Method and system for ultrasound based automated detection, quantification and tracking of pathologies

An automated system for detecting a disease state, the system comprising:

• a detection platform, comprising:
• a bone surface identifying module configured to identify a bone surface in one or more image data sets, wherein the one or more image data sets correspond to a region of interest in an object of interest, and wherein for each of the one or more image data sets the bone surface identifying module is configured to:
• obtain a first image corresponding to the region of interest in a normal subject;
• obtain a second image corresponding to the region of interest in the object of interest;
• identify an acoustic shadow in the second image;
• generate a mask based on the identified acoustic shadow;
• a joint capsule segmenting module configured to determine a joint capsule region in the one or more image data sets based on a corresponding identified bone surface, wherein the joint capsule segmenting module is configured to:
• develop an atlas of the region of interest;
• determine a point in the atlas that is disposed at a location farthest away from a determined position;
• identify one or more points in the atlas where one or more bone surfaces forming the region of interest start curving towards the farthest point;
• ascertain an initial estimate of a location of the joint capsule region;
• generate a prior likelihood for the joint capsule region based on the identified bone surface;
• segment the joint capsule region based on the prior likelihood;
• and
• an analysis module configured to identify the disease state in the segmented joint capsule region.

A non-transitory computer-readable non transitory media storing computer executable code to perform the method of:

• identifying a bone surface in each of one or more image data sets, wherein the one or more image data sets correspond to a region of interest in an object of interest, and wherein identifying the bone surface in each of the one or more image data sets comprises:
• obtaining a first image corresponding to the region of interest in a normal subject;
• obtaining a second image corresponding to the region of interest in the object of interest;
• identifying an acoustic shadow in the second image;
• generating a mask based on the identified acoustic shadow;
• segmenting a joint capsule region corresponding to the one or more image data sets based on a corresponding identified bone surface, wherein segmenting the joint capsule region comprises:
• developing an atlas of the region of interest;
• determining a point in the atlas that is disposed at a location farthest away from a determined position;
• identifying one or more points in the atlas where one or more bone surfaces forming the region of interest start curving towards the farthest point;
• ascertaining an initial estimate of a location of the joint capsule region;
• generating a prior likelihood for the joint capsule region based on the identified bone surface;
• segmenting the joint capsule region based on the prior likelihood;
• and
• analyzing the segmented joint capsule region to identify the disease state.

An imaging system, the system comprising:

• an acquisition subsystem configured to obtain a plurality of image data sets corresponding to a region of interest in an object of interest;
• a processing subsystem in operative association with the acquisition subsystem and comprising a detection platform, wherein the detection platform comprises:
• a bone surface identifying module configured to identify a bone surface in one or more image data sets, wherein the one or more image data sets correspond to a region of interest in an object of interest, and wherein for each of the one or more image data sets the bone surface identifying module is configured to:
• obtain a first image corresponding to the region of interest in a normal subject;
• obtain a second image corresponding to the region of interest in the object of interest;
• identify an acoustic shadow in the second image;
• generate a mask based on the identified acoustic shadow;
• a joint capsule segmenting module configured to determine a joint capsule region in the one or more image data sets based on a corresponding identified bone surface, wherein the joint capsule segmenting module is configured to:
• develop an atlas of the region of interest;
• determine a point in the atlas that is disposed at a location farthest away from a determined position;
• identify one or more points in the atlas where one or more bone surfaces forming the region of interest start curving towards the farthest point;
• ascertain an initial estimate of a location of the joint capsule region;
• generate a prior likelihood for the joint capsule region based on the identified bone surface;
• segment the joint capsule region based on the prior likelihood;
• and
• an analysis module configured to identify a disease state in the segmented joint capsule region.