Vishnu Kumar
Assistant Professor of Industrial and Systems Engineering
Morgan State University
Areas of Expertise: Applied AI and Machine Learning
Vishnu Kumar is an assistant professor of industrial and systems engineering at Morgan State University. His research focuses on smart systems, process optimization, and data analytics across manufacturing, supply chain, and healthcare systems. Leveraging feature engineering, predictive modeling, and interpretable machine learning on public health datasets, Kumar has authored multiple high-impact publications applying advanced computational methods to advance intelligent decision-making in public health and industrial settings.
-
V Kumar, R Butler (2026). International Journal of Mental Health and Addiction, Volume 24, pages 1865–1882
Abstract: The opioid crisis remains a significant public health challenge in the USA, with opioid-related overdose deaths continuing to rise in recent years. Understanding the patterns and key determinants of these deaths is crucial for improving prevention, intervention, and resource allocation strategies. This study analyzes temporal trends and county-level distributions of opioid overdose fatalities and applies machine learning models to predict county-level OOD rates using publicly available data from 2016 to 2023. Findings reveal that approximately 30% of US counties experienced an increase in opioid-related overdose deaths during this period, with 15 counties reporting spikes exceeding 50%. An eXtreme Gradient Boosting (XGBoost) regressor-based machine learning model was applied using 18 distinct features across 3142 counties, achieving an R 2 value of 0.93. SHapley Additive exPlanations (SHAP) were employed to assess the contribution of each feature to the model's predictions. The most influential features included "County Population," "Aver-age Mentally Unhealthy Days," "Median Age," "Percentage of the Population Uninsured," "Violent Crime Rates," and "Percentage of the Black Population." These key features were then used to develop a Risk Index for identifying counties at high risk of opioid-related overdose deaths. This machine learning-driven study offers a valuable framework for targeted opioid-related overdose death prevention and intervention efforts, and optimized resource allocation to combat the ongoing opioid epidemic in the USA.
-
V Kumar (2025). International Conference on Information and Communication Technology for Intelligent Systems, pages 495-505
Abstract: Heart disease remains a leading cause of mortality in the United States, responsible for approximately 1 in 5 deaths in 2022. Modifiable behavioral and lifestyle factors, such as smoking, physical activity, and diet, play a critical role in cardiovascular risk. This study applies a machine learning (ML) approach to predict heart disease risk in the U.S. using data from the 2022 Behavioral Risk Factor Surveillance System (BRFSS). Three ML-based classification models were developed using ten key behavioral and lifestyle features: general health perception, days of poor physical and mental health, time since the last checkup, physical activity engagement, average sleep duration, smoking status, e-cigarette use, body mass index (BMI), and alcohol consumption. Among the three ML-based classification models, XGBoost exhibited superior performance, achieving an F1-score of 0.92 with balanced precision and recall across both classes. Shapley Additive Explanations (SHAP) was then used to identify the impact of behavioral and lifestyle factors on heart disease risk. Global SHAP analysis revealed that general health, poor mental health, and BMI were the most influential features affecting heart disease risk. Local SHAP analysis showed that the importance of individual features varied across different observations, with factors such as: time since the last checkup, and smoking status significantly influencing heart disease risk for certain individuals. These findings demonstrate the potential of explainable ML techniques to identify actionable, personalized cardiovascular risk factors. The insights gained can help healthcare providers tailor interventions and prevention strategies, prioritize high-risk individuals for early detection, and allocate resources more effectively to reduce the burden of heart disease.
-
V Kumar (2025). International Journal of Mechanical Engineering Education
Abstract: Recent advancements in Artificial Intelligence (AI) have created new opportunities to enhance teaching and learning in engineering education. Among these, ChatGPT, an AI language model developed by OpenAI, has gained significant popularity for its ability to generate human-like responses, explain technical concepts, and support problem-solving. While much of the current discourse focuses on student use of ChatGPT, its potential to assist instructors in designing assessments and providing feedback remains underexplored. This gap is particularly relevant in design and manufacturing engineering courses, where instructors often face challenges in crafting open-ended questions, programming-based assessments, and grading complex student submissions with consistency and clarity. To address this gap, this paper investigates the use of ChatGPT as a digital assistant to support assessment-related tasks in design and manufacturing engineering education. Using a reflective case study approach, the work evaluates ChatGPT across five key use cases: (i) Designing Multiple-Choice Questions, (ii) Designing Descriptive Questions, (iii) Designing Computational and Analytical Questions, (iv) Designing Image and Computer Programming-Based Questions, and (v) Designing Rubrics for Grading. Outputs were assessed using a structured pedagogical framework grounded in course learning objectives and Bloom's Taxonomy to evaluate cognitive depth, relevance, and alignment with instructional goals. Findings indicate that ChatGPT can effectively generate relevant and diverse assessment items, reduce instructor workload, and support personalized feedback. However, limitations such as occasional inaccuracies, difficulty generating technical visuals, and the need for instructor oversight highlight the importance of critical evaluation. By outlining both opportunities and constraints, this study offers actionable insights for integrating ChatGPT tools into assessment practices in design and manufacturing engineering education.
-
V. Kumar (2026). Knowledge Management in Learning and Education 6(1), 7
Abstract: This study examines the factors driving perceived Study Efficiency and Exam Readiness associated with ChatGPT use among STEM students in higher education. Although prior research on generative artificial intelligence (GenAI) has largely focused on adoption and attitudes using descriptive or linear statistical approaches, limited empirical work has explored how students’ interactions with such tools relate to learning-related outcomes. To address this gap, this study applies an interpretable machine learning (ML) framework to identify key predictors of learning gains from ChatGPT use. Data were obtained from a large-scale global survey of STEM students (n = 10,525) across 109 countries and territories, capturing usage patterns, perceived capabilities, satisfaction, and academic outcomes. Two eXtreme Gradient Boosting (XGBoost)-based ML classification models were developed to predict Study Efficiency and Exam Readiness, and SHapley Additive exPlanations (SHAP) were used to interpret feature-level contributions. The models achieved strong predictive performance for the high-gain class, with an accuracy of 0.93 (F1 = 0.96) for Study Efficiency and 0.86 (F1 = 0.92) for Exam Readiness. Results indicate that motivation, personalized learning support, improved access to knowledge, facilitation of study activities, and exam-focused study assistance are key predictors of learning gains. These findings offer empirical and practical insights for educators and policymakers seeking to design effective and pedagogically sound AI-assisted learning environments in STEM education.
