Accurate and reproducible invasive breast cancer detection in whole-slide images: A Deep Learning approach for quantifying tumor extent
Problem
Manually identifying the presence and extent of invasive breast cancer in whole-slide pathology images is tedious and subject to inter- and intra-reader variability.
Why it matters
Tumor extent assessment is critical for patient staging and treatment response decisions.
What the AI did
A convolutional neural network automatically detects the extent of invasive tumor on digitized whole-slide images, evaluated for accuracy and robustness across scanners and protocols.
Results
not stated in source
Impact
Demonstrates feasibility of automated tumor-extent detection as a pathologist decision-support tool.
Record
| Venue | Scientific Reports |
|---|---|
| Published | 2017 |
| Work type | article |
| Authors | Ángel Cruz-Roa, Hannah L. Gilmore, Ajay N. Basavanhally, Michael D. Feldman, Shridar Ganesan, Natalie N. C. Shih, John Tomaszewski, Fabio Augusto Gonzalez et al. |
| Citations | 532 (OpenAlex, at import time) |
| References | 67 |
| Model or method | convolutional neural network (not named) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1038/srep46450 |
| Open access | gold |
| Keywords | breast cancer, deep learning, whole-slide images, digital pathology, convolutional neural network |
Importance rating
significant — thousands to millions affected
Abstract states tumor extent identification is critical for patient management, staging and treatment response; manual process tedious with reader variability. No disease scale given.