CALL FOR PAPER - Vol 10 - Issue 1

Pneumonia still constitutes one of the major causes of mortality across the globe, specifically targeting children, elderly people, and immunocompromised patients. Diagnosis and timely intervention can help reduce disease burden and patient morbidity. Traditional methods of diagnosing pneumonia using chest X-ray images depend largely on the involvement of radiologists, making the entire process slow and prone to observer variability, especially in limited health care facilities. This paper introduces an automated pneumonia diagnosis system based on deep learning using ResNet34 architecture. The model uses image pre-processing techniques, transfer learning, and data augmentation to boost feature extraction capabilities while reducing overfitting. The trained system classifies chest X-ray images into pneumonia and non-pneumonia classes using a user-friendly web application built on the Streamlit platform. The designed system provides a good computer-aided diagnosis system which may help medical professionals conduct initial screening for pneumonia, reduce diagnostic effort, and increase the accessibility of health care facilities in rural areas.

Heart disease is a major cause of mortality worldwide, making early and accurate prediction essential for effective clinical intervention. This paper presents a feature-optimized heart disease prediction framework using machine learning techniques to improve diagnostic performance. The proposed approach incorporates data preprocessing, feature optimization, and classification to identify the most relevant clinical attributes while reducing redundancy. Several machine learning algorithms, including Logistic Regression, Decision Tree, Random Forest, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and XGBoost, are evaluated using standard performance metrics such as accuracy, precision, recall, F1-score, and ROC-AUC. Experimental results demonstrate that feature optimization enhances prediction accuracy and model efficiency, with ensemble-based methods achieving superior performance. The proposed framework offers an effective decision-support tool for the early detection of heart disease and has the potential to assist healthcare professionals in improving diagnostic accuracy and patient care.

Despite numerous advancements in modern medicine, cardiovascular diseases still rank among the top mortality factors all around the world, with 18 million annual deaths caused by heart diseases according to the World Health Organization estimates. Identification of individuals with elevated risk of heart diseases is important in order to reduce the mortality rate especially in developing countries with limited access to medical facilities. Thanks to recent progress made in Artificial Intelligence and Machine Learning, it became possible to create intelligent diagnostic systems that can help doctors diagnose patients at an early stage. This paper aims at evaluating the performance of five supervised machine learning algorithms for heart disease prediction, namely, Random Forest, Support Vector Machine, Logistic Regression, Decision Tree, and K-Nearest Neighbors. Based on experimental evaluation, the Random Forest algorithm shows the best performance providing an accuracy of 90.16% while logistic regression shows accuracy of 85.25%, and the KNN classifier shows the lowest accuracy level of 67.21%. In addition to classification accuracy, the precision and recall values are also estimated based on the confusion matrix. The achieved results suggest that ensembles of learning algorithms provide better predictions while maintaining appropriate level of generalization.

The seismic design of tall reinforced concrete (RC) buildings requires explicit consideration of soil-structure interaction (SSI), which governs the dynamic characteristics, load distribution, and settlement behaviour of the foundation system. Conventional design practice frequently neglects SSI through the rigid-base assumption, potentially yielding unconservative estimates of footing demand. The plan geometry of isolated footings directly controls the SSI interface stiffness, contact pressure distribution, and Winkler spring response under combined gravity and seismic loading. This review synthesises four decades of theoretical, experimental, and computational research on SSI-based footing behaviour for tall buildings, with particular focus on the comparative performance of rectangular, square, oval, and elliptical isolated footings. To consolidate and critically evaluate published research on SSI modelling approaches for shallow foundations; to review comparative FEM-based studies of footing geometry effects on SSI-mediated structural response using STAAD.Pro and equivalent platforms; and to identify the most SSI-compatible footing geometry for tall RC buildings under Indian seismic conditions, with respect to shear force, axial force, support reaction uniformity, deflection, and construction cost. Oval and elliptical isolated footings consistently exhibit superior SSI performance relative to conventional rectangular and square profiles. The smooth curved perimeter eliminates corner Winkler spring concentrations, producing more uniform contact pressure distributions and lower peak structural demands. SSI-inclusive FEM analyses report reductions of 84-95% in maximum shear force, 83-90% in peak support reaction, 89% in axial force, and approximately 49% in maximum footing deflection for oval footings versus rectangular equivalents under seismic loading. Reinforcement savings of 15-17% yield proportional cost reductions. Advanced SSI modelling (Pasternak, continuum FEM) and machine learning integration are emerging research frontiers. Priority directions include nonlinear SSI analysis with plastic soil yielding, experimental V-H-M validation of oval footings, taller building (G+10 to G+20) and higher seismic zone (III-V) parametric studies, two-parameter Pasternak SSI modelling, and development of codal design provisions for non-conventional footing geometries.

The growth of high-rise construction in seismically active regions has made the control of the lateral response of buildings a governing design concern. Reinforced-concrete (RC) shear walls, steel bracing, and supplemental dampers are the three dominant strategies used to resist earthquake and wind actions, and increasingly they are combined into hybrid systems whose behaviour is evaluated numerically in finite-element platforms such as ETABS.This paper reviews the state of the art on the dynamic analysis of high-rise structures that combine shear walls, bracing, and damper systems under lateral loading. It compares the working mechanisms, analysis methods, and reported performance of these systems, and identifies the technologies and design methods that currently dominate the field.Across the reviewed studies a consistent pattern emerges: stiffness-based devices (walls and bracing) shorten the fundamental period and reduce displacement but attract larger inertial forces, whereas velocity-dependent dampers reduce drift, member forces, and floor acceleration without a comparable increase in stiffness. Hybrid wall–brace–damper systems combine these benefits, and response-spectrum and non-linear time-history analysis, together with the P-Delta effect, are the standard evaluation tools. Recent work is moving from passive devices towards semi-active, adaptive, and machine-learning-assisted design.Priorities identified include soil–structure-interaction-aware design, optimal placement and sizing of dampers, life-cycle and resilience-based cost assessment, and data-driven optimisation of hybrid configurations. These directions frame the motivation for the detailed comparative study reported subsequently by the authors.

This study examined the impact of the professional skills of school heads on teachers' performance in the Division of Camarines Sur during School Year 2024–2025. Specifically, it assessed the extent to which school heads manifested professional skills in leadership, innovation, information technology, communication, delegation, decision-making, and problem-solving and determined how these competencies influenced teachers' performance in terms of teaching efficiency, instructional productivity, work environment, motivation, engagement, and commitment. The study employed a descriptive-survey research design involving school heads and teachers from public secondary schools across the five congressional districts of the Division of Camarines Sur. A validated researcher-made questionnaire served as the primary data-gathering instrument. Statistical tools such as weighted mean, Kendall's Coefficient of Concordance (W), and ranking were utilized in analyzing the collected data. Findings revealed that school heads demonstrated a high level of professional skills across all indicators, which significantly contributed to improved teacher performance. Leadership, communication, and decision-making emerged as the strongest competencies influencing instructional effectiveness and teacher motivation. Respondents also identified challenges related to limited instructional resources, teacher workload, facilities management, and student discipline. Based on the findings, an action plan was developed to strengthen leadership competencies and sustain teacher performance. The study concludes that enhancing the professional skills of school heads is essential in promoting teacher effectiveness and improving the overall quality of basic education.