Image analysis method and system

Abstract

A system and computer-implemented method for diagnosing, monitoring, screening for or evaluating a musculoskeletal disease or condition of a subject. The method comprises quantifying one or more features segmented and identified from a medical image of the subject; assessing the quantified one or more features with a trained machine learning model trained to diagnose, monitor, screen for or evaluate one or more musculoskeletal diseases or conditions; and outputting one or more results of the assessing, the results comprising one or more disease or condition diagnoses, disease or condition monitorings, disease or condition screenings, or disease or condition evaluations.

1 . A computer-implemented method of diagnosing, monitoring, screening for or evaluating a musculoskeletal disease or condition of a subject, the method comprising:
 quantifying one or more features segmented and identified from a medical image of the subject;   assessing the quantified one or more features with a trained machine learning model trained to diagnose, monitor, screen for or evaluate one or more musculoskeletal diseases or conditions; and   outputting one or more results of the assessing, the results comprising one or more disease or condition diagnoses, disease or condition monitorings, disease or condition screenings, or disease or condition evaluations.   
     
     
         2 . A method as claimed in  claim 1 , wherein the medical image is an image of a two-dimensional region or three-dimensional volume of the subject. 
     
     
         3 . A method as claimed in  claim 1 , wherein the trained machine learning model is a model trained with training data that comprises (i) subject image data or (ii) subject image data and subject non-image data. 
     
     
         4 . A method as claimed in  claim 1 , comprising capturing the medical image of the subject with a scanner. 
     
     
         5 . A method as claimed in  claim 1 , comprising diagnosing, monitoring or evaluating bone mineral density, marrow adiposity and/or mineralization at one or more anatomical sites. 
     
     
         6 . A method as claimed in  claim 1 , wherein the trained machine learning model 
 (a) is a disease classification model;   (b) is a model trained using features extracted from patient data and labels or annotations indicating disease or non-disease; and/or   (c) comprises a deep learning neural network or other machine learning algorithms.   
     
     
         7 . A method as claimed in  claim 1 , wherein the results comprise (i) one or more disease classifications; (ii) one or more disease probabilities; and/or (iii) one or more fracture risk scores. 
     
     
         8 . A method as claimed in  claim 1 , wherein the disease or condition is arthritis, bone fracture, or osteomalacia. 
     
     
         9 . A system for diagnosing, monitoring, screening for or evaluating a musculoskeletal disease or condition, the system comprising:
 a feature quantifier configured to quantify one or more features segmented and identified from a medical image of a subject;   a feature assessor configured for assessing the quantified one or more features with a trained machine learning model trained to diagnose, monitor, screen for or evaluate one or more musculoskeletal diseases or conditions; and   an output configured to output one or more results of the assessing, the results comprising one or more disease or condition diagnoses, disease or condition monitoring, disease or condition screening and disease or condition evaluations.   
     
     
         10 . A system as claimed in  claim 9 , wherein the medical image is an image of a two-dimensional region or three-dimensional volume of the subject. 
     
     
         11 . A system as claimed in  claim 9 , wherein the trained machine learning model is a model trained with training data that comprises (i) subject image data or (ii) subject image data and subject non-image data. 
     
     
         12 . A system as claimed in  claim 9 , comprising a scanner operable to capture the medical image of the subject. 
     
     
         13 . A system as claimed in  claim 9 , configured to diagnose, monitor or evaluating bone mineral density, marrow adiposity and/or mineralization at one or more anatomical sites. 
     
     
         14 . A system as claimed in  claim 9 , wherein the trained machine learning model 
 (a) is a disease classification model;   (b) is a model trained using features extracted from patient data and labels or annotations indicating disease or non-disease; and/or   (c) comprises a deep learning neural network or other machine learning algorithms.   
     
     
         15 . A system as claimed in  claim 9 , wherein the results comprise (i) one or more disease classifications; (ii) one or more disease probabilities; and/or (iii) one or more fracture risk scores. 
     
     
         16 . A system as claimed in  claim 9 , wherein the disease or condition is arthritis, bone fracture, or osteomalacia. 
     
     
         17 . A non-transitory computer-readable medium comprising computer program code, wherein the computer program code comprises instructions configured, when executed by one or more computing devices, to implement a method of diagnosing, monitoring, screening for or evaluating a musculoskeletal disease or condition of a subject, the method comprising:
 quantifying one or more features segmented and identified from a medical image of the subject;   assessing the quantified one or more features with a trained machine learning model trained to diagnose, monitor, screen for or evaluate one or more musculoskeletal diseases or conditions; and   outputting one or more results of the assessing, the results comprising one or more disease or condition diagnoses, disease or condition monitorings, disease or condition screenings, or disease or condition evaluations.   
     
     
         18 . The non-transitory computer-readable medium of  claim 17 , wherein the medical image is an image of a two-dimensional region or three-dimensional volume of the subject. 
     
     
         19 . The non-transitory computer-readable medium of  claim 17 , wherein the trained machine learning model is a model trained with training data that comprises (i) subject image data or (ii) subject image data and subject non-image data. 
     
     
         20 . The non-transitory computer-readable medium of  claim 17 , wherein:
 the method comprises diagnosing, monitoring or evaluating bone mineral density, marrow adiposity and/or mineralization at one or more anatomical sites; and/or   the results comprise (i) one or more disease classifications, (ii) one or more disease probabilities, and/or (iii) one or more fracture risk scores; and /or   the disease or condition is arthritis, bone fracture, or osteomalacia.