Blind source separation (BSS) is a powerful technique used in biomedical applications to extract hidden signals from recordings containing mixtures of multiple sources.

In biomedicine, BSS is particularly useful because medical signals are often contaminated by interference from other sources. For instance, an electroencephalogram (EEG) might pick up muscle activity or electrical noise from the environment along with brain activity. BSS can separate these unwanted signals, allowing for clearer analysis of the brain waves. BSS offers a valuable tool for researchers and clinicians to gain deeper insights into various biological processes. By removing unwanted noise and isolating specific signals of interest, BSS helps improve the accuracy and effectiveness of medical diagnosis and treatment.

Here are some specific applications of BSS that I worked on:

• EEG analysis: Isolating specific brain activity patterns related to sleep stages, epilepsy, or cognitive function.
• EMG analysis: Isolating complex muscle activity from different muscles to better understand of human movements and classify complex hand gestures.
• Fatigue Identification: Separating the complex muscle activity to identify different fatigue levels in different age groups

My expertise in biomedical signal processing signifies a deep understanding of the techniques and tools used to analyze and interpret signals generated by the human body. I possess strong knowledge of signal processing fundamentals, including Fourier transforms, filters, and statistical analysis methods. I have an in-depth understanding of various biological signals like EEG, ECG, and EMG, encompassing their generation mechanisms, physiological meaning, and interpretation. Furthermore, I am proficient in noise reduction techniques and feature extraction to isolate signals of interest. My knowledge extends to machine learning, with experience using algorithms like Twin SVMs for signal classification. Having experience working with Matlab, R and Python and libraries allows me to implement these techniques and develop practical solutions.

I possess expertise in sleep signal analysis and algorithm development. My knowledge encompasses human sleep stages, their physiological correlates in brain waves, muscle activity, and eye movements. I am proficient in applying signal processing techniques to sleep data and have experience developing algorithms using machine learning and statistical modeling. My skills extend to software development using Matlab, R, Python and relevant libraries, allowing me to translate concepts into practical tools for sleep analysis. I have experience focusing on algorithms for upper airway collapsibility and its relevance to sleep research.

I bridge the gap between engineering and health by specializing in wearable signal processing. My expertise lies in extracting meaningful information from the raw data streams generated by wearable devices.

Key Skills:

• Signal Processing Wizard: I wield a powerful toolkit of filtering, feature extraction, and time-frequency analysis techniques to transform raw sensor data into actionable insights.
• Biomedical Signal Decoder: I fluently understand the language of physiological signals like ECG, EMG, and PPG, unlocking their secrets to assess health and well-being.
• Wearable Data Whisperer: I am adept at navigating the complexities of wearable data, tackling challenges like motion artifacts and data gaps to ensure accurate analysis.

My expertise empowers the development of next-generation wearable applications that translate raw data into actionable health insights.