BlurryScope enables compact, cost-effective scanning microscopy for HER2 scoring using deep learning on blurry images

Abstract

We developed a rapid scanning optical microscope, termed "BlurryScope", that leverages continuous image acquisition and deep learning to provide a cost-effective and compact solution for automated inspection and analysis of tissue sections. This device offers comparable speed to commercial digital pathology scanners, but at a significantly lower price point and smaller size/weight. Using BlurryScope, we implemented automated classification of human epidermal growth factor receptor 2 (HER2) scores on motion-blurred images of immunohistochemically (IHC) stained breast tissue sections, achieving concordant results with those obtained from a high-end digital scanning microscope. Using a test set of 284 unique patient cores, we achieved testing accuracies of 79.3% and 89.7% for 4-class (0, 1+, 2+, 3+) and 2-class (0/1+, 2+/3+) HER2 classification, respectively. BlurryScope automates the entire workflow, from image scanning to stitching and cropping, as well as HER2 score classification.

22 Pages, 5 Figures, 1 Table

References in corpus (16)

• All-Optical Machine Learning Using Diffractive Deep Neural Networks
• Deep Learning Microscopy
• PhaseStain: Digital staining of label-free quantitative phase microscopy images using deep learning
• Terahertz Pulse Shaping Using Diffractive Surfaces
• Deep learning enhanced mobile-phone microscopy
• Phase Imaging with Computational Specificity (PICS) for measuring dry mass changes in sub-cellular compartments
• Fourier Imager Network (FIN): A deep neural network for hologram reconstruction with superior external generalization
• Super-resolution image display using diffractive decoders
• Illumination Pattern Design with Deep Learning for Single-Shot Fourier Ptychographic Microscopy
• Unidirectional Imaging using Deep Learning-Designed Materials
• Information hiding cameras: optical concealment of object information into ordinary images
• Automated HER2 Scoring in Breast Cancer Images Using Deep Learning and Pyramid Sampling
• All-Optical Phase Conjugation Using Diffractive Wavefront Processing
• Increasing a microscope's effective field of view via overlapped imaging and machine learning
• Neural Network-Based Processing and Reconstruction of Compromised Biophotonic Image Data
• Quantifying myelin in brain tissue using color spatial light interference microscopy (cSLIM)