Methods and systems to determine cancer molecular subtypes based on ultrasound and/or optoacoustic (oa/us) features

Abstract

Methods, devices and systems are provided that utilize one or more processors in connection with, receiving OA/US feature scores in connection with OA/US images collected from a patient examination for a volume of interest. The methods, devices and systems apply the OA/US feature scores to a feature score to molecular subtype (FSMS) model. The methods, devices and systems determine, from the FSMS model, an indication of at least one of a molecular subtype or histologic grade of a pathology experienced by the patient.

1 . A method, comprising:
 utilizing one or more processors in connection with,   receiving OA/US feature scores in connection with OA/US images collected from a patient examination for a volume of interest;   applying the OA/US feature scores to a feature score to molecular subtype (FSMS) model; and   determining, from the FSMS model, an indication of at least one of a molecular subtype or histologic grade of a pathology experienced by the patient.   
     
     
         2 . The method of  claim 1 , wherein the pathology represents breast cancer and the molecular subtype represents one or more of Luminal A (LumA), Luminal B (LumB), Triple-negative (TRN) and HER2 amplified (HER2+). 
     
     
         3 . The method of  claim 1 , wherein the OA/US feature scores include at least one of:
 a) multiple US feature scores only, and no OA feature scores;   b) multiple OA feature scores only and no US feature scores; or   c) at least one US feature score and at least one OA feature score.   
     
     
         4 . The method of  claim 1 , wherein the FSMS model defines a correlation between one or more of the OA/US feature scores and at least one of one or more molecular subtypes or one or more histologic grades. 
     
     
         5 . The method of  claim 1 , wherein the FSMS model comprises a table associating pairs of molecular subtypes and the OA/US features scores, the table contains a correlation index indicative of an extent to which the corresponding OA/US feature scores differentiate between the corresponding pair of the molecular subtypes. 
     
     
         6 . The method of  claim 1 , wherein the OA/US feature scores include at least one of a US or OA boundary zone and at least one of a US or OA peripheral zone feature score. 
     
     
         7 . The method of  claim 1 , wherein the OA/US feature scores include at least one of a US or OA boundary zone feature score and at least one US/OA internal or peripheral feature score from the following: US internal zone shape feature score, US internal zone echotexture feature score, US internal zone sound transmission feature score, US peripheral zone feature score, OA internal deoxygenated blood feature score, OA internal total hemoglobin feature score, or OA peripheral zone feature score. 
     
     
         9 . The method of  claim 1 , further comprising displaying the indication as a collection of predictive results representative of probabilities of malignancy (POM) associated with a collection of the molecular subtypes and/or histologic grades. 
     
     
         10 . The method of  claim 1 , wherein the receiving, applying and determining are performed in connection with a combination of a US data set, OA data set, US images, OA images, US feature scores, and OA feature scores. 
     
     
         11 . A system, comprising:
 memory configured to store program instructions and a feature score to molecular subtype (FSMS) model;   one or more processors that, when executing the program instructions, or configured to:   receive the OA/US feature scores comprises in connection with OA/US images collected from a patient examination for a volume of interest;   apply the OA/US feature scores to the FSMS model; and   determine, from the FSMS model, an indication of at least one of a molecular subtype or histologic grade of a pathology experienced by the patient.   
     
     
         12 . The system of  claim 11 , wherein the pathology represents breast cancer and the memory is configured to store molecular subtype represents one or more of Luminal A (LumA), Luminal B (LumB), Triple-negative (TRN) and HER2 amplified (HER2+). 
     
     
         13 . The system of  claim 11 , further comprising a display configured to present a probability of malignancy (POM) indicia in a manner and format representative of a collection of probabilities associated with a collection of at least one of the molecular subtypes or histologic grades. 
     
     
         14 . The system of  claim 13 , wherein the display is further configured to display the POM indicia to include at least one of a graph, alphanumeric characters, or color-coded scale, the POM indicia noting a central point/mean, and confidence intervals for the corresponding at least one of molecular subtypes or histologic grades. 
     
     
         15 . The system of  claim 11 , wherein the OA/US feature scores include at least one of:
 a) multiple US feature scores only, and no OA feature scores;   b) multiple OA feature scores only and no US feature scores; or   c) at least one US feature score and at least one OA feature score.   
     
     
         16 . The system of  claim 11 , wherein the FSMS model defines a correlation between one or more of the OA/US feature scores and at least one of one or more molecular subtypes or one or more histologic grades. 
     
     
         17 . The system of  claim 11 , wherein the FSMS model comprises a table associating pairs of molecular subtypes and the OA/US features scores, the table contains a correlation index indicative of an extent to which the corresponding OA/US feature scores differentiate between the corresponding pair of the molecular subtypes. 
     
     
         18 . The system of  claim 11 , wherein the OA/US feature scores include at least one of a US or OA boundary zone and at least one of a US or OA peripheral zone feature score. 
     
     
         19 . The system of  claim 11 , wherein the OA/US feature scores include at least one of a US or OA boundary zone feature score and at least one US/OA internal or peripheral feature score from the following: US internal zone shape feature score, US internal zone echotexture feature score, US internal zone sound transmission feature score, US peripheral zone feature score, OA internal deoxygenated blood feature score, OA internal total hemoglobin feature score, or OA peripheral zone feature score.