My research projects combine biomedical signal processing, machine and deep learning, multimodal physiological sensing, and embedded computing to develop intelligent healthcare and neurotechnology systems.
A particular focus of my work is moving beyond offline model development towards practical systems capable of operating with wearable sensors and real-time physiological data.
Development of an end-to-end system for objective pain assessment using physiological signals, multimodal deep learning and real-time embedded inference.
EDA · BVP/PPG
Signal preprocessing · Deep learning · Multimodal fusion · Transformer-based modelling
Wearable sensing · Raspberry Pi · Continuous real-time inference
Investigation of functional near-infrared spectroscopy for decoding pain-related neurophysiological responses and estimating multiple levels of pain intensity.
fNIRS
Neurophysiological signal processing · Feature representation · Machine learning · Deep learning
Objective pain assessment · Neurophysiological decoding
Multimodal modelling of peripheral physiological and neurophysiological responses for pain assessment by combining electrodermal activity with functional near-infrared spectroscopy.
EDA · fNIRS
Multimodal fusion · Attention mechanisms · Transformer-based learning
Gradient-infused attention and Transformer-based multimodal architecture
Investigation of hybrid EEG and fNIRS for neurophysiological decoding, with a focus on motor imagery and multimodal learning for brain-computer interface applications.
EEG · fNIRS
Signal preprocessing · Deep learning · Multimodal fusion · Representation learning
Motor imagery · Brain-computer interfaces · Neurotechnology
Development and evaluation of physiological sensing pipelines that integrate wearable biosensors with signal processing, machine learning and embedded inference.
EmotiBit · BIOPAC · PLUX Biosignals · Raspberry Pi · STM32
Sensor integration · Data acquisition · Real-time preprocessing · Edge inference
Wearable healthcare · Physiological computing · Real-time intelligent systems
Prior to my current research, I co-founded and led signal and image processing research activities within the Syndicate of Electronic and Embedded Design (SEED) Research Group at UET Taxila. This work included biomedical signal processing, computer-aided diagnosis, biometric authentication and intelligent healthcare applications.
Research focused on the use of physiological biosignals for computer-aided identification and assessment of cardiac disorders.
Investigation of biometric authentication using physiological signals including ECG, PCG, PuPG and PPG.
Research on identifying pulmonary disorders from lung and breathing sounds using biomedical signal analysis.
Research addressing prediction of term and preterm delivery and classification of vaginal and caesarean deliveries.
Classification of EMG and EEG signals for the assessment of neuromuscular disorders.
Investigation of biomedical signal and image analysis for the identification of gastrointestinal diseases.