FakET: Simulating Cryo-Electron Tomograms with Neural Style Transfer
This addresses a bottleneck in cryo-electron microscopy for researchers by providing a faster and more memory-efficient way to generate synthetic training data, though it is incremental as it builds on existing Neural Style Transfer techniques.
The paper tackles the slow generation of synthetic training data for cryo-electron microscopy by introducing FakET, a method using Neural Style Transfer to simulate microscope outputs, which matches benchmark performance while boosting data generation speed 750 times and reducing memory usage by 33 times.
In cryo-electron microscopy, accurate particle localization and classification are imperative. Recent deep learning solutions, though successful, require extensive training data sets. The protracted generation time of physics-based models, often employed to produce these data sets, limits their broad applicability. We introduce FakET, a method based on Neural Style Transfer, capable of simulating the forward operator of any cryo transmission electron microscope. It can be used to adapt a synthetic training data set according to reference data producing high-quality simulated micrographs or tilt-series. To assess the quality of our generated data, we used it to train a state-of-the-art localization and classification architecture and compared its performance with a counterpart trained on benchmark data. Remarkably, our technique matches the performance, boosts data generation speed 750 times, uses 33 times less memory, and scales well to typical transmission electron microscope detector sizes. It leverages GPU acceleration and parallel processing. The source code is available at https://github.com/paloha/faket.