# US 8,788,012 B2 · Methods and apparatus for automatically registering lesions between examinations

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 8,788,012 B2
- Document ID: US8788012B2
- Status: Granted
- Kind: B2
- Application number: 11/603,701
- Filed: 2006-11-21
- Published: 2008-05-22
- Granted: 2014-07-22
- Earliest claimed: 2006-11-21
- Assignees: General Electric Company
- Applicants: Istvan Ubelhart; Balazs Cziria
- Inventors: Istvan Ubelhart; Balazs Cziria
- Web page: https://patentgrep.com/patent/US8788012B2

## CPC

- A61B 6/5235: {combining images from the same or different ionising radiation imaging techniques, e.g. PET and CT}
- A61B 5/1113: {Local tracking of patients, e.g. in a hospital or private home}
- A61B 6/032: {Transmission computed tomography [CT]}
- A61B 6/037: {Emission tomography}
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)

## Abstract

Methods and apparatus for automatically registering lesions between examinations are provided. The methods include performing a rigid registration using computed tomography (CT) images from a first image set and a second image set and performing a local non-rigid registration using positron emission tomography (PET) images from the first image set and the second image set. The methods further include automatically locating lesions on the second image set from the first image set.

## Claims

### Claim 1

A method for lesion matching in medical images, said method comprising:
obtaining a first set of computed tomography (CT) images fused to a first set of positron emission tomography (PET) images at a first point in time that form a baseline fused data set;
obtaining a second set of CT images fused to a second set of PET images at a different second point in time to form a follow-up fused data set;
performing a rigid registration between the first set of CT images and the second set of CT images to define an anatomical mapping between the baseline and follow-up fused data sets;
identifying a baseline bookmarked region of the first set of PET images using a plurality of bookmarks;
performing a local non-rigid registration between the baseline bookmarked region, that is encompassed by the plurality of bookmarks, and a corresponding region in the second set of PET images;
automatically propagating the bookmarks in the first set of PET images to the second set of PET images utilizing the local non-rigid registration, the anatomical mapping defined by the rigid registration between the baseline and follow-up fused data sets; and
displaying the bookmarks in the corresponding region in the second set of PET images to automatically match lesions on at least one PET image in the second set of PET images.

### Claim 2 (depends on 1)

A method in accordance with claim 1 wherein the first and second sets of CT images and the first and second sets of PET images are acquired using a combined PET/CT machine providing hardware based multi-modality registration of the baseline and follow-up fused data sets.

### Claim 3 (depends on 1)

A method in accordance with claim 1 wherein the baseline fused data set comprises baseline images and the follow-up fused data set comprises follow-up images acquired at a time subsequent to the baseline fused data set.

### Claim 4 (depends on 1)

A method in accordance with claim 1 further comprising displaying a plurality of views of images resulting from the rigid registration.

### Claim 5 (depends on 4)

A method in accordance with claim 4 further comprising allowing manual user adjustment of the rigid registration.

### Claim 6 (depends on 5)

A method in accordance with claim 5 wherein the user adjustment comprises one of translation, scaling and rotation adjustment.

### Claim 7 (depends on 1)

A method in accordance with claim 1 wherein the local non-rigid registration is based on the location of a plurality of bookmarks identifying the lesions.

### Claim 8 (depends on 1)

A method in accordance with claim 1 wherein the rigid registration comprises a linear registration.

### Claim 9 (depends on 1)

A method in accordance with claim 1 wherein the local non-rigid registration comprises a non-linear registration.

### Claim 10 (depends on 1)

A method in accordance with claim 1 wherein the baseline and follow-up fused data sets are each acquired using a single imaging system.

### Claim 11 (depends on 1)

A method in accordance with claim 1 further comprising:
identifying a lesion in the first set of PET images using the plurality of bookmarks; and
propagating the plurality of bookmarks from the first set of PET images to the second set of PET images to automatically locate the lesion on the second set of PET images.