Multidisciplinary Biomedical Data Scientist and Algorithm Researcher with 5+ years of experience bridging the gap between physical hardware and advanced data analytics. Currently a final-year Doctoral Researcher at the University of Turku, specializing in applied machine learning, scalable data engineering, and biosignal processing. I possess a proven track record of developing end-to-end solutions—from designing embedded sensor prototypes (C/C++, CAD) to building automated ETL cloud pipelines (Python, SQL, Apache Airflow) and deploying deep learning models (TensorFlow, PyTorch). Passionate about transforming noisy, real-world data into robust, production-ready insights for health technology, IoT, and broader industrial applications.
English: Fluent (TOEFL: 105/120)
Finnish: Basic (A2)
- Designed and implemented advanced biosignal processing algorithms in Python, explicitly utilizing Polar research dataloggers to analyze continuous ECG, Bioimpedance (BioZ), and mechanical cardiac signals.
- Translated complex physiological phenomena into robust algorithms, extracting clinically relevant features (e.g., LVET, PEP, IVCT, heart and respiration rates).
- Built and orchestrated scalable ETL and data analysis pipelines in Python and SQL using Apache Airflow, improving processing efficiency by 30%.
- Automated data validation processes, reducing manual efforts and ensuring data integrity across all environments.
- Contributed to the development of wearable research platforms, applying DSP and statistical modeling techniques to extract robust features from noisy, real-world biosignals.
- Bridged hardware and embedded software integration by designing CAD prototypes to optimize physical sensor placement and signal acquisition quality.
- Designed and implemented ML models and algorithms for anomaly detection in multimodal sensor data for real-time health monitoring.
- Contributed to technical innovation through peer-reviewed publications (IEEE JBHI, CinC).
01/2020 - 02/2022
Wize Analytics
- Developed and implemented robust data processing algorithms in Python, deploying them as functional components within a larger operational system.
- Optimized deep learning models (CNNs) and data pipelines to ensure low-latency, highly accurate algorithm performance in live, real-time traffic environments.
05/2025 - 01/2026
Hardware Integration
- Designed and implemented a wearable proof-of-concept prototype for simultaneous, high-sample-rate cardiac data acquisition, prioritizing PPG, PCG and MEMS-based (BCG/GCG) sensors alongside ECG.
- Ran feasibility studies, tuning sensor sampling frequencies and optimizing physical placement to evaluate hardware performance and maximize physiological signal fidelity.
- Programmed a low-power embedded system using Python and C/C++ to read direct sensor registries, ensuring precise, time-synchronized data streaming across multiple sensing modalities.
- Developed a live telemetry dashboard for real-time signal quality evaluation and engineered an automated pipeline for seamless, high-fidelity data transfer to the cloud.
02/2024 - 03/2026
University of Turku
- Mentored university students through hands-on technical labs utilizing electronic measurement equipment and biosignal simulators to design and test physiological circuits.
- Explained complex technical, physiological, and mathematical concepts clearly to multidisciplinary students.
08/2023 - 01/2027
University of Turku
Thesis: Unobtrusive Monitoring for Cardiovascular Disease using ML.
Core Focus: AI-driven applied solutions, Multimodal data analysis, Biosignal quality assessment.
09/2019 - 02/2022
Iran University of Science and Technology
Thesis: Ultrasound Image Despeckling with Deep Learning.
GPA: 4.0/4.0
Core Focus: Computer vision techniques, deep learning, and practical image processing.
09/2014 - 09/2018
Bu-Ali Sina University
Thesis: Applications of FibroScan.
GPA: 3.5/4.0
Core Focus: Signal processing, embedded systems, microcontroller programming, and robotics.
- S. Seifizarei et al, “Autocorrelation-Based Algorithm for Longitudinal Multi-Node Accelerometer Heart Rate Monitoring in Clinical Settings”, Biomedical Signal Processing and Control.
- S. Seifizarei et al, “Continuous Radar-based Heart Rate Monitoring using Autocorrelation-based Algorithm in Intensive Care Unit.”, IEEE Journal of Biomedical and Health Informatics
- S. Seifizarei et al, “Evaluating Piezoelectric Ballistocardiography for Post-Surgical Heart Rate Monitoring.”, Computing in Cardiology 2024
Address
Univ. of Turku, Turku, Finland
sepehr.seifizarei@utu.fi