Apparatuses and methods for digital pathology
Digital pathology is a promising alternative to manual slide analysis, but improvements in imaging and analysis methods are needed to provide an analysis data set of images that can easily be handled, stored, and importantly, delivered from one data location to another. Methods and software are described herein to improve image collection and analysis.
1. A method for generating image data for a set of stained slides that comprises a tissue sample, the method comprising:
selecting a plurality of sub-regions from one of the set of stained slides;
identifying, for each imaging channel of a first plurality of imaging channels, either a first number of variations of contrast thresholds, a first number of variations of thresholds of marker positivity, or a first number of variations of contrast thresholds and a first number of variations of thresholds of marker positivity, wherein each imaging channel is associated with a corresponding biomarker or plurality of biomarkers;
generating a plurality of masks for each of the plurality of sub-regions corresponding to each combination of the variations of contrast thresholds and variations of thresholds of marker positivity for each imaging channel of the first plurality of imaging channels, wherein each mask is segmented to identify cells;
ranking the plurality of masks based on an ability of each of the plurality of masks to accurately identify cells that are positive for the corresponding biomarker or plurality of biomarkers of each imaging channel of the first plurality of imaging channels;
selecting a set of final mask parameters from the plurality of masks based on the rankings; and
generating a set of slide images for all slides of the set of stained slides using the set of final mask parameters.
2. The method of claim 1 , wherein selecting the plurality of sub-regions from one of the set of stained slides comprises selecting the sub-regions so that at least one of the sub-regions includes a structure that is substantially free of cells.
3. The method of claim 1 , wherein identifying further comprises identifying a number of variations of image acquisition parameters for each imaging channel of the first plurality of imaging channels.
4. The method of claim 1 , further comprising selecting a plurality of data compression values.
5. The method of claim 1 , wherein the ranking is performed by a trained network.
6. The method of claim 1 , further comprising iterating the steps of identifying and generating for different imaging channels.
7. The method of claim 1 , further comprising determining, from the set of stained slides, statistics on labeled cells for a second plurality of imaging channels that is larger than the first plurality of imaging channels.
8. The method of claim 1 , wherein selecting the plurality of sub-regions from one of the set of stained slides comprises selecting a different size for each of the sub-regions.
9. The method of claim 1 , wherein each mask is segmented to identify cells using the variations of contrast thresholds and the variations of thresholds of marker positivity.
10. A method for generating image data for a set of stained slides that comprises a tissue sample, the method comprising:
selecting a plurality of sub-regions from one of the set of stained slides, wherein at least one of the sub-regions includes a structure that is substantially free of cells;
identifying, for each imaging channel of a first plurality of imaging channels: a first number of variations of contrast thresholds, a first number of variations of thresholds of marker positivity, or a first number of variations of contrast thresholds and a first number of variations of thresholds of marker positivity, wherein each imaging channel is associated with a corresponding biomarker or plurality of biomarkers;
generating a plurality of masks comprising a mask for each of the plurality of sub-regions corresponding to each combination of the variations of contrast thresholds and the variations of thresholds of marker positivity for each imaging channel of the first plurality of imaging channels, wherein each mask is segmented to identify cells using the variations of contrast thresholds and the variations of thresholds of marker positivity;
ranking the plurality of masks based on an ability of each of the plurality of masks to accurately identify cells that are positive for the corresponding biomarker or plurality of biomarkers of each imaging channel of the first plurality of imaging channels;
selecting a set of final mask parameters from the plurality of masks based on the rankings; and
generating a set of slide images for all slides of the set of stained slides using the set of final mask parameters.
11. A system for generating image data for a set of stained slides that comprises a tissue sample, the system comprising:
one or more processors;
memory coupled to the one or more processors, the memory configured to store computer-program instructions, that, when executed by the one or more processors, perform a computer-implemented method comprising:
selecting a plurality of sub-regions from one of the set of stained slides;
identifying, for each imaging channel of a first plurality of imaging channels, either a first number of variations of contrast thresholds, a first number of variations of thresholds of marker positivity, or a first number of variations of contrast thresholds and a first number of variations of thresholds of marker positivity, wherein each imaging channel is associated with a corresponding biomarker or plurality of biomarkers;
generating a plurality of masks for each of the plurality of sub-regions corresponding to each combination of the variations of contrast thresholds and variations of thresholds of marker positivity for each imaging channel of the first plurality of imaging channels, wherein each mask is segmented to identify cells;
ranking the plurality of masks based on an ability of each of the plurality of masks to accurately identify cells that are positive for the corresponding biomarker or plurality of biomarkers of each imaging channel of the first plurality of imaging channels;
selecting a set of final mask parameters from the plurality of masks based on the rankings; and
generating a set of slide images for all slides of the set of stained slides using the set of final mask parameters.
12. The system of claim 11 , wherein selecting the plurality of sub-regions from one of the set of stained slides comprises selecting each of the sub-regions so that each sub-region includes a structure that is substantially free of cells.
13. The system of claim 11 , wherein identifying further comprises identifying a number of variations of image acquisition parameters for each imaging channel of the first plurality of imaging channels.
14. The system of claim 11 , further comprising selecting a plurality of data compression values.
15. The system of claim 11 , wherein the ranking is performed by a trained network.
16. The system of claim 11 , further comprising iterating the steps of identifying and generating for different imaging channels.
17. The system of claim 11 , further comprising determining, from the set of stained slides, statistics on labeled cells for a second plurality of imaging channels that is larger than the first plurality of imaging channels.
18. The system of claim 11 , wherein selecting the plurality of sub-regions from one of the set of stained slides comprises selecting a different size for each of the sub-regions.
19. The system of claim 11 , wherein each mask is segmented to identify cells using the variations of contrast thresholds and the variations of thresholds of marker positivity.