Systems and methods for processing electronic images for computational assessment of disease

Abstract

Systems and methods are disclosed for receiving a digital image corresponding to a target specimen associated with a pathology category, wherein the digital image is an image of tissue specimen, determining a detection machine learning model, the detection machine learning model being generated by processing a plurality of training images to output a cancer qualification and further a cancer quantification if the cancer qualification is an confirmed cancer qualification, providing the digital image as an input to the detection machine learning model, receiving one of a pathological complete response (pCR) cancer qualification or a confirmed cancer quantification as an output from the detection machine learning model, and outputting the pCR cancer qualification or the confirmed cancer quantification.

1 - 20 . (canceled) 
     
     
         21 . A computer-implemented method for processing electronic images, the method comprising:
 receiving a digital image corresponding to a target specimen associated with a pathology category, wherein the digital image is an image of tissue specimen;   determining a detection machine learning model, the detection machine learning model being generated by processing a plurality of training images to output a cancer qualification and further output a cancer quantification if the cancer qualification is an confirmed cancer qualification;   providing the digital image as an input to the detection machine learning model;   receiving a confirmed cancer quantification comprising of a minimal residual disease (MRD) as an output from the detection machine learning model, wherein the detection machine learning model comprises a treatment effect machine learning model, and wherein the treatment effect machine learning model is initialized with one of weights and/or layers from a trained version of the detection machine learning model; and   outputting the MRD cancer qualification based on the treatment effects machine learning model.   
     
     
         22 . The computer-implemented method of  claim 21 , further including receiving one of a pathological complete response (pCR) cancer qualification. 
     
     
         23 . The computer-implemented method of  claim 21 , wherein the MRD cancer qualification is protocol specific. 
     
     
         24 . The computer-implemented method of  claim 21 , wherein the MRD cancer qualification corresponds to a number of cancer cells below a MRD threshold. 
     
     
         25 . The computer-implemented method of  claim 21 ,wherein the MRD cancer qualification identifies one or more diseases that remains occult within the patient, but may eventually lead to a relapse. 
     
     
         26 . The computer-implemented method of  claim 21 , wherein the treatment effect machine learning model is trained based on tagged treatment effects in the plurality of training images. 
     
     
         27 . The computer-implemented method of  claim 21 , wherein receiving the confirmed cancer quantification also comprises receiving a type of cancer when the output of the detection machine learning model comprises a confirmed cancer quantification 
     
     
         28 . The computer-implemented method of  claim 27 , wherein the type of cancer is determined based on the digital image and one or more of a tissue characteristics, slide type, glass type, tissue type, tissue region, chemical used, or stain amount. 
     
     
         29 . The computer-implemented method of claim  1 , wherein the digital image is from a pathology category, the pathology category selected from one or more of histology, cytology, frozen section, immunohistochemistry (IHC), immunofluorescence, hematoxylin and eosin (H&E), hematoxylin alone, molecular pathology, and/or 3D imaging. 
     
     
         30 . A system for processing electronic images, the system comprising:
 at least one memory storing instructions; and   at least one processor executing the instructions to perform operations comprising:
 receiving a digital image corresponding to a target specimen associated with a pathology category, wherein the digital image is an image of tissue specimen; 
 determining a detection machine learning model, the detection machine learning model being generated by processing a plurality of training images to output a cancer qualification and further output a cancer quantification if the cancer qualification is a confirmed cancer qualification; 
 providing the digital image as an input to the detection machine learning model; 
 receiving a confirmed cancer quantification comprising of a minimal residual disease (MRD) as an output from the detection machine learning model, wherein the detection machine learning model comprises a treatment effect machine learning model, and wherein the treatment effect machine learning model is initialized with one of weights and/or layers from a trained version of the detection machine learning model; and 
 outputting the MRD cancer qualification based on the treatment effects machine learning model. 
   
     
     
         31 . The system of  claim 30 , further including receiving one of a pathological complete response (pCR) cancer qualification. 
     
     
         32 . The system of  claim 30 , wherein the MRD cancer qualification is protocol specific. 
     
     
         33 . The system of  claim 30 , wherein the MRD cancer qualification corresponds to a number of cancer cells below a MRD threshold. 
     
     
         34 . The system of  claim 30 , wherein the MRD cancer qualification identifies one or more disease that remains occult within the patient, but may eventually lead to a relapse. 
     
     
         35 . The system of  claim 30 , wherein receiving the confirmed cancer quantification also comprises receiving a type of cancer when the output of the detection machine learning model comprises a confirmed cancer quantification. 
     
     
         36 . The system of  claim 35 , wherein the type of cancer is determined based on the digital image and one or more of a tissue characteristics, slide type, glass type, tissue type, tissue region, chemical used, or stain amount. 
     
     
         37 . The system of  claim 30 , wherein the digital image is from a pathology category, the pathology category selected from one or more of histology, cytology, frozen section, immunohistochemistry (IHC), immunofluorescence, hematoxylin and eosin (H&E), hematoxylin alone, molecular pathology, and/or 3D imaging. 
     
     
         38 . A non-transitory computer-readable medium storing instructions that, when executed by processor, cause the processor to perform operations for processing electronic images, the operations comprising:
 receiving a digital image corresponding to a target specimen associated with a pathology category, wherein the digital image is an image of tissue specimen;   determining a detection machine learning model, the detection machine learning model being generated by processing a plurality of training images to output a cancer qualification and further output a cancer quantification if the cancer qualification is a confirmed cancer qualification;   providing the digital image as an input to the detection machine learning model;   receiving a confirmed cancer quantification comprising of a minimal residual disease (MRD) as an output from the detection machine learning model, wherein the detection machine learning model comprises a treatment effect machine learning model, and wherein the treatment effect machine learning model is initialized with one of weights and/or layers from a trained version of the detection machine learning model; and   outputting the MRD cancer qualification based on the treatment effects machine learning model.   
     
     
         39 . The non-transitory computer-readable medium of  claim 38 , further including receiving one of a pathological complete response (pCR) cancer qualification. 
     
     
         40 . The non-transitory computer-readable medium of  claim 38 , wherein the MRD cancer qualification is protocol specific.