac arrest/OHCA (n = 62), cardiovascular diseases/CVDs (n = 19), and trauma (n = 24). The performance of these ML algorithms varied, with a median area under the receiver operating characteristic curve (AUC) of 85.6%, accuracy of 88.1%, sensitivity of 86.05%, and specificity of 86.5%. Within the operational studies, the operational task of most ML algorithms was ambulance allocation (n = 21), followed by ambulance detection (n = 5), ambulance deployment (n = 5), route optimization (n = 5), and quality assurance (n = 3). The performance of all operational ML algorithms varied and had a median AUC of 96.1%, accuracy of 90.0%, sensitivity of 94.4%, and specificity of 87.7%. Generally, neural network and ensemble algorithms, to some degree, out-performed other ML algorithms. CONCLUSION: Triaging and managing different prehospital medical conditions and augmenting ambulance performance can be improved by ML algorithms. Future reports should focus on a specific clinical condition or operational task to improve the precision of the performance metrics of ML models.

Electrocardiographic interpretation by emergency medical services professionals in Saudi Arabia: A cross sectional study.

BACKGROUND: Management of acute myocardial infarction (AMI) and cardiac arrhythmias in prehospital settings is largely determined by providers of emergency medical services (EMS) who can proficiently interpret the electrocardiography (ECG). The aim of this study was to assess the ECG competency of EMS providers in Saudi Arabia. METHODS: Between Aug and Sep 2022, we invited all EMS providers working for the Saudi Red Crescent Authority in Makkah, Riyadh, and Sharqiyah regions to complete a cross-sectional survey. The survey was used to assess the ability of EMS providers to interpret 12 ECG strips. Characteristics and ECG competency were summarized using descriptive statistics. Differences in ECG competency across paramedics with lower and higher qualifications were assessed. RESULTS: During the study period, 231 participants completed the survey, and all were included. The overall mean age was 33.4, and most participants were male (94.8%). Nearly half of the participants were paramedics with an associate degree and 46.4% were paramedics with higher degrees. The average rate of correct answers to the 12 ECG strips was 43.3% (95% CI: 35.4%, 51.3%). Atrial flutter, ventricular fibrillation, atrial fibrillation, 3rd degree heart block, and ventricular tachycardia were identified by 52.8%, 60.2%, 42.0%, 40.7%, and 49.4% of the participants, respectively. The strip with an AMI was identified by 41.1%, while a pathological Q wave and ventricular extrasystole were identified by 19.1% and 24.7%, respectively. Paramedics with higher qualifications were as 28.0%-61.0% more likely to correctly interpret the 12 ECG strips compared to those with an associate degree (p-value across all variables was ≤ 0.001). CONCLUSION: While the majority of participants in our region were unable to correctly answer the 12 ECG questionnaire, paramedics with higher qualifications were. Our study indicates that there is a need for evidenced-based ECG curricula targeting different levels of EMS professionals.

The Perceptions and Expectations of the Clinical Learning Environment in Saudi Arabia: a Multidisciplinary Study.

Background: The clinical learning environment is a fundamental component of healthcare education. In this setting, students can develop the skills and knowledge necessary to become competent and efficient healthcare practitioners. Due to the importance of clinically based education, it is crucial to have a valid and reliable tool to enable its evaluation. Objective: The aim of this study was to use the Clinical Learning Environment Inventory (CLEI) to examine the perceptions of Saudi undergraduate healthcare students regarding their actual and preferred clinical learning environment and explore the differences between the two viewpoints. Methods: A cross-sectional survey design was utilised with a cohort of Saudi undergraduate healthcare students. Data regarding perceptions of the 'actual' and 'preferred' clinical learning environments were gathered with the Clinical Learning Environment Inventory tool. Results: A total of 194 students participated and nine healthcare disciplines were represented. The highest mean score for both 'actual' and 'preferred' clinical learning environment was for the subscale Task Orientation. Significant differences between 'actual' and 'preferred' environments were demonstrated for Innovation and Individualization, with both subscales scoring higher for the 'preferred' environment. All five subscales-Individualization, Innovation, Involvement, Personalization, and Task Orientation-appear to be important aspects contributing to student satisfaction with their clinical learning environment. Conclusion: Saudi healthcare students demonstrate a preference for a clinical learning environment with the utilization of new and interesting experiences, as well as recognition and accommodation of student individuality. Additionally, student satisfaction appears to be multifactorial in origin. Therefore, there may be many avenues available to enhance the clinical experiences of healthcare students, which is vitally important for the optimization of clinical learning opportunities.

Performance of Manufactured and Recycled Steel Fibres in Restraining Concrete Plastic Shrinkage Cracks.

Early-age plastic shrinkage cracks can reduce the durability of concrete slabs by creating direct paths for the ingress of aggressive agents and thus accelerating degradation due to environmental attack, in particular, in hot and windy environments. The elimination of such cracks is essential for durable and sustainable concrete structures. This paper parametrically investigates the effect of manufactured steel fibres (MSF) and recycled tyre steel fibres (RTSF) on restraining plastic shrinkage and micro cracks at different dosages (10, 20, and 30 kg/m3). The plastic shrinkage tests were carried out in a specially designed chamber, according to ASTM C1579. Various environmental conditions are examined, and their impact on compressive strength and crack potential is assessed. A digital image analysis technique is used to measure length, width, and the area of the crack on the exposed surface to gain additional insights into crack behaviour. The results show a slight early-age (one-day) increase in compressive strength for the concrete exposed to the various environmental conditions, mostly as a result of higher temperatures. Through the use of the crack reduction ratio (CRR), both RTSF and MSF are shown to be successful in controlling plastic shrinkage and micro cracks, with the RTSF being superior due to the fact that they are better distributed in the concrete volume. The addition of 30 kg/m3 of RTSF was effective in preventing crack development in most environments or restraining cracks in extremely harsh environments. The adoption of these results will lead to more sustainable concrete slabs in the harsher environmental conditions created by climate change.

The Effect of Harsh Environmental Conditions on Concrete Plastic Shrinkage Cracks: Case Study Saudi Arabia.

Due to climate change and population expansion, concrete structures are progressively being subjected to more extreme environments. As the environment affects plastic shrinkage directly, there is a need to understand the effect of environmental changes on plastic shrinkage cracking. This paper examines the plastic shrinkage crack development parametrically at low, mid, and high drying environmental conditions, corresponding to different environments in three Saudi cities. The effects of water-cement ratios and quantities of recycled tire steel fibers (RTSF) in concrete are also investigated. The different environmental conditions for the plastic shrinkage tests were simulated in a specially designed chamber as per ASTM C1579, 2006. A digital image processing (DIP) technique was used to monitor crack initiation and development. Through the use of the crack reduction ratio (CRR), it was found that 30 kg/m3 of RTSF can control plastic shrinkage cracks at low and mid conditions. For the more extreme (high) conditions, the use of 40 kg/m3 of RTSF fiber was sufficient to completely eliminate surface plastic shrinkage cracks. This work can help develop more sustainable concrete structures in a wider set of environmental conditions and help mitigate the impact of climate change on concrete infrastructure.

Experimental investigation of air cooler using local palm tree waste.

Middle Saudi Arabia has weather conditions where the temperature is high in summer with low humidity. Conventional air conditioning systems operated by a vapor compression cycle are not economical because of the high electrical power consumption. Therefore, evaporative cooling through evaporative coolers is one of the best and most economical solutions. The present study experimentally investigates the factors affecting the performance of evaporative coolers. Pad materials and airflow rate are the main variables to investigate the evaporative cooler's performance in terms of saturation effectiveness, pressure drop across the pads, and coefficient of performance (COP). Pads material are the local palm tree "Nakheel" waste that are leaflet, leaf base, bulb, and roots. The maximum COP of the cooling system in the case of bulb pad material is 80% more than that of leaflet pad material. The saturation effectiveness of the bulb pad was a maximum which is 61.93% at an airflow of 2.25 m/s, which is more than two times that of the saturation effectiveness of the leaflet pad. The pressure drop across the bulb pad is almost 2.5 times to 9.5 times than that of leaflet pad. Results show that bulb pad performance best, whereas the leaflet pad material has the lowest performance in terms of pressure drop, saturation effectiveness, and COP.

Saudi Healthcare Students' Perceptions and Beliefs About Immunizations: a Descriptive, Cross-Sectional Study.

Background: Immunization has been one of the most successful public health measures ever undertaken. However, a degree of hesitancy about vaccine use still exists. Healthcare professionals are in a unique position to provide advice and education to the public and may influence the decision to undergo immunization. Objective: The aim of this study was to explore perceptions and beliefs regarding immunizations and immunization-preventable diseases. Methods: A descriptive cross-sectional study was undertaken at the Imam Abdulrahman bin Faisal University, located in Dammam, Saudi Arabia. In the Kingdom of Saudi Arabia, a survey of 564 Saudi undergraduate healthcare students was conducted. 77.8% of participants replied (439). Information was collected regarding perceptions of; severity of immunization-preventable diseases, contracting these diseases, immunization safety, and immunization beliefs. The statistical analysis was performed using the Statistical Package for the Social Sciences (IBM SPSS v25). Non-parametric analyses were utilized. Descriptive data were generated as appropriate, including frequencies, median, and inter-quartile range. Statistical relationships of demographic variables were explored using Kruskal Wallis H-Test and Spearman's Rank-Order Correlation. A p-value < 0.05 was considered statistically significant. Results: Meningitis was perceived as the most severe disease and COVID-19 as having the highest likelihood of infection. Concern regarding vaccine side effects was most evident for the COVID-19 vaccine. Student year level and profession resulted in statistically significant differences for all three assessed perceptions. Substantial differences were also identified regarding views on immunization belief statements. Conclusion: This study identified considerable heterogeneity in Saudi healthcare students' perceptions and beliefs regarding immunization-preventable diseases and vaccination. Further education is required to produce well-informed and confident healthcare professionals around these issues.

Haematological and radiological-based prognostic markers of COVID-19.

BACKGROUND: Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) has emerged in 2019 and caused a global pandemic in 2020, manifesting in the coronavirus disease 2019 (COVID-19). The majority of patients exhibit a mild form of the disease with no major complications; however, moderate to severe and fatal cases are of public health concerns. Predicting the potential prognosis of COVID-19 could assist healthcare workers in managing cases and controlling the pandemic in an effective way. Therefore, the objectives of the study were to search for biomarkers associated with COVID-19 mortality and predictors of the overall survival (OS). METHODS: Here, clinical data of 6026 adult COVID-19 patients admitted to two large centers in Saudi Arabia (Riyadh and Hafar Al-Batin cities) between April and June 2020 were retrospectively analysed. RESULTS: More than 23% of the study subjects with available data have died, enabling the prediction of mortality in our cohort. Markers that were significantly associated with mortality in this study were older age, increased d-dimer in the blood, higher counts of WBCs, higher percentage of neutrophil, and a higher chest X-ray (CXR) score. The CXR scores were also positively associated with age, d-dimer, WBC count, and percentage of neutrophil. This supports the utility of CXR scores in the absence of blood testing. Predicting mortality based on Ct values of RT-PCR was not successful, necessitating a more quantitative RT-PCR to determine virus quantity in samples. Our work has also identified age, d-dimer concentration, leukocyte parameters and CXR score to be prognostic markers of the OS of COVID-19 patients. CONCLUSION: Overall, this retrospective study on hospitalised cohort of COVID-19 patients presents that age, haematological, and radiological data at the time of diagnosis are of value and could be used to guide better clinical management of COVID-19 patients.

OpenSHS: Open Smart Home Simulator.

This paper develops a new hybrid, open-source, cross-platform 3D smart home simulator, OpenSHS, for dataset generation. OpenSHS offers an opportunity for researchers in the field of the Internet of Things (IoT) and machine learning to test and evaluate their models. Following a hybrid approach, OpenSHS combines advantages from both interactive and model-based approaches. This approach reduces the time and efforts required to generate simulated smart home datasets. We have designed a replication algorithm for extending and expanding a dataset. A small sample dataset produced, by OpenSHS, can be extended without affecting the logical order of the events. The replication provides a solution for generating large representative smart home datasets. We have built an extensible library of smart devices that facilitates the simulation of current and future smart home environments. Our tool divides the dataset generation process into three distinct phases: first design: the researcher designs the initial virtual environment by building the home, importing smart devices and creating contexts; second, simulation: the participant simulates his/her context-specific events; and third, aggregation: the researcher applies the replication algorithm to generate the final dataset. We conducted a study to assess the ease of use of our tool on the System Usability Scale (SUS).

Maxillary osteosarcoma: two case reports and literature review.

Osteosarcoma is a rare malignant tumor of the maxilla. Although clear surgical margin is the only predictor for the prognosis of the disease, neoadjuvent chemotherapy showed a reasonable effect on the tumor with variable degree of necrosis. In this article, we report two cases where neoadjuvent chemotherapy was used with review of the literature.