IP Library Granted Patent US 12694483
Granted Patent B2
US 12694483 · App. 18/416,567 · Granted Jul 28, 2026

Systems and methods for a task-specific deep-learning-based denoising approach for myocardial perfusion SPECT

Inventors: Abhinav Kumar Jha (St. Louis, MO); Md Ashequr Rahman (St. Louis, MO); Zitong Yu (St. Louis, MO); Barry Siegel (St. Louis, MO)
Assignee: Washington University
G06T5/70G06T7/0012G06T2207/10108G06T2207/20081
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12694483
App. No.
18/416,567
Granted
Jul 28, 2026
Kind
B2
Abstract

A system for single-photon emission computed tomography (SPECT) is provided. The system includes a computer device comprises at least one processor in communication with at least one memory device. The at least one processor is programmed to: a) store a model trained to denoise computer tomography (CT) scans of a subject being examined; b) receive a CT scan of a first subject being examined; c) execute the model with the CT scan of the first subject as an input, wherein the model performs denoising on the CT scan while accounting for an observer loss function; and d) output a denoised-CT scan of the first subject.

Claims (29)

1 . A system for single-photon emission computed tomography (SPECT) comprising a computer device comprises at least one processor in communication with at least one memory device, wherein the at least one processor is programmed to:

store a model trained to denoise computer tomography (CT) scans of a subject being examined, wherein the at least one processor is further programmed to train the model with a plurality of historical low-dose CT scans and a plurality of normal dose CT scans, wherein the plurality of historical low-dose CT scans are images of where the corresponding subject received a lower dose of a radioactive tracer than subjects of the normal dose CT scans;

receive a CT scan of a first subject being examined;

execute the model with the CT scan of the first subject as an input, wherein the model performs denoising on the CT scan while accounting for an observer loss function; and

output a denoised-CT scan of the first subject.

2 . The system of claim 1 , wherein the lower dose is one of 12.5% and 6.25% of a normal dose.

3 . The system of claim 1 , wherein the at least one processor is further programmed to execute the model for a plurality of CT scans.

4 . The system of claim 3 , wherein the plurality of CT scans is a part of a video.

5 . The system of claim 1 , wherein the CT scan was taken during a myocardial perfusion.

6 . The system of claim 1 , wherein the at least one processor is further programmed to train the model with an observer loss function.

7 . A method for single-photon emission computed tomography (SPECT), the method implemented by a computer device comprising at least one processor in communication with one or more memory devices, the method comprises:

storing a model trained to denoise computer tomography (CT) scans of a subject being examined;

receiving a CT scan of a first subject being examined, wherein the at least one processor is further programmed to train the model with a plurality of historical low-dose CT scans and a plurality of normal dose CT scans, wherein the plurality of historical low-dose CT scans are images of where the corresponding subject received a lower dose of a radioactive tracer than subjects of the normal dose CT scans;

executing the model with the CT scan of the first subject as an input, wherein the model performs denoising on the CT scan while accounting for an observer loss function; and

outputting a denoised-CT scan of the first subject.

8 . The method of claim 7 , wherein the lower dose is one of 12.5% and 6.25% of a normal dose.

9 . The method of claim 7 further comprising executing the model for a plurality of CT scans.

10 . The method of claim 9 , wherein the plurality of CT scans is a part of a video.

11 . The method of claim 7 , wherein the CT scan was taken during a myocardial perfusion.

12 . The method of claim 7 further comprising training the model with an observer loss function.

13 . A computer device for single-photon emission computed tomography (SPECT) comprising at least one processor in communication with at least one memory device, wherein the at least one processor is programmed to:

store a model trained to denoise computer tomography (CT) scans of a subject being examined, wherein the at least one processor is further programmed to train the model with a plurality of historical low-dose CT scans and a plurality of normal dose CT scans, wherein the plurality of historical low-dose CT scans are images of where the corresponding subject received a lower dose of a radioactive tracer than the subjects of normal dose CT scans;

receive a CT scan of a first subject being examined;

execute the model with the CT scan of the first subject as an input, wherein the model performs denoising on the CT scan while accounting for an observer loss function; and

output a denoised-CT scan of the first subject.

14 . The computer device of claim 13 , wherein the lower dose is one of 12.5% and 6.25% of a normal dose.

15 . The computer device of claim 13 , wherein the at least one processor is further programmed to execute the model for a plurality of CT scans.

16 . The computer device of claim 15 , wherein the plurality of CT scans is a part of a video.

17 . The computer device of claim 13 , wherein the CT scan was taken during a myocardial perfusion.