Systems and methods to process electronic images to provide image-based cell group targeting
Systems and methods are disclosed for grouping cells in a slide image that share a similar target, comprising receiving a digital pathology image corresponding to a tissue specimen, applying a trained machine learning system to the digital pathology image, the trained machine learning system being trained to predict at least one target difference across the tissue specimen, and determining, using the trained machine learning system, one or more predicted clusters, each of the predicted clusters corresponding to a subportion of the tissue specimen associated with a target.
1. A method of targeted sampling, comprising:
receiving one or more patient-specific medical images of a tissue sample of a patient;
determining, by providing the patient-specific medical images to a machine learning system, one or more target regions in the tissue sample, each target region being associated with a target, each target being associated with a different clinically relevant tissue feature;
determining, using the machine learning system, one or more optimal tissue sampling locations within each determined target region of the tissue sample to maximize information gained about the tissue sample; and
providing the one or more target regions and the one or more optimal tissue sampling locations for output to a display.
2. The method of claim 1 , further comprising:
receiving sequencing data of the tissue sample, the sequencing data indicating a target composition of one or more samples at the one or more optimal tissue sampling locations; and
determining, using the machine learning system and based on the sequencing data, a spatial distribution of the target across each of the target regions.
3. The method of claim 1 , wherein the one or more optimal tissue sampling locations are determined using segmentation and a clustering heuristic.
4. The method of claim 1 , wherein the one or more optimal tissue sampling locations comprises optimal genetic sampling locations.
5. The method of claim 1 , wherein each of the target regions corresponds to a different biomarker cluster.
6. The method of claim 1 , further comprising:
determining, using the one or more target regions and the machine learning system, a mapping of each of the target regions to a biological target; and
providing the biological target for output to the display.
7. The method of claim 1 , further comprising:
determining, using the one or more target regions and the machine learning system, a mapping of each of the target regions to a predicted mutation; and
providing the predicted mutation for output to the display.
8. The method of claim 1 , further comprising:
determining, using the machine learning system, one or more pixel masks, each pixel mask corresponding to one of the target regions; and
providing the pixels masks for output to the display.
9. The method of claim 1 , further comprising:
determining, based on a spatial distribution and spatial relation of the target regions, a treatment decision associated with the patient.
10. The method of claim 1 , further comprising:
determining, using flow cytometry and/or mass spectrometry techniques, one or more physical and/or chemical characteristics of one or more samples from the one or more determined optimal sampling locations.
11. The method of claim 1 , wherein determining the one or more optimal tissue sampling locations is configured to maximize information gained about (i) heterogeneity across the tissue sample and/or (ii) one or more target differences across the tissue sample.
12. A system, comprising:
at least one memory storing instructions; and
at least one processor configured to execute the instructions to perform operations comprising:
receiving one or more patient-specific medical images of a tissue sample of a patient;
determining, by providing the patient-specific medical images to a machine learning system, one or more target regions in the tissue sample, each target region being associated with a target, each target being associated with a different clinically relevant tissue feature;
determining, using the machine learning system, one or more optimal tissue sampling locations within each determined target region of the tissue sample to maximize information gained about the tissue sample; and
providing the one or more target regions and the one or more optimal tissue sampling locations for output to a display.
13. The system of claim 12 , the operations further comprising:
receiving sequencing data of the tissue sample, the sequencing data having been collected based on the target regions; and
determining, using the machine learning system and based on the sequencing data, a spatial distribution of the target across each of the target regions.
14. The system of claim 12 , wherein the one or more optimal tissue sampling locations are determined using segmentation and a clustering heuristic.
15. The system of claim 12 , wherein each of the target regions corresponds to a different biomarker cluster.
16. The system of claim 12 , the operations further comprising:
determining, using the one or more target regions and the machine learning system, a mapping of each of the target regions to a biological target; and
providing the biological target for output to the display.
17. The system of claim 12 , the operations further comprising:
determining, using the one or more target regions and the machine learning system, a mapping of each of the target regions to a predicted mutation; and
providing the predicted mutation for output to the display.
18. The system of claim 12 , the operations further comprising:
determining, using the machine learning system, one or more pixel masks, each pixel mask corresponding to one of the target regions; and
providing the pixels masks for output to the display.
19. The system of claim 12 , the operations further comprising:
determining, based on a spatial distribution and spatial relation of the target regions, a treatment decision associated with the patient.
20. A non-transitory computer-readable medium storing instructions that, when executed by a processor, cause the processor to perform operations, the operations comprising:
receiving one or more patient-specific medical images of a tissue sample of a patient;
determining, by providing the patient-specific medical images to a machine learning system, one or more target regions in the tissue sample, each target region being associated with a target, each target being associated with a different clinically relevant tissue feature;
determining, using the machine learning system, one or more optimal tissue sampling locations within each determined target region of the tissue sample to maximize information gained about the tissue sample; and
providing the one or more target regions and the one or more optimal tissue sampling locations for output to a display.