Deep learning driven label-free image guided automatic patch clamp system for human and rodent in vitro slice physiology
Krisztián Koós1, Gáspár Oláh2, Tamás Balassa1, Norbert Mihut2, Márton Rózsa2, Attila Ozsvár2, Ervin Tasnádi1, József Molnár1, Pál Barzó3, Gábor Molnár2, Gábor Tamás2, Peter Horvath1, 4
Patch clamp recording of neurons is labor intensive and time consuming procedure. We have developed a tool that fully automatically performs electrophysiological recordings in label-free tissue slices. The automation covers the detection of cells in label-free images based on deep learning, calibration of the micropipette, approach to the cell with the pipette, formation of the whole-cell configuration, and the recording. The model was trained on a new image database of brain tissue slices. The pipette tip detection and approaching phase uses image analysis techniques for precise movements. Approaching trajectory was modulated by the system dynamically in order to avoid untargeted tissue components. High quality measurements were performed on hundreds of human and rodent neurons. We also demonstrate that further molecular and anatomical analysis can be performed on the recorded cells. Our tool can multiply the number of daily measurements to help brain research.