System and method for diagnostic vector classification support

• US 10,026,170 B2
• Filed: 07/19/2016
• Issued: 07/17/2018
• Est. Priority Date: 03/15/2013
• Status: Active Grant

• Current Assignee: Seno Medical Instruments, Inc.
• Original Assignee: Seno Medical Instruments, Inc.
• Inventors: Stavros, Anthony Thomas, Butler, Reni S., Lavin, Philip T., Zalev, Jason, Miller, Thomas G.
• Primary Examiner(s): Mistry, Oneal R
• Application Number:
  US15/213,889
• Publication Number:
  US 20160343132A1
• Time in Patent Office: 728 Days
• CPC Class Codes: A61B 2576/00 Medical imaging apparatus involving image processing or analysis A61B1/00009, A61B6/52 and A61B8/52 take precedence; image processing per se G06T; details of algorithms for analysing medical images G06T7/0012
• CPC Class Codes: A61B 5/0035 adapted for acquisition of images from more than one imaging mode, e.g. combining MRI and optical tomography
• CPC Class Codes: A61B 5/0095 by applying light and detecting acoustic waves, i.e. photoacoustic measurements
• CPC Class Codes: A61B 8/0825 for diagnosis of the breast, e.g. mammography
• CPC Class Codes: A61B 8/5223 for extracting a diagnostic or physiological parameter from medical diagnostic data for algorithms to analyse biomedical images G06T7/0012
• CPC Class Codes: A61B 8/5261 combining images from different diagnostic modalities, e.g. ultrasound and X-ray
• CPC Class Codes: A61B 8/5292 using additional data, e.g. patient information, image labeling, acquisition parameters
• CPC Class Codes: G06F 18/2113 by ranking or filtering the set of features, e.g. using a measure of variance or of feature cross-correlation
• CPC Class Codes: G06F 18/241 relating to the classification model, e.g. parametric or non-parametric approaches
• CPC Class Codes: G06T 2207/10132 Ultrasound image
• CPC Class Codes: G06T 2207/30096 Tumor; Lesion
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment

Abstract

The diagnostic vector classification support system and method disclosed herein may both reduce the time and effort required to train radiologists to interpret medical images, and provide a decision support system for trained radiologists who, regardless of training, have the potential to miss relevant findings. In an embodiment, a morphological image is used to identify a zone of interest in a co-registered functional image. An operator'"'"'s grading of a feature at least partially contained within the zone of interest is compared to one or more computer-generated grades for the feature. Where the operator and computer-generated grades differ, diagnostic support can be provided such as displaying additional images, revising the zone of interest, annotating one or more displayed images, displaying a computer-generated feature grade, among other possibilities disclosed herein.

Claims

1. A method for providing support in classifying a lesion using an optoacoustic image of a volume of tissue, wherein the volume of tissue comprises a tumor having a central nidus, the method comprising:

obtaining an ultrasound image of the volume of tissue, the ultrasound image presenting at least part of the central nidus and at least part of a peritumoral region;

obtaining an optoacoustic image of the volume of tissue, the optoacoustic image being coregistered with the ultrasound image and presenting the optical contrast of at least a portion of the part of the central nidus and at least a portion of the part of the peritumoral region;

identifying on the ultrasound image a tumoral boundary curve, the tumoral boundary curve approximating at least a portion of a perimeter of the central nidus of the tumor;

presenting on a display at least a portion of the optoacoustic image with the tumoral boundary curve superimposed thereon;

defining an internal zone within the displayed image based on the tumoral boundary curve; and

obtaining from an operator an operator feature score for a tumoral feature contained within the internal zone;

obtaining from the operator an operator feature score for an extra-tumoral feature contained at least partially outside the internal zone;

calculating by computer one or more computer-generated feature scores for the tumoral feature, based at least in part on information falling within the internal zone of the displayed image;

calculating by computer one or more computer-generated feature scores for the extra-tumoral feature, based at least in part on information falling outside the internal zone of the displayed image;

obtaining one or more supplementary inputs from the operator if either of the operator feature scores differ from the corresponding at least one computer-generated feature scores; and

classifying the lesion based on the operator feature scores.

2. The method of claim 1, wherein at least one of the one or more supplementary inputs is a modification to the operator feature score, the method further comprising re-obtaining the one or more supplementary inputs from the operator if the operator feature score differs from at least one of the one or more computer-generated feature scores.

3. The method of claim 1, wherein the at least one of the one or more supplementary inputs is a confirmation of the operator feature score.

4. A method for providing support in classifying a lesion using an optoacoustic image of a volume of tissue, wherein the volume of tissue comprises a tumor having a central nidus, the method comprising:

obtaining an ultrasound image of the volume of tissue, the ultrasound image presenting at least part of the central nidus and at least part of a peritumoral region;

obtaining an optoacoustic image of the volume of tissue, the optoacoustic image being coregistered with the ultrasound image and presenting the optical contrast of at least a portion of the part of the central nidus and at least a portion of the part of the peritumoral region;

identifying on the ultrasound image a tumoral boundary curve, the tumoral boundary curve approximating at least a portion of a perimeter of the central nidus of the tumor;

presenting on a display at least a portion of the optoacoustic image with the tumoral boundary curve superimposed thereon;

defining an internal zone within the displayed image based on the tumoral boundary curve; and

obtaining from an operator a classification for the lesion;

calculating by computer one or more computer-generated feature scores for the tumoral feature, based at least in part on information falling within the internal zone of the displayed image;

calculating by computer one or more computer-generated feature scores for the extra-tumoral feature, based at least in part on information falling outside the internal zone of the displayed image;

determining one or more computer-generated classifications for the lesion based upon at least the one or more computer-generated feature scores for the extra-tumoral feature and the at least one or more computer-generated feature scores for the tumoral feature;

obtaining one or more supplementary inputs from the operator if the operator classification differs from all of the one or more computer-generated classifications for the lesion; and

classifying the lesion based on the operator classification for the lesion.

5. A method for providing support for classifying a tumor having a central nidus and a peritumoral region adjacent to the central nidus, the method comprising:

presenting on a display an optoacoustic image of at least a portion of a volume of tissue comprising the tumor;

identifying on the display an internal zone of the optoacoustic image approximating at least a portion of the central nidus of the tumor;

evaluating at least a portion of the optoacoustic image falling within the internal zone to determine a grade for at least one internal feature contained within the central nidus of the tumor;

identifying on the display a boundary zone of the optoacoustic image approximating at least a portion of the peritumoral region of the tumor;

evaluating at least a portion of the optoacoustic image falling within the boundary zone to determine a grade for at least one peritumoral feature contained at least partially within the peritumoral region of the tumor; and

evaluating a portion of the optoacoustic image falling outside the internal zone and outside the boundary zone to determine a grade for at least one peripheral feature external to the central nidus of the tumor and at least partially external to the peritumoral region of the tumor;

identifying a classification of the tumor based on the at least one internal feature grade, the at least one peritumoral feature grade, and the at least one peripheral feature grade,

wherein determining a grade comprises;

calculating by computer one or more computer-generated feature scores for the at least one feature;

obtaining an operator feature score for the at least one feature; and

comparing at least one of the one or more computer-generated feature scores to the operator feature score;

wherein the at least one peripheral feature is selected from the set of;

a) vascularity;

b) oxygenation;

c) speckle;

d) blush;

e) amount of hemoglobin;

f) amount of blood;

g) ratio of oxygenated to deoxygenated blood; and

h) amount of radiating arteries;

i) amount of radiating veins;

j) amount of tumor neovessels;

k) amount of vessels oriented substantially parallel to a surface of the tumor;

l) amount of vessels oriented substantially perpendicular to a surface of the tumor;

m) length of vessels;

n) straightness of vessels; and

o) amount of interfering artifacts.