Stow, Ohio, United States
I have extensive expertise in analyzing biosignals from digital devices, including smartwatches, smartphones, and wearable sensors, to develop objective measures of motor symptoms in neurodegenerative diseases. My research focuses on designing, applying, and validating AI and signal processing algorithms to quantify motor symptoms in Parkinson’s disease, such as tremor and slowness, as well as medication side effects such as dyskinesia. I have also built models to identify candidates for advanced therapies such as deep brain stimulation using multimodal wearable sensor data. In addition, I developed unsupervised models to quantify tremor severity and its impact on quality of life in essential tremor, independently of clinical ratings and patient-reported outcomes. My work lies at the intersection of signal processing, AI, and digital health. I hold a Ph.D. in Biomedical Engineering, where my dissertation focused on blood pressure regulation and postural control in Parkinson’s disease, aiming to better understand balance mechanisms and prevent falls. Through this work, I gained expertise in study design, clinical data collection, physiological signal analysis, and machine learning, integrating engineering with human physiology to address unmet clinical needs. In my current research, I lead the development of objective measures of motor symptoms and digital biomarkers from wearable sensor data, including motion, physiological, and behavioral, to support clinical trials and improve patient care in neurodegenerative diseases. I am proficient in Python, MATLAB, and R.
• Market research analysis to identify new product opportunities. • Early medical devices research, design, and development. • Build cardiovascular bench-top models, collect, analyze and interpret physiological data. • Work and collaborate with different research teams on various active projects.
• Tested equipment and developed user interface to synchronize different medical devices for data collection. • Performed transcranial magnetic stimulation to study fatigue in Parkinson’s disease patients. • Collected clinical data from Parkinson’s patients and analyzed physiological signals.
Project 1: Blood pressure and postural control in Parkinson’s disease • Designed experiment to study the effect of lower limb muscles on blood pressure and postural balance. • Tested equipment, recorded blood pressure, center of pressure, electrocardiogram and electromyogram. • Analyzed the data using MATLAB, and R, interpreted and documented the study’s outcomes. Project 2: Design and evaluation of a dynamic aircushion • Helped develop an Arduino-based system to control inflation/deflation cycles of aircushion cells. • Designed experiment and recorded tissue oxygen saturation, skin blood flow, and interface pressure. • Investigated the effect of the designed dynamic aircushion to relieve pressure and enhance perfusion. • Analyzed the collected signals using MATLAB and R and reported the findings. Project 3: Early detection of Parkinson’s disease • Designed experiment and recorded postural sway data from Parkinson’s disease (PD) patients and healthy controls (HC). • Analyzed sway patterns and extracted meaningful features using python and R. • Developed and implemented supervised machine learning models to differentiate between PD patients and HC. Project 4: Early detection of breast microcalcifications • Utilized python to implement supervised machine learning techniques for detecting breast microcalcifications on mammograms. • Analyzed and extracted relevant features from each mammogram using python and R. • Developed a Random Forest based model which was able to detect breast macrocalcifications on mammographic images with 95% accuracy.
Developed image processing algorithms to detect breast microcalcifications.