Digital images of biopsy tissue can contain more than 14 billion pixels, making them about 287,000 times larger than images used to train the best AI image-recognition models to date.
Notes on verification
Multiple peer-reviewed sources support gigapixel-scale whole-slide biopsy imaging (10+ billion pixels), and the comparison to standard AI training image sizes (e.g., 224x224 or similar) is mathematically plausible, but the specific '287,000 times larger' figure is a derived calculation not directly stated in sources, so precision is unverified.
Sources
- Why it's so hard to use AI to diagnose cancer | MIT Technology Review (technologyreview.com)
- https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4197933/ (ncbi.nlm.nih.gov)
- https://doi.org/10.3390/diagnostics14050524 (doi.org)
- https://pmc.ncbi.nlm.nih.gov/articles/PMC9324951/ (pmc.ncbi.nlm.nih.gov)