Convolutional neural networks that teach microscopes how to image
Horstmeyer R, Chen RY, Kappes B, Judkewitz B (2017)
Publication Language: English
Publication Type: Other publication type
Publication year: 2017
URI: https://arxiv.org/abs/1709.07223
Open Access Link: https://arxiv.org/abs/1709.07223
Abstract
Deep learning algorithms offer a powerful means to automatically analyze the content of medical images. However, many biological samples of interest are primarily transparent to visible light and contain features that are difficult to resolve with a standard optical microscope. Here, we use a convolutional neural network (CNN) not only to classify images, but also to optimize the physical layout of the imaging device itself. We increase the classification accuracy of a microscope's recorded images by merging an optical model of image formation into the pipeline of a CNN. The resulting network simultaneously determines an ideal illumination arrangement to highlight important sample features during image acquisition, along with a set of convolutional weights to classify the detected images post-capture. We demonstrate our joint optimization technique with an experimental microscope configuration that automatically identifies malaria-infected cells with 5-10% higher accuracy than standard and alternative microscope lighting designs.
Authors with CRIS profile
Involved external institutions
United States (USA) (US)
Germany (DE)How to cite
APA:
Horstmeyer, R., Chen, R.Y., Kappes, B., & Judkewitz, B. (2017). Convolutional neural networks that teach microscopes how to image.
MLA:
Horstmeyer, Roarke, et al. Convolutional neural networks that teach microscopes how to image. 2017.
BibTeX: Download