# US 10,371,836 B2 · Digital positron emission tomography (DPET) energy calibration method

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

- Number: US 10,371,836 B2
- Document ID: US10371836B2
- Status: Granted
- Kind: B2
- Application number: 14/405,015
- Filed: 2013-05-30
- Published: 2015-06-11
- Granted: 2019-08-06
- Earliest claimed: 2012-06-27
- Assignees: Koninklijke Philips N.V.
- Applicants: Koninklijke Philips N.V.
- Inventors: Sharon Xiaorong Wang; Thomas Leroy Laurence
- Web page: https://patentgrep.com/patent/US10371836B2

## CPC

- G01T 1/2985: {In depth localisation, e.g. using positron emitters; Tomographic imaging (longitudinal and transverse section imaging; apparatus for radiation diagnosis sequentially in different planes, steroscopic radiation diagnosis); (using external radiation sources A61B6/02)}
- G01T 1/1647: {Processing of scintigraphic data (not related to a particular imaging system G01T1/2992)}
- G01T 1/249: {specially adapted for use in SPECT or PET (SPECT imaging G01T1/1642; PET imaging G01T1/2985; detecting prohibited goods, e.g. weapons, explosives, hazardous substances, contraband or smuggled objects G01V5/20)}
- G01T 7/005: {calibration techniques (stabilization of spectrometer G01T1/40)}
- G01T 1/247: {Detector read-out circuitry (for processing gain or off-set correction H04N)}
- G01T: MEASUREMENT OF NUCLEAR OR X-RADIATION (radiation analysis of materials, mass spectrometry G01N23/00; tubes for determining the presence, intensity, density or energy of radiation or particles H01J47/00)

## Abstract

A system (10) and method for energy correction of positron emission tomography (PET) event data by at least one processor. Event data for a plurality of strike events corresponding to gamma events is received. Each strike event is detected by a pixel of a detector module (50) and includes an energy and a time. The energy of the strike events is linearized using an energy linearity correction model including one or more parameters. Clusters of the strike events are identified based on the times of the strike events, and sub-clusters of the clusters are identified based on the pixels corresponding to the strike events of the clusters. Energies of the sub-clusters are corrected using a first set of correction factors, and energies of clusters including a plurality of sub-clusters are corrected using a second set of correction factors.

## Claims

### Claim 1

A system for energy correction of positron emission tomography (PET) event data, said system comprising:
at least one processor programmed to:
receive event data for a plurality of strike events corresponding to gamma events, each strike event detected by a pixel of one of a plurality of detector modules around an imaging volume and including an energy and a time, the detector modules detecting pairs of gamma photons from a region of interest of a patient in the imaging volume;
linearize the energy of the strike events using an energy linearity correction model including one or more parameters to compensate for energy non-linearity of scattered strike events;
identify clusters of the strike events based on the times of the strike events, wherein the strike events in each of the identified clusters are within a temporal window that corresponds to a single gamma event, each cluster including a plurality of strike events that correspond to a single gamma event;
identify sub-clusters of the clusters based on the pixels corresponding to a subset of the strike events of each cluster;
correct energies of the strike events of the sub-clusters using a first set of correction factors;
correct energies of the strike events of each cluster of a plurality of sub-clusters using a second set of correction factors;
reconstruct the energy corrected event data into an image of the region of interest of the patient; and
a display device configured to display the image.

### Claim 2 (depends on 1)

The system according to claim 1, wherein the linearization of the energy of the strike events includes for each of the strike events:
determining parameter values corresponding to the parameters of the energy linearity correction model, the parameter values specific to the pixel corresponding to the strike event;
update the parameters of the energy linearity correction model with the determined parameter values; and,
correct the energy of the strike event using the updated energy linearity correction model.

### Claim 3 (depends on 1)

The system according to claim 1, wherein the energy linearity correction model is:
P
=
-
k
1
⁢
A
⁢
⁢
log
10
⁡
(
1
-
(
k
2
⁢
P
0
A
)
k
3
)
where k1, k2, and k3 are the parameters, P0 is the energy of the strike event, P is the corrected energy of the strike event, and A is the number of active cells.

### Claim 4 (depends on 1)

The system according to claim 1, wherein the energy of the strike events is linearized in a range between 100 keV and 500 keV.

### Claim 5 (depends on 1)

The system according to claim 1, wherein each of the identified clusters corresponds to a single one of the gamma events and includes the strike events corresponding to the gamma event.

### Claim 6 (depends on 1)

The system according to claim 1, wherein the identification of the sub-clusters includes:
grouping pixels of one or more detector modules corresponding to the strike events into a plurality of non-overlapping blocks, the pixels of the detector modules including the pixels corresponding to the strike events;
wherein each of the sub-clusters corresponds to a single one of the non-overlapping blocks and includes strike events of the corresponding cluster which correspond to the single non-overlapping block.

### Claim 7 (depends on 1)

The system according to claim 1, wherein each correction factor of the first set of correction factors and/or the second set of correction factors is one or more of:
specific to a pixel of a detector module of the strike events; and,
a ratio of a known peak energy and a corresponding measured peak energy.

### Claim 8 (depends on 1)

The system according to claim 1, wherein the correction of the sub-clusters includes:
for each of the sub-clusters:
determining energy of the sub-cluster by summing energies of one or more strike events of the sub-cluster;
determining a main strike event of the sub-cluster, the main strike event including the greatest energy of all the strike events of the sub-cluster;
determining a correction factor using the first set of correction factors, the correction factor being specific to a pixel corresponding to the main strike event; and,
applying the correction factor to the energy of the sub-cluster.

### Claim 9 (depends on 1)

The system according to claim 1, wherein the correction of the clusters including a plurality of sub-clusters includes:
for each of the clusters including a plurality of sub-clusters:
determining energy of the cluster by summing energies of one or more sub-clusters of the cluster;
determining a main strike event of the cluster, the main strike event including the greatest energy of all strike events of the cluster;
determining a correction factor using the second set of correction factors, the correction factor being specific to a pixel corresponding to the main strike event; and,
applying the correction factor to the energy of the cluster.

### Claim 10

A system for energy correction of positron emission tomography (PET) event data, said system comprising:
a plurality of solid state detector modules detecting gamma photons from a region of interest of a patient and generating the event data in response to the gamma photons;
at least one energy correction processor programmed to:
receive event data for a plurality of strike events corresponding to gamma events, each strike event detected by a pixel of a detector module and including an energy and a time,
linearize the energy of the strike events using an energy linearity correction model including one or more parameters,
identify clusters of the strike events based on the times of the strike events,
identify sub-clusters of the clusters based on the pixels corresponding to the strike events of the clusters,
correct energies of the sub-clusters using a first set of correction factors, and,
correct energies of clusters including a plurality of sub-clusters using a second set of correction factors; and,
a reconstruction processor processing the event data as corrected by the energy correction processor to create an image representation of the region of interest.

### Claim 11 (depends on 1)

The system according to claim 1, wherein each gamma event is 511 keV and further including:
a calibration processor determining parameter values for the energy linearity correction model using event data for a plurality of calibration sources, including Co57 and Na22.

### Claim 12

A method for energy correction of positron emission tomography (PET) event data, said method comprising:
receiving by at least one processor event data for a plurality of strike events corresponding to 511 keV gamma photons from a region of interest of a patient disposed in an imaging volume of a PET scanner, each of the strike events being detected by a pixel of a detector module and including an energy and a time, individual ones of the 511 keV gamma photons being scattered in a scintillation crystal of a first detector, the scattering causing a first portion of the 511 keV gamma photon to be converted to light in a first strike event and a remainder of the gamma photon to travel into a scintillation crystal of an adjoining detector converting a second portion of the gamma photon to light in a second strike event;
linearizing by the at least one processor the energy of the strike events of the event data using an energy linearity correction model including one or more parameters;
identifying by the at least one processor clusters of the strike events based on the times of the strike events, each cluster corresponding to the strike events caused by a single 511 keV gamma photon;
identifying by the at least one processor sub-clusters of the clusters based on the pixels corresponding to the strike events of the clusters;
correcting by the at least one processor energies of the strike events of the sub-clusters of the event data using a first set of correction factors; and,
further correcting by the at least one processor energies of the strike events of each cluster including a plurality of sub-clusters of the event data using a second set of correction factors;
reconstructing the linearized and corrected event data into an image depicting the region of interest of the patient; and,
displaying the image with a display device.

### Claim 13 (depends on 12)

The method according to claim 12, wherein the energy linearity correction model is:
P
=
-
k
1
⁢
A
⁢
⁢
log
10
⁡
(
1
-
(
k
2
⁢
P
0
A
)
k
3
)
where k1, k2, and k3 are the parameters, P0 is the energy of the strike event, P is the corrected energy of the strike event, and A is the number of active cells.

### Claim 14 (depends on 12)

The method according to claim 12, wherein the identification of the sub-clusters includes:
grouping pixels of one or more detector modules corresponding to the strike events into a plurality of non-overlapping blocks, the pixels of the detector modules including the pixels corresponding to the strike events;
wherein each of the sub-clusters corresponds to a single one of the non-overlapping blocks and includes strike events of the corresponding cluster which correspond to the single non-overlapping block.

### Claim 15 (depends on 12)

The method according to claim 12, wherein each correction factor of the first set of correction factors and/or the second set of correction factors is one or more of:
specific to a pixel of a detector module of the strike events; and,
a ratio of a known peak energy and a corresponding measured peak energy.

### Claim 16 (depends on 12)

The method according to claim 12, wherein the correction of the sub-clusters includes:
for each of the sub-clusters:
determining energy of the sub-cluster by summing energies of one or more strike events of the sub-cluster;
determining a main strike event of the sub-cluster, the main strike event including the greatest energy of all the strike events of the sub-cluster;
determining a correction factor using the first set of correction factors, the correction factor being specific to a pixel corresponding to the main strike event; and,
applying the correction factor to the energy of the sub-cluster.

### Claim 17 (depends on 12)

The method according to claim 12, wherein the correction of the clusters including a plurality of sub-clusters includes:
for each of the clusters including a plurality of sub-clusters:
determining energy of the cluster by summing energies of one or more sub-clusters of the cluster;
determining a main strike event of the cluster, the main strike event including the greatest energy of all strike events of the cluster;
determining a correction factor using the second set of correction factors, the correction factor being specific to a pixel corresponding to the main strike event; and,
applying the correction factor to the energy of the cluster.

### Claim 18 (depends on 12)

The method according to claim 12, further including:
determining by the at least one processor parameter values for the energy linearity correction model using event data for a plurality of calibration sources, including Co57 and Na22.

### Claim 19

A positron emission tomography (PET) system with energy correction of positron emission tomography (PET) event data, said system comprising:
a ring of gamma photon detectors, each gamma photon detector including a scintillator which converts gamma energy into light scintillations and light radiation sensitive elements which convert each light scintillation into event data, the ring of gamma photon detectors encircling an imaging volume configured to receive 511 keV gamma photons from a region of interest of a patient, each 511 keV photon striking a scintillator being a gamma event;
wherein each 511 keV gamma photon strikes a first scintillation crystal causing a first scintillation and is scattered into a second scintillation crystal causing a second scintillation, the output from each light radiation sensitive element in response to a scintillation being a strike event, the plurality of strike events corresponding to a single 511 keV gamma event being a cluster;
at least one processor programmed to:
receive the event data for a plurality of the strike events corresponding to each gamma event, each strike event detected by a pixel of a detector module and including an energy and a time;
linearize the energy of the strike events between 100 keV and 500 keV using an energy linearity correction model;
identify the clusters of the strike events based on the times of the strike events;
correct energies of the strike events of sub-clusters of the identified clusters using a first level of clustering energy correction factors;
correct the energies of the strike events corresponding to each cluster using second level clustering energy correction factors;
combine the corrected energies of the strike events of a common cluster;
compare the combined corrected energies with a preselected threshold; and
reconstruct the clusters with combined energies exceeding the preselected threshold into a PET image of the region of interest.

### Claim 20 (depends on 19)

The system according to claim 19, further including a display device configured to display the PET image.