Assessment of Morphological Changes in H&E Stained Intestinal Crypts
Scientists at Genentech demonstrate that automated image analysis can increase the efficiency of scoring tasks that traditionally require expert pathologists - the scoring of crypt loss and infiltration in a mouse model of colitis.
- Scoring of mouse colon in DSS induced colitis in 800 animals
- Assessment of morphological changes in H&E stained crypts - a task that traditionally requires a specialized pathologist
- Infiltration cell count in corresponding IHC stained sections
- 20-fold decrease in manual labor - skilled (and expensive) pathologists can dedicate their time to other tasks
- 5-fold throughput increase: 500 slides/day
- Robust data, R2>0.7 correlation with manual assessment
- Continuous quantitative readout instead of discrete pathologist score captures multiple aspects of colitis and supports more comprehensive evaluation of drug candidates
- The study highlights the significant productivity gains that can be achieved
- It also shows that even complex image analysis tasks can successfully be automated:
Overall, we have used Definiens Developer to analyze images in a similar way to a pathologist's eye. The methods described here are highly applicable to many other morphology-based tissue scoring techniques that pathologists perform routinely on slides stained with H&E, for which few other effective automated analysis methods exist.
The study was conducted at Genentech and was presented by Cleopatra Kozlowski at the 2nd International Definiens Symposium.