LJ McPherson

I’m an aspiring neuroengineer leveraging “cross-functional communication” and “four years of neuroscience research and programming experience” to innovate minimally invasive wireless devices. I’ve contributed to an interdisciplinary micropublication team by building a Python-based deep learning model for motor activity classification, grounded in rigorous research collaboration.

Right now, I’m “currently developing a pipeline using Python and Neurodesk” to build a tool for analyzing inferior colliculus anomalies, while also partnering on computational modeling for MRI imaging and STEM cell research. I’ve helped improve BCI therapies by designing and training a feature extraction model using Python, PyTorch, and Random Forest on large-scale ECoG datasets, refining adaptability through careful feature learning and signal-focused preprocessing.

Across technical projects and research work, I’ve paired strong experimentation with practical engineering: I enhanced T1 MRI readability through skull stripping and intensity normalization, and built classification pipelines using nnU-Net and MONAI. In NeuroMatch Academy, I attenuated noise and extracted relevant components from large ECoG datasets using MNE and SciPy with bandpass filtering, comparing linear (Spatial) vs. non-linear (CEBRA) filtering and maximizing SNR using GED-based spatial filtering—while maintaining organized workflows and using tools like Overleaf, Jupyter Notebooks, and GitHub to keep research moving forward.