WO2008034182A1

WO2008034182A1 - Method and system of automated image processing - one click perfusion

Info

Publication number: WO2008034182A1
Application number: PCT/AU2007/001388
Authority: WO
Inventors: Qing Yang
Original assignee: Apollo Medical Imaging Technology Pty Ltd
Priority date: 2006-09-20
Filing date: 2007-09-20
Publication date: 2008-03-27

Abstract

A method for automated processing of image data of a subject by a user, the method comprising the steps of loading the image data into a memory, detecting a global arterial input function (AIF) from a feeding blood vessel, detecting a baseline before arrival of a contrast agent, converting the signal intensity profile of the image data to the contrast concentration profile depending on the type of image data loaded, and calculating perfusion indices. The user undertakes a single action in order to enable the above steps and display the results of the steps.

Description

"Method and system of automated image processing - one click perfusion"

Cross-Reference to Related Applications

The present application claims priority from Australian Provisional Patent Application No 2006905183 filed on 20 September 2006, the content of which is incorporated herein by reference.

Field of the Invention

This invention relates to a method and system -for automated image processing of medical imaging data, and more particularly relates to a method and system for automated processing of dynamic perfusion imaging data of a subject in order to obtain desired imaging maps of functional information by a single user action.

Background to the Invention The process of measuring blood flow within a body of a subject non-invasively is useful in diagnosing and treating the subject. This is particularly the case where a part of a subject or patient, such as a tissue or organ, suffers from diseases due, for example, to cancer or malfunction. Determining perfusion indices including the blood flow through such a tissue or organ can provide important information to a physician in order to determine an appropriate treatment regime for the patient.

Existing systems pertaining to blood flow information have been disclosed. In general, the systems involve a contrast agent which is delivered as an intravascular bolus during a dynamic imaging session such as computerised tomography (CT), magnetic resonance imaging (MRI), nuclear medicine (NM) or positron emission tomography (PET). The temporal profile of the image intensity in a pixel or region of interest (ROI) reflects the characteristics of the contrast agent and hence the blood passing through the tissue.

A typical method of obtaining quantitative perfusion indices involves several steps including: (a) loading the dynamic imaging data into a computer memory;

(b) applying motion correction to compensate for motion artefacts due to the region of interest, such as a tissue or organ, that moves over time;

(c) measuring the arterial input function (AIF) from a feeding vessel to the tissue of interest, and setting the baseline time points before the contrast arrival; (d) converting the signal intensity profile to the contrast concentration profile depending on the type of imaging modality; (e) measuring the tissue profile;

(f) extracting the tissue impulse residue function (IRF) from the AIF and tissue profile using deconvolution; and

(g) calculating quantitative perfusion indices including blood flow, blood volume and mean transit time using the IRF.

However, the above process often requires the user to go through each step manually, particularly involving the steps for motion correction, selection of the appropriate AIF and setting the baseline time points. It is desirable to automate the entire process with minimum user intervention, which is particularly useful in a busy clinical environment.

The present invention seeks an automated post-processing method to substantially overcome or at least ameliorate, any one or more of the above-mentioned disadvantages associated with, for example, user controls for motion correction, selection of the appropriate AIF and setting the baseline time points or any other process associated with obtaining quantitative perfusion indices. The automated process can be initiated by the user with a single action of a button.

Summary of Invention

According to a first aspect of the invention there is provided a method for automated processing of image data of a subject by a user, the method comprising the steps of:

(a) loading the image data into an electronic memory means;

(b) detecting a global arterial input function (AIF) from a feeding blood vessel; (c) detecting a baseline before arrival of a contrast agent;

(d) converting the signal intensity profile of the image data to the contrast concentration profile depending on the type of image data loaded;

(e) calculating perfusion indices; wherein the user undertakes a single action in order to undertake steps (a) to (e) and display the results of steps (a) to (e).

The method may further include the step of applying motion correction after loading the data to compensate for motion artefacts due to the movement over time of, for example, a tissue or organ. The step of determining a global arterial input function (AIF) may be performed by searching through all pixels of imaging slices and identifying a major feeding blood vessel such as an artery. The step of determining the baseline may include time points before the arrival of the contrast agent on an AIF profile. The AIF may be identified by pixels having the properties of early contrast arrival with high and narrow peak profile of contrast enhancement within an arterial first-pass period. The method may further comprise the step of determining a venous output function (VOF) from a draining vein where a selected artery has partial voluming, such as in the brain, the vein being larger than the artery. The VOF can be used to scale up the AIF profile to minimize a partial voluming effect (PVE) from the AIF.

The type of image may be any one of CT, MRI, NM or PET. The single action that the user undertakes may be the mouse click of a control button on a computer interface or a key stroke on a keyboard of a computer.

The method may further include the feature of measuring the tissue profile and extracting a tissue impulse residue function (IRF) from the AIF and tissue profile by applying a selected model using deconvolution. The perfusion indices calculated may include blood flow, blood volume, mean transit time, arterial delay time and permeability related parameters using the IRF.

The perfusion indices may also include dynamic angiography by subtracting the mean baseline image from the dynamic contrast enhanced images. According to a second aspect of the invention there is provided a computer program means for instructing a computer processor, in response to a user undertaking a single action on the processor to perform the steps (a) through to (e) according to the first aspect.

Brief Description of the Drawings

The invention will hereinafter be described in a preferred embodiment, by way of example only, with reference to the drawings wherein:

Figure 1 is a block diagram showing a communication network including a number of scanners linked to a data storage system and a processing system;

Figure 2 shows a database interface on the data storage system or the processing system; and Figure 3 is a flow diagram showing steps performed by a computer program to enable the automated processing of image data and display of results.

Detailed Description of Preferred Embodiment

The present invention is particularly applicable to CT, MRI, NM and PET imaging systems. Raw data and/or images collected by a scan, such as from a CT scanner 20, MRI scanner 25, NM scanner 30 or PET scanner 35 are forwarded to a data storage system 40 in the form of a Picture Archiving Communications System (PACS) in Figure 1. A computer program operating on a processor 50, in the form of a computer, is used to retrieve or receive the various images or raw data from any one of the scanners 20, 25, 30 or 35 or from the data storage system 40. The program then processes those images to provide an improved data set for a clinician to use, particularly in relation to perfusion indices including blood flow, blood volume, mean transit time, arterial delay time, permeability related parameters, etc. The computer program need not reside on computer 50, but may reside in a console computer linked to any one of the scanners 20, 25, 30 or 35. Alternatively the program may reside in a workstation (stand-alone or in a system) or in the data storage system 40.

Medical image data is usually accessed on the workstation (or processor 50) via a typical database interface 200 as shown in Figure 2. The user can browse the database and select the image dataset for processing by highlighting or viewing it. For example, a user can select a dataset of series number 220 under patient name 202, a patient ID 204 or a Study ID 206, etc. The highlighted dataset 208 is for a CT perfusion of a stroke patient, as shown at 212. Different control buttons can be provided on the database interface panel for different processing models associated with different organs (such as brain, kidney, liver or other body organs) and different types of image data (such as CT, MRI, NM or PET). The automated image processing method can be executed by a mouse click on one control button 210 for the desired function.

Once the user has clicked on button 210 after selecting the dataset that they wish to have processed, the computer program will instruct the computer processor 50 or workstation to load the image data for the CT stroke dataset 208 into a memory of the computer processor. The processor 50 may apply motion correction to the data if it is applicable by software configuration to a particular organ, such as the brain. The automatic motion correction can be applied to the whole image of the subject and need not define a specific Region of Interest (ROI). The processor 50 then detects a global AIF from a feeding blood vessel by searching through all pixels of imaging slices and identifying a major feeding blood vessel such as an artery. Thereafter, the processor 50 detects or determines a baseline, including time points before the arrival of a contrast agent on the AIF profile. The baseline is the image data without contrast enhancement. The AIF is identified by pixels having the properties of early contrast arrival with high and narrow peak profile of contrast enhancement within an arterial first-pass period. The processor 50 further converts the signal intensity profile of the data to the contrast concentration profile depending on the type of image data loaded. The processor 50, by configuration, may further detect a VOF from a draining vein in order to scale up the AIF to minimize a PVE from the AIF if it is applicable to a particular organ, such as of the brain. Thereafter the perfusion indices are calculated by applying a model of a region of interest of the subject, in this case the brain of the selected dataset 208. All of these are carried out by the processor as a result of the user clicking an icon or button on screen, using a mouse, or using a keyboard. The results are then displayed on the processor 50 to the user. The user may alternatively click on screen, using a mouse, or use a keyboard, to inspect any one or more of the detected baseline, AIF or VOF profiles after automatically detecting any the baseline, global AIF or VOF, allowing the user to ascertain if these are the correct baseline or AIF, VOF profiles to use for calculating the perfusion indices.

The method can be configured to apply a motion correction method depending on the specific function being executed via the mouse click of a control button. For example, by clicking on the button 210 designed for CT perfusion stroke model, a conventional 3D or 2D rigid body image registration method can be applied automatically. Since the movement of a patient's head during data acquisition is mostly within the axial imaging plane, the automatic image registration method can be further restricted to a 2D rigid body method. The image registration method involves aligning the image of each time frame to a reference frame such as the first time frame. The 2D rigid body alignment involves an iterative optimization of two translational parameters and one rotational parameter by minimizing a cost function such as that defined by the mean least squares of the differences between the aligned image and the reference image. However, for body organs such as the kidneys or liver, non-uniform distortions are normally expected due to local tissue stretching associated with cardiac and/or respiratory motion. Some nonlinear motion compensation methods may be used such as an image registration method involving nonlinear warping (United States Patent 5,359,513 Kano, et al.) Alternatively, a virtual gating method (such as discussed in International Patent

Application No PCT/AU2006/000205 to the present applicant) may be used to minimize motion artefacts. For simplicity, the automated method can be configured to not apply motion correction for organs other than the brain.

The process of selecting the AIF involves the program searching through all pixels and identifying the pixels of a feeding blood vessel such as an artery which has the properties of early contrast arrival with high and narrow peak profile of contrast enhancement within a typical arterial first-pass period. The selected AIF is a global AIF which will be used in subsequent processing of perfusion maps for all pixels from all imaging slices. After the AIF is selected, the baseline time points can be set to include those time points before the contrast arrival on the AIF profile.

In the brain, many cerebral arteries are small, subjecting to partial volurning. The automated method may be configured to further comprise determining a venous output function (VOF) from a draining vein where a selected artery has partial voluming, the vein being larger than the artery. The VOF can be identified as the pixels with the maximum contrast enhancement peak, after the AIF first-pass time period, due to lack of partial voluming. The AIF may then be scaled up for having the same peak height or the same first-pass bolus area as the VOF to minimize a partial voluming effect (PVE) from the AIF. The first-pass AIF and VOF profiles can be obtained by fitting the profiles to gamma-variate function (GVF) profiles respectively to remove contrast recirculation effects.

The method, by software configuration, may pause after automatically detecting the AIF and VOF, allowing the user to inspect the detected baseline, AIF and VOF profiles. The software configuration can involve modifying a parameter in a configuration file. During the pause, the user can simply click a "Processing" button to continue processing if satisfied with the detected baseline, AIF and VOF profiles. Otherwise the user may manually place a ROI covering an anatomical region containing an artery and then click a button labelled with "Auto AIF in ROI" to instruct the computer processor 50 to automatically find a new AIF within the ROI. Similarly the user can place a different ROI covering another anatomical region covering a vein, and click on a button labelled with "Auto VOF in ROI" to enable the processor 50 to automatically find a new VOF within that ROI. The user may further manually adjust the baseline time frames and/or apply additional motion correction before continuing the processing by clicking on the "Processing" button.

The method further involves the step of applying the functional perfusion model on a pixel-by-pixel basis for processing of various perfusion maps (as disclosed in International Patent Application No PCT/AU2004/000821 to the present applicant). The perfusion maps may also include dynamic angiography by subtracting the mean baseline image from the dynamic contrast enhanced images. The results are finally presented in colour display mode for easy visualization, analysis, reporting and archiving by the user. With reference to Figure 3 there is shown a flow diagram 300 of the steps taken by a computer program loaded into memory of a computer processor 50. Computer program code is written to perform the various steps and functions depicted in the flow diagram 300 whereby upon clicking the button 210 on the database interface of Figure 2, at step 302, the computer program loads the selected dataset into the memory of the computer processor 50 at step 304. At step 306, if it is applicable to a particular scan, such as of the brain, then motion correction can be applied by the computer program. At step 308 an arterial input function is detected from a feeding blood vessel and, if applicable, a venous output function is detected from a draining vein and used to scale up the arterial input function. At step 310 baseline time points are set before the arrival of the contrast agent. By configuration, at optional step 311 the user is prompted to accept or refine any one or more of the detected AIF, VOF and baseline manually or semi-automatically. At step 312 the program converts the signal intensity profile of the data to the contrast concentration profile depending on the type of image scanned. At step 314 the tissue profile is measured and at step 316 the computer program extracts the tissue impulse residue function from the AIF and tissue profile by applying the selected model using deconvolution techniques. At step 318 the perfusion indices are calculated and at step 320 the results are then displayed of the whole process to the user. This embodiment has been described using an example of a CT perfusion of a stroke patient. The invention is equally applicable to other disease such as brain tumor or diseases in other body organs, not just of humans but animals as well, and using MRI or NM or PET scans.

It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the invention as shown in the specific embodiments without departing from the spirit or scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

Claims

CLAIMS:

1. A method for automated processing of image data of a subject by a user, the method comprising the steps of:

(a) loading the image data into an electronic memory means; (b) detecting a global arterial input function (AIF) from a feeding blood vessel;

(c) detecting a baseline before arrival of a contrast agent;

(d) converting the signal intensity profile of the image data to the contrast concentration profile depending on the type of image data loaded; (e) calculating perfusion indices; wherein the user undertakes a single action in order to undertake steps (a) to (e) and display the results of steps (a) to (e).

2. A method according to claim 1 further comprising applying motion correction after loading the image data to compensate for motion artefacts.

3. A method according to claim 1 or claim 2 wherein the step of determining a global arterial input function (AIF) is performed by searching through all pixels of imaging slices and identifying a major feeding blood vessel such as an artery.

4. A method according to any one claims 1 to 3 further comprising the step of determining the baseline including time points before the arrival of the contrast agent on an AIF profile.

5. A method according to claim 3 wherein the global AIF is identified by pixels having the properties of early contrast arrival with high and narrow peak profile of contrast enhancement within an arterial first-pass period.

6. A method according to claim 4 or claim 5 further comprising detecting a venous output function (VOF) from a draining vein where a selected artery has partial voluming, the draining vein being larger than the selected artery.

7. A method according to claim 6 further comprising the step of using a detected VOF to scale up a profile of the AIF to minimize a partial voluming effect from the AIF.

8. A method according to claim 6 or claim 7 wherein the VOF is identified by pixels having the maximum contrast enhancement peak, after the AIF first-pass time period, due to lack of partial voluming.

9. A method according to claim 7 or claim 8 wherein the scaling up is to the same peak height or the same first-pass bolus area as the VOF so as to minimize the partial voluming effect.

10. A method according to claim 9 further comprising the step of obtaining first- pass AIF and VOF profiles by fitting the AIF and VOF profiles to a respective gamma- variate function profile in order to remove contrast recirculation effects.

11. A method according to any one of the previous claims further comprising pausing, by performing an action or by software configuration, after automatically detecting any one or more of the baseline, global AIF or VOF, allowing the user to inspect the detected baseline, AIF or VOF profiles.

12. A method according to claim 11 further comprising the steps of allowing the user to manually place a Region of Interest (ROI) covering an anatomical region covering an artery or a vein and then performing an action to instruct a computer processor to automatically find a new AIF or VOF within the ROI.

13. A method according to claim 11 further comprising the step of allowing the user to manually adjust the baseline time frames and/or apply additional motion correction before continuing the processing by performing an action.

14. A method according to any one of the previous claims further comprising the steps of measuring a tissue profile and extracting a tissue impulse residue function (IRF) from the AIF and the tissue profile by applying a selected model using deconvolution.

15. A method according to any one of the previous claims wherein the perfusion indices calculated include any one or more of blood flow, blood volume, mean transit time, arterial delay time or permeability related parameters using the IRF.

16. A method according to any one of the previous claims wherein the perfusion indices calculated include dynamic angiography by subtracting a mean baseline image from dynamic contrast enhanced images.

17. A method according to any one of the previous claims wherein the type of image data is from any one of CT, MRI, NM or PET scans.

18. A method according to any one of the previous claims wherein the single action that the user undertakes is the mouse click of a control button on a computer interface or a key stroke on a keyboard of a computer.

19. A method according to any one of claims 11 to 13 wherein the action is the mouse click of a control button on a computer interface or a key stroke on a keyboard of a computer and the software configuration is by modifying a parameter in a configuration file.

20. Computer program means for instructing a computer processor to undertake the steps of any one of the previous claims and to display the results of the steps in response to the user undertaking the single action.