Methods for reducing motion induced artifacts in a projection imaging system

• US 5,271,055 A
• Filed: 08/19/1992
• Issued: 12/14/1993
• Est. Priority Date: 08/19/1992
• Status: Expired due to Term

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Hsieh, Jiang, Gard, Michael F., Ritchie, Cameron J.
• Primary Examiner(s): Porta, David P.
• Application Number:
  US07/932,446
• Time in Patent Office: 482 Days
• Field of Search: 324/309, 324/307, 378/8, 378/4, 378/20, 378/95, 378/106, 364/413.13-413.15
• US Class Current: 378/95
• CPC Class Codes: A61B 5/1135 by monitoring thoracic expansion
• CPC Class Codes: A61B 5/7285 for synchronising or triggering a physiological measurement or image acquisition with a physiological event or waveform, e.g. an ECG signal A61B6/5205, A61B6/541, A61B8/5207, A61B8/543 take precedence
• CPC Class Codes: A61B 6/032 Transmission computed tomography [CT]
• CPC Class Codes: A61B 6/541 involving acquisition triggered by a physiological signal

Abstract

A medical imaging system has a source of X-rays and a detector which produces an electrical signal in response to X-rays. A signal generated by the detector, during an acquisition interval, is used to construct an image of a medical patient, which may contain artifacts due to motion of patient. A signal is produced indicative of the motion, which has a quiescent period of minimum movement. A first portion of the signal is selected such that if image acquisition commenced at the end of the first portion, the acquisition interval would occur during a quiescent period. This first portion is initially used as a reference signal portion. Then subsequent portions of the signal are compared to the reference signal portion to produce a descriptor of the degree of similarity. An image acquisition is commenced when the descriptor indicates a given degree of similarity that is above a defined threshold. Periodically a subsequent portion of the signal is combined with the previous reference signal portion to create an updated reference signal portion which thereafter is used in the comparison step.

Claims

1. A method for reducing artifacts in an image that result from motion of an object being imaged, where the image data is obtained during an acquisition interval, the motion having a quiescent period during which the motion is at a minimum;

said steps of said method comprising;

producing a signal indicative of the motion;

defining a point in the signal such that if image acquisition commenced at that point a quiescent period would occur during an acquisition interval;

performing an analysis of a waveform shape of the signal which analysis uses an adaptive shape criterion that is updated over time based on information from the signal;

predicting from results of the analysis when a quiescent period is approaching; and

commencing an image acquisition at a time prior to the predicted quiescent period so that the acquisition will occur during the quiescent period.

2. The method as recited in claim 1 wherein said step of analyzing a waveform shape of the signal utilizes either a least means square or a recursive least squares technique.

3. The method as recited in claim 1 wherein said step of analyzing a waveform shape of the signal utilizes adaptive moving correlation.

4. A method for reducing artifacts in an image that result from motion of an object being imaged, wherein image data is obtained during an acquisition interval, the motion having a quiescent period during which the motion is at a minimum;

said steps of said method comprising;

producing a signal indicative of the motion;

selecting a first portion of the signal which occurs for a given time interval, such that a quiescent period will occur during the acquisition interval if image acquisition commenced at an end of the given time interval;

thereafter selecting subsequent portions of the signal, which occur during other time intervals;

comparing each selected subsequent portion to the reference signal portion and providing an descriptor of the similarity between the two portions; and

commencing an image acquisition when the descriptor indicates a given degree of similarity.

5. The method as recited in claim 4 wherein said step of producing includes digitizing a signal indicative of the motion to produce a series of digital signal samples;

and wherein said steps of selecting a first portion and selecting subsequent portions of the signal comprise selecting groups of digital signal samples.

6. The method as recited in claim 4 wherein said step of comparing utilizes either a least means square or a recursive least squares technique to produce the descriptor of the similarity.

7. The method as recited in claim 4 wherein said step of comparing utilizes an adaptive moving correlation process to produce the descriptor of the similarity.

8. The method as recited in claim 4 further comprising revising the reference signal portion by combining a previous reference signal portion with a selected subsequent portion.

9. A method for reducing artifacts in an image that result from motion of an object being imaged, wherein image data is obtained during an acquisition interval, the motion having a quiescent period that is shorter in duration that the acquisition interval and during which the motion is at a minimum;

said steps of said method comprising;

(a) producing a signal indicative of the motion;

(b) locating a position in the signal which position occurs during a quiescent period;

(c) defining a point in the signal prior to the position such that if image acquisition commenced at that point a quiescent period would occur during an acquisition interval;

(d) selecting a first portion of the signal which occurred for a given interval of time prior to the defined point as a correlation kernel;

(e) thereafter selecting other portions of the signal which occur for periods of time having like duration to the given interval of time;

(f) comparing each another selected portion to the correlation kernel and providing an descriptor of the similarity between the two portions; and

(g) commencing an image acquisition when a descriptor indicates a given degree of similarity.

10. The method as recited in claim 9 further comprising deriving a correlation kernel R_temp from another selected portion;

and revising the correlation kernel by combining the correlation kernel R_temp with a previous correlation kernel.

11. The method as recited in claim 9 further comprising deriving a correlation kernel R_temp from another selected portion;

and revising the correlation kernel by averaging the correlation kernel R_temp with a previous correlation kernel.

12. The method as recited in claim 11 wherein the averaging weights the previous correlation kernel more heavily than correlation kernel R_temp.

13. The method as recited in claim 9 further comprising:

periodically locating another position in the signal which other position occurs during a quiescent period;

defining another point in the signal prior to the other position such that if image acquisition commenced at that other point a quiescent period would occur during an acquisition interval;

selecting a section of the signal which occurred prior to the defined point as a temporary correlation kernel; and

producing a revised correlation kernel by averaging the temporary correlation kernel with a previous correlation kernel, which revised correlation kernel then is used in the comparing step.

14. The method as recited in claim 9 wherein said step of producing a signal includes digitizing a signal indicative of the motion to produce a series of digital signal samples.

15. The method as recited in claim 13 wherein said step of defining a point in the signal involves counting a predefined number of signal samples from the position.

16. The method as recited in claim 13 wherein said steps of selecting a first portion and selecting other portions of the signal comprise selecting groups having a specified number of signal samples.

17. The method as recited in claim 9 wherein said step of locating a position in the signal which position occurs during a quiescent period comprises identifying a position at which the signal reaches a maximum level.

18. The method as recited in claim 9 wherein said step of locating a position in the signal which position occurs during a quiescent period comprises identifying a position at which the signal reaches a minimum level.