Multi-lens microscopy efficiently analyzes specimens using overlapping sensor views. Its software accurately detects malaria and WBC counting features.
Integrating an illumination model into deep CNNs enables learning task-specific LED patterns, significantly improving fluorescence image inference from unstained microscopy.
In this work, we investigate an approach to jointly optimize multiple microscope settings, together with a classification network, for improved performance with such automated tasks.
Using reinforcement learning to discover the best way to illuminate microscope samples for faster, more accurate classification.
A method to enhance automated image classification speed and accuracy by co-optimizing microscope illumination and a deep neural network classification pipeline.
