Risini Dinara Kumarasinghe
Passionate about Biomedical Engineering, Signal Processing & Neural Interfaces
I am a Biomedical Engineering undergraduate at the University of Moratuwa, Sri Lanka, deeply passionate about the intersection of engineering and medicine. My interests primarily lie in biomedical signal processing, neural engineering, and wearable healthcare systems, where I strive to develop innovative solutions that bridge the gap between technology and clinical rehabilitation.
In my research experience, I worked as a Research Intern at the School of Computer Science, University of Sydney. My work involved electrotactile stimulation for sensory rehabilitation, including waveform design (monophasic, biphasic, sine, chirp) and real-time multi-electrode control using precise PWM timing. I also explored Ear-EEG acquisition, signal analysis techniques, and safety considerations for non-invasive monitoring, and investigated mycelium-based conductive composites for biosensing and bioelectrical interface stability.
Alongside research, I have contributed to biomedical product development through the MeasureUP project, a World Bank–funded initiative to develop a portable human height-measuring device for clinical settings. My role included algorithmic support for handling low ultrasound reflection caused by dense scalp hair and mechanical enclosure design improvements, providing hands-on exposure to medical device engineering and translational development.
Academically, I maintain a GPA of 3.90/4.0 and have been on the Dean’s List for multiple semesters. I achieved All-Island 2nd place at GCE O/L and All-Island 73rd place at GCE A/L, earning gold medals for academic excellence. My coursework and projects reflect a strong foundation in DSP, adaptive filtering, time–frequency analysis, machine learning, and multichannel biosignal analysis.
Beyond academics, I am currently involved in developing a non-invasive EEG-based BCI game for a locked-in pediatric patient, focusing on signal processing pipelines, lightweight classifiers, and user-specific adaptation. My broader interests lie at the intersection of biomedical signal processing, wearable systems, and human-centered neurotechnology, with the goal of building clinically meaningful, technically robust systems.
I conducted research on electrotactile stimulation systems for sensory rehabilitation, focusing on real-time waveform generation and multi-electrode control. This involved designing and implementing monophasic, biphasic, sine, and chirp stimulation patterns using precise PWM timing for embedded haptic interfaces. Additionally, I investigated Ear-EEG signal acquisition, covering sensor placement strategies, signal quality analysis, and safety considerations for non-invasive monitoring. I also designed and evaluated mycelium-based conductive composites for biosensing applications, analyzing their bioelectrical conductivity and stability, while contributing to the end-to-end research workflow from experimental design to technical documentation.
University of Moratuwa, Sri Lanka
GPA: 3.90 / 4.00 · Dean’s List: Semesters 1, 2, 4, 5, and 6
Relevant Coursework: Biosignal Processing, Digital Signal Processing, Signals and Systems, Pattern Recognition, Neural Networks, Medical Imaging, Modelling and Analysis of Physiological Systems, Biomedical Device Design, Medical Electronics, Applied Statistics, Numerical Methods, Linear Algebra, Differential Equations
Devi Balika Vidyalaya, Colombo
All-Island Rank: 73 out of ~34,389 candidates
Results: A passes in Mathematics, Physics, Chemistry, and English
Awards: Gold Medal for Best Performance at GCE A/L Examination
Samudradevi Balika Vidyalaya
All-Island Rank: 2 out of 312,464 candidates
Awards: Gold Medal for Best Performance at GCE O/L Examination
Developing a non-invasive EEG-based BCI that enables a locked-in pediatric patient to play computer games. Key contributions include designing robust signal processing pipelines, real-time lightweight classifiers, and patient-specific calibration. Additionally, a custom dry-electrode headset and Unity-based game interface are being built to ensure comfort and effective training.
This project involved implementing end-to-end biomedical signal processing pipelines for ECG and related physiological signals using MATLAB. Key contributions include the design and evaluation of digital filters (FIR, IIR, moving average, comb), adaptive filtering techniques (LMS, RLS, Wiener) for non-stationary noise suppression, and wavelet-based time–frequency analysis for feature exploration. The work emphasized reproducibility through reusable scripts and experimental validation.
A fully analog ECG acquisition system was designed and implemented using wet electrodes and discrete electronic components. The project focused on low-noise analog front-end design, including amplification, band-pass filtering, and power-line interference mitigation. The system successfully extracted clinically interpretable ECG waveforms, providing strong practical grounding in biomedical instrumentation and analog signal conditioning.
This project addressed the challenge of unreliable ultrasound-based height measurements caused by dense scalp hair. The work included contributing to data clustering approaches to separate valid reflections from noise and designing an improved mechanical enclosure to enhance sensor alignment and measurement stability. The project formed part of the MeasureUP initiative and progressed toward a deployable clinical device.
This ongoing project involves the development of a motorized body-weight support system for rehabilitation applications. Contributions focus on designing the wearable interface, prioritizing patient comfort, load distribution, and adjustability. The work includes enclosure design and mold development aligned with industry standards, supporting safe and repeatable use in clinical rehabilitation settings.
This project explored mathematical and computational models of physiological processes, including cardiac electrophysiology, neuronal dynamics, and respiratory mechanics. Key components included dendritic tree approximations, Hodgkin–Huxley modeling, compartmental system analysis, and simulation-based evaluation of system behavior, strengthening foundations in physiology-informed modeling.
English (Professional proficiency), Sinhala (Native proficiency)
Python, MATLAB, C/C++
Digital Signal Processing (DSP), Time–Frequency Analysis, Adaptive Filtering, Statistical Signal Processing, Feature Extraction, Classical Machine Learning, Deep Learning (CNNs, RNNs/LSTMs)
Biomedical Signal Processing (ECG, EEG, PCG, EMG), Noise and Artifact Modeling, Multichannel Biosignal Analysis
Python Scientific Stack: NumPy, SciPy, Pandas, Matplotlib
Machine Learning: PyTorch, TensorFlow
Signal Processing: MATLAB Signal Processing Toolbox
Version Control: Git
Microsoft Office Suite, LaTeX