I'm a recent Biomedical Engineering graduate from Toronto Metropolitan University, currently working as a Medventions Fellow
at Sunnybrook, where I focus on developing temporary pacing technology for patients at risk of arrhythmias.
Previously, I contributed to the development of the world’s first ML-based liver screening device at Oncoustics, and disease detection using microfluidics at iBEST.
• Performed 150+ hours of clinical observations in pre-operative, intra-operative, and post-operative patient care to identify 100 unmet clinical needs in interventional cardiology.
• Assessed the technical, clinical, IP, and regulatory feasibility and risk of addressing unmet needs with a commercial medtech device by conducting market and patent research, as well as literature reviews.
• Translated clinical needs into actionable user requirements and technical design inputs by conducting extensive consultations and interviews with 8+ key clinician stakeholders.
• Implemented the technical design specifications of an outpatient temporary pacemaker to reduce hospital length of stay and unnecessary permanent pacemaker implantations.
• Led the design and implementation of tools for automating data processing and analysis, supporting FDA clinical trials and R&D for an ML-driven liver screening device.
• Built a novel ultrasound artifact quantification algorithm that reduces manual data validation time by 75% and measures acoustic shadowing in liver ultrasound images with 85% accuracy, using Python and OpenCV.
• Developed a data processing pipeline to calculate and summarize scan quality metrics from raw ultrasound data, saving 8 hours of weekly manual work, using Python, Numpy, and Google Cloud API.
• Facilitated 510(k) submissions by performing linear regression analysis on tissue characterization datasets to establish statistical equivalence between the company's medical device and its predicate.
• Established data collection partnerships with 5 healthcare providers across North America, scaling FDA clinical trial operations by 200% to enable wider participant access and enriched data acquisition.
• Accelerated digital signal processing (DSP) runtime by 45% through the optimization of API calls and parallel processing with Python’s Google Cloud API and multiprocessing libraries.
• Implemented an interactive data visualization dashboard for real-time tracking of 25+ data collection KPIs to ensure alignment with clinical study objectives using SQL and Looker.
• Streamlined FDA compliance processes by drafting and proofreading 510(k) pre-submissions, device descriptions, and data protocols, accelerating approval timelines for the company’s medical device.
• Led a collaborative project with iBEST, St. Michael's Hospital, and The Hospital for Sick Children to investigate the role of extracellular vesciles on cost-effective approaches towards lung disease detection.
• Identified 3+ biphasic solutions for biomarker partitioning from blood samples, leading to more cost-effective and in-vitro approaches to detecting early-stage lung disease.
• Ensured laboratory safety by implementing continuous improvements in 3+ SOPs related to lab instrument operation, bacterial sample transportation, and decontamination procedures.
An image-guided software algorithm that converts photos of a user's hand into a 3D-printable, custom wrist brace. Check it out here.
A Discord bot which scrapes info from transfermarkt.us and displays statistical insights of 20,000+ football players/teams using BeautifulSoup and Discord.py. Check it out here.
The personal website you see here was built using CSS, JavaScript, and HTML! I'm planning to expand this website further using React. Feel free to refer to the GitHub page here.
A supervised ML model capable of classifying 3 distinct sleep stages with ~85% accuracy, using MATLAB's Classification Learner. Check it out here.
A working simulation model of a piezoelectric electromyogram sensor, programmed using MATLAB and C. Check it out here.