solaiman ahmed asif

Machine Learning and signal processing Practitioner in Healthcare Monitoring with wearable platform.

I worked as a postdoctoral researcher in advanced medical research institute in Japan. I am experienced with MATLAB and Python programming for Signal Processing and Machine Learning application.

My doctoral thesis represents wearable sensor-based healthcare monitoring with Machine Learning (ML) technique where some novel features were developed for classification of human activities along with physiological data (ECG, PPG, EDA). The first study of this dissertation tried to create such angle variance based vibrational intensity estimation produced by human activity which can give movement intensity as well as posture information.  In the second study signal quality index-based stress beat estimation method is proposed from PPG. The morphology of the PPG signal was quantified as signal quality index by comparing the template beat (extracted from resting conditions) to the PPG beats recorded during vigorous physical activity. Whereas the morphology results corroborate the findings from traditional HRV parameters generally used for stratifying stress. Lastly, physiological signal was utilized to classify human activities along with accelerometer features and several classifiers was used to train model. Features from ECG and EDA contribute to classify different stressful activities which are difficult to classify with wrist and chest-based accelerometers.