In our most recent study, we demonstrate in silico that three-dimensional action potential waves can be reconstructed from ventricular muscle deformation that occurs in response to the action potential waves. We provide a proof-of-principle in computer simulations and anticipate that we will be able to make similar findings in isolated hearts ex vivo. We used a deep learning approach that is based on training a neural network with many thousands of examples of ventricular arrhythmias consisting of corresponding pairs of electrical action potential wave patterns and ventricular deformation, respectively. Our findings suggest that electrical arrhythmia circuits underlying heart rhyhtm disorders could be characterized non-invasively in patients using ultrasound imaging. We published the results on arXiv (Link). Congratulations and big thank you to Jan Lebert and Daniel Deng, who were the driving force behind this study. This study is Jan Lebert's 6th publication resulting from his doctoral work.