Causal Machine Learning for Optimal Antibiotic Timing in Sepsis
Website • LinkedIn • Google Scholar • ORCID • IEEE Explore • ResearchGate
I am a Ph.D. Candidate in Computer Engineering at Case Western Reserve University, with research interests at the intersection of causal machine learning, healthcare AI, intelligent systems, data science, and computer engineering.
My current doctoral research focuses on using causal machine learning for optimal timing of antibiotic prescribing in sepsis, with the broader goal of developing data-driven clinical decision support systems that can improve patient outcomes and support evidence-based healthcare.
I am especially interested in building AI systems that are not only technically strong, but also clinically meaningful, interpretable, ethical, and useful in real-world environments.
| Area | Current Focus | Application |
|---|---|---|
| Causal Machine Learning | Treatment timing and outcome modeling | Sepsis and clinical decision support |
| Healthcare AI | Interpretable and useful AI systems | Data-driven medicine |
| Machine Learning & Data Science | Predictive modeling, optimization, analytics | Research and real-world datasets |
| Intelligent Systems | Decision-oriented AI workflows | Clinical and engineering systems |
| Computer Networks | Intrusion detection and optimization | Cybersecurity and network intelligence |
- Causal Machine Learning
- Healthcare Artificial Intelligence
- Sepsis Prediction and Clinical Decision Support
- Optimal Timing of Antibiotic Prescribing
- Machine Learning and Data Science
- Data Engineering and Analytics
- Network and Intelligent Systems
- Optimization and Decision-Making Systems
- AI-Driven Research Workflows
My doctoral research focuses on developing temporal causal machine learning methods to study how clinical decisions, antibiotic timing, and patient outcomes interact in emergency department sepsis care.
The goal is to estimate patient-specific timing-response patterns and support interpretable, clinically meaningful decision-support systems that can help identify when delaying antibiotics may become harmful.
Key themes include:
- Modeling temporal clinical events and irregular EHR data
- Estimating counterfactual outcomes under alternative antibiotic timing strategies
- Identifying meaningful treatment windows and patient-specific delay tolerance
- Studying clinician decision-making patterns and workflow-driven confounding
- Applying causal reasoning to healthcare data
- Supporting interpretable and actionable AI in clinical settings
- Building foundations for intelligent clinical decision support systems
Technical focus: causal inference, temporal machine learning, healthcare AI, EHR data, sepsis, clinical decision support.
A MATLAB-based machine learning and optimization project for network intrusion detection.
The project combines Multi-Layer Perceptron neural networks with the Grasshopper Optimization Algorithm to reduce intrusion detection error by optimizing model parameters such as weights and biases.
Technical focus: MATLAB, neural networks, intrusion detection, optimization, metaheuristic algorithms, cybersecurity, KDD/UNSW datasets.
Related publication:
GOAMLP: Network Intrusion Detection With Multilayer Perceptron and Grasshopper Optimization Algorithm
Repository:
View GOAMLP on GitHub
A Python-based reinforcement learning project for optimizing electric vehicle charging behavior.
The system models charging decisions using reinforcement learning to balance charging efficiency, grid load, cost, and user demand.
Technical focus: Python, reinforcement learning, optimization, Gym environment, smart grids, EV charging, energy systems, decision-making.
Engineering value: This project demonstrates applied AI for infrastructure optimization and intelligent resource management.
Repository:
View EV Charging Optimization on GitHub
My academic identity is built around the intersection of computer engineering, machine learning, healthcare AI, and intelligent decision systems.
I am interested in developing AI methods that connect technical rigor with real-world impact, especially in healthcare environments where timing, interpretability, and decision quality matter.
-
GOAMLP: Network Intrusion Detection With Multilayer Perceptron and Grasshopper Optimization Algorithm
IEEE Access, 2020. Focused on machine learning, neural networks, metaheuristic optimization, and network intrusion detection.
CWRU Repository • GitHub Code -
Machine Learning in Apache Spark Environment for Diagnosis of Diabetes
Preprints, 2021. Focused on scalable machine learning, Apache Spark, healthcare-related data analysis, and diabetes diagnosis.
CWRU Repository -
Present a New Method for Energy Efficient Routing for Wireless Body Area Networks
Focused on wireless body area networks, healthcare-oriented sensor networks, and energy-efficient routing optimization.
CWRU Repository
I am open to research and engineering collaborations in:
- Causal machine learning and healthcare AI
- Sepsis prediction and clinical decision support
- Data science and machine learning systems
- AI-driven optimization and intelligent decision-making
- Applied AI projects with real-world impact
I aim to develop intelligent, ethical, and clinically meaningful AI systems that can support better decision-making, improve healthcare outcomes, and contribute to a more reliable future for data-driven medicine.
My long-term vision is to combine machine learning, causal reasoning, healthcare data, and intelligent systems to create technologies that are technically rigorous and genuinely valuable for people.