Project: ECG Signal Processing for HRV Insights
In this project, I implemented a full-workflow analysis to convert raw ECG data into meaningful heart rate variability (HRV) metrics. Using the MIT-BIH Arrhythmia Database, I:

• Performed signal preprocessing (filtering, noise removal) to extract clean ECG waveforms.

• Detected R-peaks and measured R–R intervals using differentiation, squaring, moving-window integration, and adaptive thresholding for high precision.

• Validated R–R interval detection results with millisecond-level accuracy and preserved short-term HRV measures such as RMSSD and SDNN.

• Framed the broader context of wearables and HRV interpretation, setting the stage for deriving physiological insights around stress, recovery, and autonomic balance.

Technologies & Skills:
Python (signal processing libraries, Jupyter/Kaggle notebook), statistical methods for variability analysis, time-domain HRV metrics, and an understanding of physiological data flows from raw signals to actionable insights.

Outcome:

This work demonstrates how to bridge raw biosignal data with meaningful health-monitoring metrics — highlighting how technical signal processing underpins modern wearable readiness/recovery scores.

Blog Article:  Signal to Meaning (Part I): The Signal Processing That Makes HRV Possible

Data & Notebook on Kaggle
Companion Notebook: Signal to Meaning (Part I): The Signal Processing That Makes HRV Possible

The dataset — the MIT-BIH Arrhythmia Database from Kaggle — includes reference annotations, allowing us to validate how well our ECG signal processing HRV workflow performed.