rahulrkm0038@gmail.com | +61 435 844 977 | Melbourne, Australia
LinkedIn | GitHub | Portfolio | ORCID | Google Scholar
Full-scholarship PhD researcher in biomedical signal processing and machine learning, focused on extracting vital signs from optical and video signals. Experienced in end-to-end rPPG pipelines, robust feature extraction, CNN/Transformer model development, and real-world validation for heart rate, SpO₂, and cuffless blood pressure estimation. Strong fit for research on physiological monitoring from video images through a combination of signal processing, reproducible experimentation, and deployment-aware model design.
- Biomedical Signal Processing: rPPG, time-series analysis, bandpass filtering, spectral analysis, signal preprocessing, feature extraction, evaluation and validation
- Machine Learning / AI: CNN, Transformer, self-attention, PyTorch, TensorFlow, Keras, Scikit-learn, XGBoost, Random Forest, LSTM
- Data and Research Systems: Python, MATLAB, SQL, Bash, OpenCV, NumPy, Pandas, PostgreSQL, MongoDB, Docker, Git, LaTeX
Monash University, Melbourne
- Developed smartphone-video pipelines for contactless heart rate, SpO₂, and blood pressure estimation, covering face ROI extraction, rPPG preprocessing, feature engineering, and model training.
- Built CNN- and Transformer-based physiological monitoring models with color transforms, bandpass filtering, and self-attention to improve robustness in real-world mobile-health settings.
- Analyzed 5,000+ video samples across multiple datasets, reporting 95%+ accuracy on heart-rate and blood-pressure tasks and MAE below 5 mmHg for contactless blood pressure estimation.
- Designed reproducible cross-subject evaluation, ablation, and model-optimization workflows, reducing inference time by about 40% while keeping model size below 10M parameters.
Department of Physics, BITS Pilani, India
- Investigated phase transitions in ferromagnetic systems using Monte Carlo simulations with the Metropolis algorithm on 2D and 3D Ising lattice models.
- Calculated energy, magnetization, susceptibility, and specific heat across temperature sweeps, with emphasis on extracting stable signal-like patterns from noisy simulations.
- Reduced runtime by 40% through vectorization and parallel temperature computations.
- Applied finite-size scaling and power-law fitting to estimate critical exponents and analyze the effect of disorder on transition behavior.
Aglow Engineers, Kolkata
- Architected the company’s first centralized SQL-based data infrastructure by migrating manual-entry systems into a queryable operational database.
- Automated data comparison and reporting workflows in Python, improving issue visibility for non-technical stakeholders.
- Built machine-learning forecasting models using Random Forest and LSTM, reporting 89% accuracy and a 22% reduction in system downtime.
Centre for Railway Information Systems (CRIS), New Delhi
- Queried rolling-stock maintenance records from the WISE database to build structured training datasets for predictive modelling.
- Developed anomaly-detection workflows in Python and XGBoost with reported 94% accuracy for 48-hour failure prediction.
- Contributed to predictive-maintenance deployment across 200+ railway assets, with reported 18% reduction in workshop downtime.
Xilinx (now AMD), Hyderabad
- Developed the “Data Flow Synchronicity Checker” for automated file-hash verification between Head and Artifactory directories.
- Supported nightly validation workflows covering 1,000+ file checks per day through Bash scripting, cron scheduling, and email reporting.
- Collaborated with design, verification, and build teams using Helix Swarm and Perforce-based workflows.
Evolving Blood Pressure Estimation: From Feature Analysis to Image-Based Deep Learning Models
Roha, V. S., Ranjan, R., & Yuce, M. R. (2025)
Journal of Medical Systems, 49(1), 97
VITAL Net: A Hybrid Framework for SpO₂ and HR Estimation Using Smartphone rPPG Video
Ranjan, R., Roha, V. S., & Yuce, M. R. (2026)
Published in: 2026 IEEE Applied Sensing Conference (APSCON)
🎓 PhD, Electrical & Computer Systems Engineering (Expected 2029, March 2026)
Monash University
🎓 Master of Artificial Intelligence (2023–2025)
Monash University
Result: Distinction
🎓 M.Sc. (Hons.) Physics; B.E. (Hons.) Electronics & Instrumentation (2017–2022)
BITS Pilani
Thesis: Effect of Disorder on Critical Exponents
Supervisor: Prof. P. K. Thiruvikraman
🥈 The Duke of Edinburgh’s International Award, Silver (2015)
Note: Research code is released per publication guidelines. For detailed project information, visit my portfolio website.
- VITAL Net Framework: Hybrid CNN-Transformer architecture for simultaneous SpO₂ and HR estimation
- Contactless BP Monitoring: Deep learning models for cuffless blood pressure measurement
- rPPG Robustness: Improving signal quality across lighting, motion, and demographic variations
- 🔄 MLOps and reproducible research workflows
- 🎯 Motion-aware modeling and sensor fusion for biosignals
- 🏗️ Scalable ML infrastructure with Rust
I'm always interested in collaborating on:
- Mobile health and wearable sensing research
- Biomedical signal processing challenges
- Real-world ML robustness and generalization
- Healthcare AI applications
📧 Email: rahulrkm0038@gmail.com
🌐 Website: rahul201722.github.io
💼 LinkedIn: linkedin.com/in/rahul-ranjan-b595891b1
