Scatter reject method via energy calibration

1. A medical nuclear imaging system comprising:

a plurality of pixels which detect radiation events and estimate the energy of the detected radiation events;

a memory which stores a plurality of energy windows, the energy windows corresponding to the pixels;

an event verification module which windows the radiation events with the energy windows corresponding to the detecting pixels;

a reconstruction processor which reconstructs the windowed radiation events into an image representation.

2. The medical nuclear imaging system according to claim 1 , wherein the energy spectra describe the relationship between event counts and event energy, an event corresponding to the deposition of energy by a gamma photon.

3. The medical nuclear imaging system according to claim 1 , wherein the medical nuclear imaging system is one of a single-photon emission computed tomography (SPECT) imaging system and a positron emission tomography (PET) imaging system.

4. The medical nuclear imaging system according to claim 1 , further including:

a plurality of detectors including silicon photomultipliers (SiPMs), the SiPMs including the pixels.

5. The medical nuclear imaging system according to claim 1 , wherein the radiation events are gamma events.

6. The medical nuclear imaging system according to claim 1 , further including:

an energy calibration module configured to:

receive calibration energy spectra corresponding to the plurality of pixels, the calibration energy spectra generated from gamma photons from a calibration phantom;

fit the calibration energy spectra with Gaussian curves;

fit trend lines to both sides of the Gaussian curves of the calibration energy spectra; and

determine lower level discriminators (LLDs) and upper level discriminators (ULDs) from the fitted trend lines, the LLDs and the ULDs defining the energy windows.

7. The medical nuclear imaging system according to claim 6 , wherein the energy calibration module is further configured to:

determine the LLDs by identifying the energy of trend lines on the lower energy sides of the Gaussian curves at zero event counts; and

determine the ULDs by identifying the energy of trend lines on the higher energy sides of the Gaussian curves at zero event counts.

8. The medical nuclear imaging system according to claim 6 , wherein the pixels generate the calibration energy spectra from only gamma events of coincidence events.

9. The medical nuclear imaging system according to claim 6 , wherein the energy calibration module is further configured to:

for a plurality of energy levels to which the pixels can be calibrated, determine individual energy windows for the pixels.

10. A medical nuclear imaging method comprising:

detecting radiation events by a plurality of pixels;

estimating the energy of the detected radiation events;

receiving a plurality of energy windows, wherein the pixels correspond to the energy windows;

windowing the radiation events with the energy windows corresponding to the detecting pixels; and

reconstructing the windowed radiation events into an image representation.

11. The medical nuclear imaging method according to claim 10 , further including:

detecting the radiation events by silicon photomultipliers (SiPMs), which include the plurality of pixels.

12. The medical nuclear imaging method according to claim 10 , further including:

receiving calibration energy spectra corresponding to the plurality of pixels, the calibration energy spectra generated from gamma photons from a calibration phantom;

fitting the calibration energy spectra with Gaussian curves;

fitting trend lines to both sides of the Gaussian curves of the calibration energy spectra; and

determining lower level discriminators (LLDs) and upper level discriminators (ULDs) from the fitted trend lines, the LLDs and the ULDs defining the energy windows.

13. The medical nuclear imaging method according to claim 12 , further including:

determining the LLDs by identifying the energy of trend lines on the lower energy sides of the Gaussian curves at zero event counts; and

determining the ULDs by identifying the energy of trend lines on the higher energy sides of the Gaussian curves at zero event counts.

14. The medical nuclear imaging method according to claim 12 , further including:

generating the calibration energy spectra from only gamma events of coincidence events; and

for a plurality of energy levels to which the pixels can be calibrated, determining individual energy windows for the pixels.

15. A non-transitory computer readable medium carrying software which controls one or more processors to perform the method according to claim 10 .