# US 2017/0287113 A1 · Classified truncation compensation

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

- Number: US 2017/0287113 A1
- Document ID: US20170287113A1
- Status: Published application
- Kind: A1
- Application number: 15/518,596
- Filed: 2015-10-20
- Published: 2017-10-05
- Earliest claimed: 2014-10-20
- Assignees: Koninklijke Philips N.V.
- Applicants: Koninklijke Philips N.V.
- Inventors: Shekhar DWIVEDI
- Web page: https://patentgrep.com/patent/US20170287113A1

## CPC

- G06T 5/77: Retouching; Inpainting; Scratch removal
- G06F 18/24: Classification techniques
- G06T 5/50: using two or more images, e.g. averaging or subtraction
- G06T 7/11: Region-based segmentation
- G06T 7/174: involving the use of two or more images
- G06T 12/30: Image post-processing, e.g. metal artefact correction
- G06T 2207/10088: Magnetic resonance imaging [MRI]
- G06T 2207/10104: Positron emission tomography [PET]
- G06T 2207/20152: Watershed segmentation
- G06T 2207/20224: Image subtraction
- G06T 2211/432: Truncation
- G06V 2201/03: Recognition of patterns in medical or anatomical images
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

PET/MR images are compensated with simplified adaptive algorithms for truncated parts of the body. The compensation adapts to a specific location of truncation of the body or organ in the MR image, and to attributes of the truncation in the truncated body part. Anatomical structures in a PET image that do not require any compensation are masked using a MR image with a smaller field of view. The organs that are not masked are then classified as types of anatomical structures, the orientation of the anatomical structures, and type of truncation. Structure specific algorithms are used to compensate for a truncated anatomical structure. The compensation is validated for correctness and the ROI is filled in where there is missing voxel data. Attenuation maps are generated from the compensated ROI.

## Claims

### Claim 1

A truncation compensation system, comprising:
a PET image memory configured to store a volume of PET imaging data;
a MR image memory configured to store a volume of MR imaging data having a MR field of vision;
one or more processors configured to:
mask the volume of the PET imaging data with the volume of the MR imaging data to generate a masked image of truncated regions of interest (ROIs);
classify the masked image as types of anatomical structures outside the MR FOV; and
compensate for truncation in the truncated ROIs based on the anatomical structure in the masked image outside the MR FOV to generate compensated ROIs.

### Claim 2 (depends on 1)

The system according to claim 1, wherein the one or more processors are further configured to:
classify a plurality of anatomical structures in the masked image.

### Claim 3 (depends on 1)

The system according to claim 1, wherein the one or more processors are further configured to:
classify an orientation classification for each classified anatomical structure.

### Claim 4 (depends on 1)

The system according to claim 1, wherein the one or more processors are further configured to:
classify a type of truncation based on structure and characteristics for each truncated ROI.

### Claim 5 (depends on 1)

The system according to claim 1, wherein the one or more processors are further configured to:
compensate for truncation in the truncated ROI using a classification specific truncation compensation for each truncated ROI.

### Claim 6 (depends on 1)

The system according to claim 1, wherein the one or more processors are further configured to:
compensate for truncation in a truncated ROI with a watershed algorithm.

### Claim 7 (depends on 1)

The system according to claim 1, wherein the one or more processors are further configured to:
merge the compensated ROIs with the MR imaging data to form a compensated MR map.

### Claim 8 (depends on 7)

The system according to claim 7, wherein the one or more processors are further configured to:
analyze missing voxels in the compensated MR map across three-dimensions to estimate the missing voxels; and
correct the compensated MR map with the estimated voxels.

### Claim 9

A method for compensating truncated MR
images, the method including:
receiving a volume of PET imaging data of a patient;
receiving a volume of truncated MR imaging data of a patient having a MR field of vision;
masking the volume of the PET imaging data with the volume of the MR imaging data to generate a masked image of truncated regions of interest (ROIs);
classifying the masked image as types of anatomical structures outside the MR FOV; and
compensating for truncation in the truncated ROIs based on the anatomical structure in the masked image outside the MR FOV to generate compensated ROIs.

### Claim 10 (depends on 9)

The method according to claim 9, wherein the classifying includes:
classifying a plurality of anatomical structures in the masked image.

### Claim 11 (depends on 9)

The method according to claim 9, wherein the classifying includes:
classifying an orientation classification for each classified anatomical structure.

### Claim 12 (depends on 9)

The method according to claim 9, wherein the classifying includes:
classifying a type of truncation based on structure and characteristics for each truncated ROI.

### Claim 13 (depends on 9)

The method according to claim 9, further including:
compensating for truncation in the truncated ROI using a classification specific truncation compensation for each truncated ROI.

### Claim 14 (depends on 9)

The method according to claim 9, wherein the compensating includes:
compensating for truncation in a truncated ROI with a watershed algorithm.

### Claim 15 (depends on 9)

The method according to claim 9, including:
merging the compensated ROIs with the MR imaging data to form a compensated MR map.

### Claim 16 (depends on 15)

The method according to claim 15, including:
analyzing missing voxels in the compensated MR map across three-dimensions to estimate the missing voxels; and
correcting the compensated MR map with the estimated voxels.

### Claim 17 (depends on 9)

A non-transitory computer readable medium carrying instructions for controlling a processor to perform the steps of claim 9.

### Claim 18

A system for truncation compensation of a PET image, the system comprising:
a processor configured to:
mask the PET image with a smaller volume anatomical image to generate a masked image of truncated regions truncated in the anatomical region;
identify anatomical structures in the masked image;
retrieve an anatomical image model of at least one of the identified anatomical structures;
scale and convert the anatomical image model;
combine the scaled and converted anatomical image model with the anatomical image; and
generate an attenuation map from the combined scaled and converted anatomical image model and the anatomical image.

### Claim 19 (depends on 18)

The system according to claim 18, wherein the processor is further configured to:
estimate missing voxels in the combined model and anatomical image.

### Claim 20 (depends on 18)

The system according to claim 18, wherein the processor is further configured to:
compensate the truncated regions using algorithms specific to the anatomical structure identified as being in the truncated regions.
