Research

I currently work on applications of machine learning to neural signal processing, behavioral neuroscience, and biomedical imaging. Examples of my work in these areas include

• Domain-generalized deep learning for detecting neuronal spikes in whole-cell patch-clamp postsynaptic current traces

• Forecasting lab animal behavioral events with computer vision and sequence modeling

• Predicting neuroanatomical sources of detailed electrophysiology spike traces

• Benchmarking of self-supervised learning methods for mammograms

In my neural data science research, I have noticed two main engineering and informatics bottlenecks: (i) parallel acquistition and modulation of multiscale neural activity; (ii) integration of many modalities of neuroanatomical and neurophysiological data. In order to contribute to our accelerated ability to understand the mammalian nervous system, I aim to develop technologies for recording, modulating, and understanding large-scale neural activity on fast timescales. On the recording and modulating sides, I am especially interested in utilizing optical and acoustic metamaterials for minimally-invasive in vivo imaging and control of nerve activity – especially deep brain structures – and implementing image and signal processing libraries for these application areas. However, to really understand underlying physiology, we must correlate these various signals with each other and with manifestations of behaviors; I will utilize interpretable machine learning to discover relationships between different neural signals, such as spatial transcriptomics and calcium dynamics.