US9336457B2 — Adaptive anatomical region prediction

Claim 1

Independent Claim 99 words

A method for adaptive region prediction, comprising: receiving, by a processor, a set of exemplar images including annotated first landmarks; obtaining, by the processor, user definitions of first anatomical regions in the exemplar images; detecting, by the processor, second landmarks in a subject image; computing, by the processor, anatomical similarity scores between the subject image and the exemplar images based on the first and second landmarks; and predicting, by the processor, a second anatomical region in the subject image by adaptively combining coordinates of points representing the first anatomical regions weighted by coefficients derived from the anatomical similarity scores.

Claim 2

Independent Claim 99 words

A method for adaptive region prediction, comprising: receiving, by a processor, a set of exemplar images including annotated first landmarks; obtaining, by the processor, user definitions of first anatomical regions in the exemplar images; detecting, by the processor, second landmarks in a subject image; computing, by the processor, anatomical similarity scores between the subject image and the exemplar images based on the first and second landmarks; and predicting, by the processor, a second anatomical region in the subject image by adaptively combining coordinates of points representing the first anatomical regions weighted by coefficients derived from the anatomical similarity scores.

Claim 3

Dependent Claim 46 words

The method of claim 1 further comprising acquiring the set of exemplar images and the subject image by magnetic resonance (MR) imaging, computed tomography (CT), helical CT, x-ray, positron emission tomography (PET), PET-CT, fluoroscopy, ultrasound, single-photon emission computed tomography (SPECT), SPECT-CT, MR-PET, or a combination thereof.

Claim 4

Dependent Claim 40 words

The method of claim 1 further comprising: annotating a set of the first landmarks in each exemplar image; training detectors for the annotated first landmarks, and using the trained detectors to automatically detect the second landmarks in the subject image.

Claim 5

Dependent Claim 31 words

The method of claim 3 wherein annotating the set of the first landmarks in each exemplar image comprises annotating, in each exemplar image, a subset of a set of defined landmarks.

Claim 6

Dependent Claim 33 words

The method of claim 3 further comprising: defining a set of defined landmarks; and storing the set of the defined landmarks and the user definitions of the first anatomical regions in a database.

Claim 7

Dependent Claim 35 words

The method of claim 1 wherein obtaining the user definitions of the first anatomical regions in the exemplar images comprises obtaining the user definitions via a graphical interface, wherein the user definitions comprise geometrical shapes.

Claim 8

Dependent Claim 48 words

The method of claim 1 , wherein computing the anatomical similarity score comprises: identifying a common set of corresponding first and second landmarks in the subject image and each exemplar image; and computing the anatomical similarity scores based on the common sets of corresponding first and second landmarks.

Claim 9

Dependent Claim 42 words

The method of claim 7 further comprising: identifying a set of relevant first and second landmarks from each common set of corresponding first and second landmarks; and computing the anatomical similarity scores based on the sets of relevant first and second landmarks.

Claim 10

Dependent Claim 51 words

The method of claim 8 wherein identifying the set of relevant first and second landmarks comprises identifying pairs of corresponding first and second landmarks from the common set of corresponding first and second landmarks, wherein the pairs include first landmarks that are within a pre-determined distance to the first anatomical regions.

Claim 11

Dependent Claim 27 words

The method of claim 8 wherein the set of relevant first and second landmarks is identified based on common semantic description shared with the first anatomical regions.

Claim 12

Dependent Claim 21 words

The method of claim 8 wherein the set of relevant first and second landmarks is identified using a machine learning algorithm.

Claim 13

Dependent Claim 30 words

The method of claim 1 further comprising applying a global alignment on the subject image prior to computing the anatomical similarity scores between the subject image and the exemplar images.

Claim 14

Dependent Claim 25 words

The method of claim 12 wherein applying the global alignment comprises applying a Procrustes alignment to optimally superimpose the subject image on each exemplar image.

Claim 15

Dependent Claim 58 words

The method of claim 1 wherein computing the anatomical similarity scores between the subject image and the exemplar images based on the first and second landmarks comprises: calculating distances between corresponding first landmarks in the subject image and second landmarks in each exemplar image; and calculating the anatomic similarity score of each exemplar image based on the distances.

Claim 16

Dependent Claim 28 words

The method of claim 14 wherein calculating the anatomical similarity score includes combining a mean of the distances and a standard error for the mean of the distances.

Claim 17

Dependent Claim 23 words

The method of claim 14 further comprising removing outlier distances from computation prior to calculating the anatomical similarity score of each exemplar image.

Claim 18

Dependent Claim 15 words

The method of claim 1 wherein the first and second anatomical regions correspond to fields-of-view.

Claim 19

Dependent Claim 53 words

The method of claim 1 wherein predicting the second anatomical region in the subject image comprises: deriving the coefficients from the anatomical similarity scores; and calculating, based on the coefficients, a weighted linear sum of corresponding elements of the coordinates of the first anatomical regions to determine points representing the second anatomical region.

Claim 20

Dependent Claim 99 words

A non-transitory computer-readable medium embodying a program of instructions executable by machine to perform steps for adaptive region prediction, the steps comprising: receiving a set of exemplar images including annotated first landmarks; obtaining user definitions of first anatomical regions in the exemplar images; detecting second landmarks in a subject image; computing anatomical similarity scores between the subject image and the exemplar images based on the first and second landmarks; and predicting a second anatomical region in the subject image by adaptively combining coordinates of points representing the first anatomical regions weighted by coefficients derived from the anatomical similarity scores.

Claim 21

Dependent Claim 99 words

A non-transitory computer-readable medium embodying a program of instructions executable by machine to perform steps for adaptive region prediction, the steps comprising: receiving a set of exemplar images including annotated first landmarks; obtaining user definitions of first anatomical regions in the exemplar images; detecting second landmarks in a subject image; computing anatomical similarity scores between the subject image and the exemplar images based on the first and second landmarks; and predicting a second anatomical region in the subject image by adaptively combining coordinates of points representing the first anatomical regions weighted by coefficients derived from the anatomical similarity scores.

Claim 22

Independent Claim 115 words

A system for adaptive region prediction, comprising: a non-transitory memory device for storing computer-readable program code; and a processor in communication with the memory device, the processor being operative with the computer-readable program code to perform steps comprising: receiving a set of exemplar images including annotated first landmarks; obtaining user definitions of first anatomical regions in the exemplar images; detecting second landmarks in a subject image; computing anatomical similarity scores between the subject image and the exemplar images based on the first and second landmarks; and predicting a second anatomical region in the subject image by adaptively combining coordinates of points representing the first anatomical regions weighted by coefficients derived from the anatomical similarity scores.

Claim 23

Independent Claim 115 words

A system for adaptive region prediction, comprising: a non-transitory memory device for storing computer-readable program code; and a processor in communication with the memory device, the processor being operative with the computer-readable program code to perform steps comprising: receiving a set of exemplar images including annotated first landmarks; obtaining user definitions of first anatomical regions in the exemplar images; detecting second landmarks in a subject image; computing anatomical similarity scores between the subject image and the exemplar images based on the first and second landmarks; and predicting a second anatomical region in the subject image by adaptively combining coordinates of points representing the first anatomical regions weighted by coefficients derived from the anatomical similarity scores.

Annuity Stage	Window Opens	Payment Deadline	Surcharge Period Until	Estimated Fee	Status
3.5 Years (Year 4)	2019-05-10	2019-11-09	2020-05-10	$2000.00	Window Closed / Lapsed
7.5 Years (Year 8)	2023-05-10	2023-11-09	2024-05-10	$3760.00	Window Closed / Lapsed
11.5 Years (Year 12)	2027-05-10	2027-11-09	2028-05-10	$7700.00	Pending Window

Adaptive anatomical region prediction

Abstract

Inventors

Bibliographic Details

Claims Analysis

References Cited (Backward Citations)

USPTO Annuity & Maintenance Windows

Assigned Categories & Tags

Assign Category Tag