Intelligent Nanoelectronics

There remains a pressing need for electrophysiology techniques that are capable of accurate and high throughput recording of electrical signals for applications in disease modeling and drug screening. Extracellular microelectrode arrays can record signals from hundreds of cells in a non-invasive fashion and are successfully used in live animals and humans. However, extracellular electrodes can only detect the firing patterns- namely the presence and frequency of an action potential (AP), missing critical information about their true waveforms. To obtain the true waveform of an AP, the electrodes must obtain “intracellular access” by penetrating the cell membrane. This intracellular recording provides a depth of essential information including ion channel activities, cell subtype, and effects of drugs on the AP shape. However, intracellular recordings are more invasive, lower-throughput, and not currently possible in vivo. <b>Critical Need:</b> A method that could reliably reconstruct the true waveform of APs merely from non-invasive and high-throughput extracellular recordings would be of great interest to the cardiac electrophysiology field. In this work we present an intelligent nanoelectrode that can reconstruct intracellular signals from extracellular recordings. This is interdisciplinary research project that leverages the developments from two fields: 1) <b>nanoelectrode arrays</b> with the promise to simultaneously record intra- and extra- cellular signals from thousands of interconnected cardiac cells for <b>single cell electrophysiology</b> and 2) Advancements in AI, especially <b>physics guided learning</b> that allow fast and precise signal reconstruction. We believe our new technique will dramatically accelerate new discoveries in the cardiac biophysical and biomedical research fields.