Towards an Intelligent Microscope: adaptively learned illumination for optimal sample classification

TL;DR

This paper introduces a reinforcement learning system that adaptively optimizes microscope illumination patterns to improve sample classification accuracy while minimizing image acquisition costs.

Contribution

It presents a novel reinforcement learning approach that learns optimal illumination strategies for microscopes to enhance classification performance.

Findings

Adaptive illumination improves classification accuracy.

Reinforcement learning balances image quality and acquisition cost.

System outperforms naive illumination methods.

Abstract

Recent machine learning techniques have dramatically changed how we process digital images. However, the way in which we capture images is still largely driven by human intuition and experience. This restriction is in part due to the many available degrees of freedom that alter the image acquisition process (lens focus, exposure, filtering, etc). Here we focus on one such degree of freedom - illumination within a microscope - which can drastically alter information captured by the image sensor. We present a reinforcement learning system that adaptively explores optimal patterns to illuminate specimens for immediate classification. The agent uses a recurrent latent space to encode a large set of variably-illuminated samples and illumination patterns. We train our agent using a reward that balances classification confidence with image acquisition cost. By synthesizing knowledge over multiple snapshots, the agent can classify on the basis of all previous images with higher accuracy than from naively illuminated images, thus demonstrating a smarter way to physically capture task-specific information.