Alignment of electron optical beam shaping elements using a convolutional neural network

•A deep convolutional neural network is used to evaluate and correct the alignment of an orbital angular momentum sorter.•The convolutional neural network automatically analyses OAM spectra at a rate of 50 ms/spectrum, allowing for real time implementation.•The method is easily transferable to any programmable beam shaping technique. A convolutional neural network is used to align an orbital angular momentum sorter in a transmission electron microscope. The method is demonstrated using simulations and experiments. As a result of its accuracy and speed, it offers the possibility of real-time tuning of other electron optical devices and electron beam shaping configurations.