Resolution and contrast enhancement in weighted subtraction microscopy by deep learning

In subtraction microscopy, the negative sidelobes are inevitably generated by the difference between the en-velopes of Gaussian and doughnut point spread functions (PSFs), resulting in undesired information loss. There-fore, the trade-off between high resolution and information loss hinders further improvement in the performance of subtraction microscopy. Moreover, the postprocessing subtraction algorithms derived from PSF algebra tend to cause artifacts in dense samples. Herein, we propose an adaptive algorithm for assignment of the subtractive coefficient based on deep learning, termed Deep-IWS, to enhance the performance of subtraction microscopy. Both simulation and experiment reveal that Deep-IWS increases the resolution 1.8 times better than confocal microscopy, and significantly outperforms the previous subtraction microscopy. Furthermore, the reconstructed images also have fewer artifacts with a higher signal-to-noise ratio (SNR), demonstrating the validity and supe-riority of our method.

Automated summary

In subtraction microscopy, the difference between Gaussian and doughnut point spread functions (PSFs) leads to negative sidelobes and information loss. This trade-off hampers the improvement in performance. Our proposed Deep-IWS algorithm based on deep learning assigns subtractive coefficients adaptively, leading to 1.8 times higher resolution than confocal microscopy and fewer artifacts with higher SNR in reconstructed images.