Middle East Journal of Applied Science & Technology
Volume 8 Issue 1 January-March 2025
Research Article
Mst. Sumaiya Akter Mim, Md. Julker Nayeem, Sohel Rana & Md. Rabiul Islam
Abstract: The application of classification approaches utilizing multi-variable with machine learning methods holds immense implications, particularly in the realm of healthcare and disease prediction. Accurate classification of medical conditions, such as hepatitis, is critical for early diagnosis and timely intervention. In order to identify people based on important hepatitis-related characteristics, this study applies advanced machine learning with statistical techniques. It also examines a real dataset in order to create a reliable early detection predictive model. Through this model, we aspire to raise awareness and guide affected individuals toward timely treatment. The paper focuses on comprehensive data preprocessing, including outlier removal, handling class imbalance problem, missing values and extract highly correlated features in order to improve model performance. In our research paper, we have applied mean/mode imputation technique to deal with missing values. Furthermore, we have used z score approach to detect and remove outliers from out dataset and handle class imbalance problem by using oversampling technique. To identify features that are highly correlated, we have used the embedded feature selection approach in our paper. Classic machine learning algorithms, notably K-Nearest Neighbors (KNN), Naive Bayes (NB) and Random Forest (RF) have employed to predict either a person is affected by hepatitis disease or not. To assess the efficacy of our model, we have utilized the 10-fold cross validation procedure. At 97.44%, we have the highest classification accuracy from RF, with Precession, Recall, F1 score and ROC values of, respectively, 0.99, 0.96, 0.97 and 1.00.
Review Article
Ravi Yadav
Abstract: Background: The reliability of neuromuscular monitoring is essential in optimizing anesthesia outcomes. Quantitative Train-of-Four (TOF) monitoring has emerged as a tool for ensuring precise neuromuscular blockade management. Objectives: This study aims to compare the efficacy and safety of neuromuscular blockade reversal guided by quantitative TOF monitoring with reversal conducted without such monitoring. Methods: A literature-based analysis was conducted, evaluating existing data and trends regarding the use of quantitative TOF monitoring in clinical practice. Performance metrics, such as precision in blockade reversal, efficiency, and associated outcomes, were assessed to highlight the advantages and limitations of each approach. Results: Findings demonstrated that quantitative TOF monitoring facilitates a more consistent and reliable reversal of neuromuscular blockade. It reduces the risk of residual paralysis and enhances patient safety. In contrast, practices without TOF monitoring showed variability in outcomes, reflecting potential gaps in blockade management. Conclusions: Incorporating quantitative TOF monitoring into anesthesia practices improves the accuracy of neuromuscular blockade reversal, emphasizing its importance in modern anesthetic care. The study underscores the need for broader adoption of such technologies to standardize outcomes and enhance procedural safety.
Review Article
Review on renewable energy-based KY boost converter and seven level-inverter systems
Gopika B.S. & Rajeshwari
Abstract: Static VAR compensators (SVC), power rectifiers, and thyristor converters are examples of power electronics components that significantly contribute to harmonics in a range of applications. The use of power electronic converters, especially DC/AC PWM inverters, has been expanding in the industry due to the advantages they provide, including reduced energy consumption, improved system efficiency, higher-quality products, simplicity of maintenance, and more. One of the most basic and well-known topologies for multilevel inverters is cascaded H-Bridge (CHB) MLI. A KY boost converter with seven-level inverters is suggested in this study. A Matlab simulation is used to evaluate the suggested work. According to the simulation findings, the output voltage climbed from 121V to 155V, the motor speed increased from 940 rpm to 1050 rpm, the motor torque increased from 0.92 N/m to 1.80 N/m, and the output current THD decreased from 31.2% to 19.01% when a KY boost converter with a seven-level inverter was used. According to the simulation results, the conventional boost converter with a five-level inverter system performs worse than the suggested KY boost converter with seven-level inverters.
Research Article
A Machine Learning Approach for Identifying Gender Based on Bengali Vocal Cues
Al Arman Ovi, Iffath Tanjim Moon & Md. Julker Nayeem
Abstract: In this research, an advanced system getting-to-know model is being carried out to properly examine and analyze multiple Bangladeshi vocal cues to diagnose gender correctly. This study is implemented in quite a few domain names, including customer service, where determining the gender of a caller can provide extra-personalized interactions. Additionally, voice-activated assistants use it to customize responses and beautify the consumer experience. To beautify provider shipping and offer demographic insights and gender recognition the use of voice evaluation is also utilized in safety systems, transcription services and sociolinguistic research. Voice is stricken by environmental way of life elements such as smoking, acid reflux sickness, air pollutants, warm weather weight reduction, city air pollutants and the horrible effect of energy on Bangladeshi fitness. In this study, we obtained data on voices from male and female participants living in Bangladesh. To provide a consistent and convenient method of research, we first converted each recording to Waveform Audio File Format (WAV). We then extracted the most significant voices from these WAV recordings and converted them to statistical data. Then, we preprocessed this statistical information to put it together for in-depth analysis. After preprocessing, we used report visualization to understand the traits and patterns observed within the voice recordings. This holistic approach enables a complete assessment of voice data to gain the goals of our study. So, the idea of this venture is to train a machine learning model with updated information processing techniques which can be expected should be gender according to voice notes. We goal to establish a dependable gender identification set of rules based on modern-day findings and big-scale facts.
Research Article
Kieu Quoc Hoan
Abstract: This study investigates the determinants of cooperation intensity among local governments in promoting Foreign Direct Investment (FDI) within Vietnam's Northern Key Economic Zone (KEZ), which is emerging as a critical hub for FDI but faces significant disparities in FDI distribution across its provinces. The study applies the Institutional Collective Action (ICA) Framework. It is complemented by the Theory of Planned Behavior (TPB) and Risk Aversion Theory to form a comprehensive understanding of how gaps in perception and expectation of benefits, costs, and risks shape cooperative decisions. The findings indicate that while the gap between perception and expectation of benefits and transaction costs does not significantly impact the propensity to intensity of the cooperation, the gap of risks of collaboration is a key determinant. This research emphasizes the need for coordinated strategies to mitigate risks and enhance trust among local governments, offering actionable insights for policymakers to foster more effective intergovernmental collaboration, ultimately optimizing resource allocation and boosting FDI attraction.
Research Article
Oxovanadium Complex of Mebendazole: Synthesis, Spectroscopic and Physico-chemical Characterization
Olufunso O. Abosede & Stephen Princewill Osi
Abstract: The greener and cost-effective synthesis of an oxovanadium complex of mebendazole is described. The synthesis involves the use of distilled water as a solvent and refluxing, providing environmentally friendly advantages, simple work-up procedures, and a short reaction time with excellent yield. Characterization included UV-Vis and FTIR spectroscopy. The UV-Vis spectra showed intra-ligand transitions at 242 and 269 nm and charge transfer bands at 310 and 380 nm, indicating the formation of oxovanadium complex of mebendazole. The FTIR spectra confirmed coordination via the nitrogen and oxygen atoms of the ligand, evidenced by shifts in C-H, C=C, and C-O stretching bands, the appearance of a new V=O band at 977.94 cm⁻¹, and an OV-N bending band at 497.65 cm⁻¹. The metal complex with the formula [VO(L)]SO₄ was deduced, where L is mebendazole.
Review Article
Smarter Surgeries: AI-Driven Innovations in the OT and Anaesthesia Management
Ms. Anjali, Mr. Rajdeep Thidwar & Ms. Piyush Yadav
Abstract: The field of medicine, particularly surgery and anaesthesia, has witnessed remarkable advancements over the years. This article explores the evolution of the operation theatre and anaesthesia, emphasizing the transformative impact of technology and Artificial Intelligence (AI) on these critical medical domains. In the realm of surgery, technological progress has led to the expansion of minimally invasive procedures, such as laparoscopic surgery and robotic-assisted surgeries. These innovations contribute to reduced patient discomfort, faster recovery, and improved precision. 3D printing technology further enhances surgical preparation by allowing surgeons to create accurate organ replicas for preoperative planning. Anaesthesia practices have also evolved, with the increased adoption of local anaesthesia for awake surgeries, epidural anaesthesia for childbirth, and regional anaesthesia to minimize opioid use in orthopedic procedures. These advancements aim to enhance patient comfort and reduce complications associated with traditional anaesthesia methods. The integration of AI into operation theatres and anaesthesia brings real-time data analysis capabilities, enabling continuous monitoring of patient vital signs during surgery. AI-assisted robotic surgery systems enhance surgical precision, providing surgeons with detailed 3D visualizations. In the field of anaesthesia, AI analyzes physiological data to predict and minimize complications, adjust anaesthesia administration, and estimate post-surgery opioid requirements, mitigating the risk of addiction. While AI contributes significantly to surgical advancements, it is emphasized that technology should complement, not replace, the expertise of medical professionals. The collaborative efforts of skilled surgeons, anaesthetists, and innovative technology promise a future where patients can benefit from efficient, safe, and remarkable surgical experiences.
Review Article
Mr. Ravi Yadav, Ms. Neha Singla & Ms. Shabnam Gurmeetchand
Abstract: Introduction: Augmented Reality (AR) is revolutionizing surgical practice through improved visualization, accuracy, and training. This paper examines approaches to integrating AR and its potential future applications. Methods: Literature review and expert critique were used to determine AR effectiveness, challenges, and how to implement it in surgery. Results: Evidence suggests that AR enhances surgical precision and training. The adoption is impeded by costs, technical restrictions, and specialist training requirements. Future development will involve the use of AI and robotics to develop more advanced capabilities. Conclusion: Although useful, AR's application in surgery demands overcoming financial and technical constraints. Strategic planning and additional research must be conducted to achieve its greatest potential in practice.
Research Article
Subsurface Reservoir Fluids Mapping of Mature Oil Fields Using Dynamic Modeling
Ghareb Hamada, Abdel Sattar Dahab & Mahmoud Salim
Abstract: Development of mature oil reservoir blocks strategy in optimizing hydrocarbon production from mature oil fields. Using dynamic modeling techniques together with subsurface interpretations that were based on sharply adjacent well data, the study has presented new development well placements. Production data studies, pressure surveys, and detailed subsurface studies will be integrated in improving the understanding of the reservoir behavior to predict reservoir performance. The key findings indicate high increments in oil production on account of the implementation of new development wells. The current research provides not only empirical evidence supporting the efficacy of strategic drilling but also lends a structured workflow applicable for similar mature oil fields, thus providing practical means for increasing production and betterment in field life. Opportunities that deliver high value yet have limited associated risks and costs are typically characterized in general by short payout periods and reinvestment of savings. Unlock residual potential from low-productivity or shut-in wells, maximizing asset value, and information driving decisions on optimizing is an asset development strategy. These strategies can only be effectively implemented if inter-disciplinary teams work together to ensure that data is comprehensively analyzed and all solutions are formulated in line with these analyses. This work discusses the efficiency of developing no oil reservoir blocks strategy in optimizing hydrocarbon production from mature oil fields. Using dynamic modeling techniques together with subsurface interpretations that were based on sharply adjacent well data, the study has presented new development well placements. The study of the A/R field has laid the foundation for long-term optimization, integrating production data, pressure surveys, and geological studies into a comprehensive reservoir management strategy. The interdisciplinary approach ensured well-informed and targeted decisions, resulting in enhanced hydrocarbon recovery and improved field performance. By leveraging data-driven strategies and advanced technology, the field's sustainability and efficiency have been significantly improved, providing a strong foundation for future optimization efforts.
Research Article
Effiong, E.C. & Asionye, E.I.
Abstract: Roof-harvested rainwater (RHRW) has remained one of the oldest sources of alternative potable water in most rural communities in the world. In this study, two major samples were obtained from the cafeteria/canteen and female hostel at the Hezekiah University, Umudi. Total heterotrophic bacterial count, total coliform count and Staphylococcal count. The coliform composition was determined using the Most Probable Number (MPN) 3-tube technique. Biochemical tests and colonial morphology were employed in the identification of the bacterial isolates associated with the RHRW samples. Total heterotrophic bacterial and coliform counts for the samples obtained from the cafeteria/canteen was 5.0 Log10CFU/mL and 3.0 Log10 CFU/mL while the coliform content was 10 and 36 CFU/100mL was observed for the samples obtained from the female hostel and cafeteria respectively. The physicochemical composition of the RHRW from the cafeteria was observed to have a pH of 6.9, electrical conductivity of 110.56 (µS/cm), total dissolved solids was 11.97, total hardness was 13.49 mg/L while total acidity was 15 mg/L. The bacterial flora identified from the RHRW were E. coli, Enterococcus sp., Klebsiella sp. and Pseudomonas sp., while E. coli was identified to be the most frequent. There is a need for the University management to sensitize both staff and students on the potential dangers of using roof-harvested rainwater for domestic purposes. Non-governmental organizations must sensitize rural communities to the need for the treatment of rainwater before usage.
Research Article
Application of Support Vector Machine for Effective Prediction of Election for Sentiment Analysis
Asoshi Paul Anule & Chukwudi Jennifer Ifeoma
Abstract: This study proposes the use of machine learning models, namely Support Vector Machine (SVM), for effective sentiment analysis on a dataset from the Kaggle repository. Considering the Tinubu 2023 election dataset, it can be seen that SVM having been fed with the cleansed dataset feature obtained an accuracy score of 93.2%, considering the result of each algorithm on the 2023 Nigerian election datasets. The study investigates data preprocessing techniques, feature selection, and correlation metrics to optimize the sentiment detection process. Results show that the SVM model achieves the highest accuracy, making it a potential tool for political analysis, business marketing, and public policy implementation. However, future research may explore deep learning techniques and data balancing strategies to enhance the models' performance further.
Research Article
Effiong, E.C. & Asionye, E.I.
Abstract: The quality of most roof-harvested rainwater obtained from most rural communities with poor access to potable water has become a major pitfall in the longevity of the populace. Five different clean catch of the rainwater were collected from three different tertiary institutions namely Hezekiah University, Umudi, Nkwerre, Kingsley Ozumba, Mbadiwe University, Ideato Imo State and College of Health Technology, Amaigbo, Imo State. In situ parameters were measured and the other batch was aseptically transported to the laboratory. Quality indices were determined using standard reagents and methodologies. The pH of the rainwater samples obtained from roofs at the College of Health Technology was 6.4 and 6.6 for the hostel and canteen respectively. The electrical conductivity of samples obtained from the Hezekiah University hostel section was 69.1 µS/cm and 110.56 µS/cm for the canteen. The rainwater samples obtained from Hezekiah University had a faecal composition of 3.0 MPN/100mL and 7.4 MPN/100mL for the hostel and canteen section respectively while the samples obtained from K.O. Mbadiwe University had a coliform concentration of 6.1 and 7.4 MPN/100ml for the hostel and canteen respectively. The bacterial isolates obtained from the HezUni female hostel harvested rainwater in March were E. coli, Citrobacter sp., Staphylococcus sp., Shigella sp. and Aerobacter sp. In contrast, the canteen-harvested rainwater had Shigella sp., E. coli, Staphylococcus sp. and Micrococcus sp. The presence of coliforms in most of the rainwater available to students in tertiary institutions in the Orlu zone further defeats the classification of these sources as potable. There is a need for the universities to intensify the call for the provision and monitoring of rural water supplies as a roadmap for stemming the tides of water-borne diseases.
Except where otherwise noted, content on this site is licensed under Creative Commons Attribution-Share Alike 4.0 International License.
