Computational model for analyzing images of a biological specimen
Inventors
Jackson, Timothy • Holtz, Nevine • Edlund, Christoffer • Sjögren, Rickard
Assignees
Sartorius Stedim Data Analytics AB • Essen Bioscience Inc • Sartorius Bioanalytical Instruments Inc
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11803963-B2
Publication Date
2023-10-31
Expiration Date
Abstract
A method of analyzing images of a biological specimen using a computational model is described, the method including processing a cell image of the biological specimen and a phase contrast image of the biological specimen using the computational model to generate an output data. The cell image is a composite of a first brightfield image of the biological specimen at a first focal plane and a second brightfield image of the biological specimen at a second focal plane. The method also includes performing a comparison of the output data and a reference data and refining the computational model based on the comparison of the output data and the reference data. The method also includes thereafter processing additional image pairs according to the computational model to further refine the computational model based on comparisons of additional output data generated by the computational model to additional reference data.
Core Innovation
A method analyzes images of a biological specimen using a computational model. The method processes a cell image of the biological specimen and a phase contrast image of the biological specimen using the computational model to generate output data, wherein the cell image is a composite of a first brightfield image at a first focal plane and a second brightfield image at a second focal plane.
The method performs a comparison of the output data and reference data, and the computational model is refined based on the comparison. Thereafter, additional image pairs are processed according to the computational model to further refine the computational model based on comparisons of additional output data to additional reference data.
In described implementations, the computational model is applied to cell-by-cell analysis from label-free inputs, including a phase contrast image and a composite cell image derived from brightfield images at different focal planes. Output includes segmentation masks used for apoptosis time courses and supports label-free cell counting validation.
Claims Coverage
The document provides three independent claims, each centered on computational-model-based image analysis using a composite brightfield-derived cell image and a phase contrast image, with output compared to reference data and iterative model refinement using additional image pairs.
Composite cell image from two brightfield focal planes
Processing a cell image of the biological specimen and a phase contrast image of the biological specimen using the computational model to generate output data, wherein the cell image is a composite of a first brightfield image at a first focal plane and a second brightfield image at a second focal plane.
Comparison to reference data and refinement
Comparing the output data and reference data, and refining the computational model based on the comparison.
Iterative refinement using additional image pairs
Processing additional image pairs according to the computational model to further refine the computational model based on comparisons of additional output data to additional reference data.
Pixel-wise generation of cell image
Generating a cell image of the biological specimen by performing a pixel-wise mathematical operation upon the first brightfield image and the second brightfield image.
Assaying biological specimens with an optical microscope and processors
Capturing, via the optical microscope, a first brightfield image of a biological specimen at a first focal plane and a second brightfield image of the biological specimen at a second focal plane; and performing the subsequent comparison and refinement of the computational model based on output data and reference data, with further refinement using additional image pairs.
Across the independent claims, the core coverage is computational-model-based image analysis using a composite cell image from two brightfield focal planes together with a phase contrast image to generate output data, compare the output data with reference data, and refine the computational model, including further refinement using additional image pairs. One system claim further specifies generating the cell image via a pixel-wise mathematical operation.
Stated Advantages
Improved robustness when adding the cell image channel, as assessed using Box mAP and Mask mAP.
Documented Applications
Segmentation masks for apoptosis time courses.
Label-free cell counting validation.
Interested in licensing this patent?