JP2007160108A - System and method for image based physiological monitoring of cardiovascular function - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a new system and method for image based physiological monitoring. <P>SOLUTION: A system and method for monitoring of cardiac function includes selecting a plane through a heart from which to acquire heart function data, acquiring a two-dimensional image from the selected planes, projecting a plurality of beams into the image through the chamber center, sampling image intensities along each point of the beam, and calculating an image intensity to obtain a series of time measurements wherein an image intensity along each the beam can be plotted as a function of time. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  This application claims priority to “Image Based Physiological Monitoring of Cardiovascular Function” of US Provisional Application No. 60/748558 filed December 12, 2005 by Lorenz, et al. The content of this document is included in the disclosure content of the present application by reference.

  The present invention uses magnetic resonance imaging (MRI) and real-time monitoring of cardiac function performed by interleaving with stress tests or imaging used to guide or perform interventional cardiovascular procedures About.

  Monitoring the electrocardiogram (ECG) of ischemic changes that occurred during dobutamine stress testing or during MR guided intervention is not possible due to magnetohydrodynamic effects in the MR environment. The reason for this effect is that ions in the blood flowing through the body in the presence of a static magnetic field generate their own voltage. If the patient is in this static magnetic field, such additional voltage will add to the ECG and cause distortion. However, cardiac function must be monitored for possible ischemic changes or dyssynchrony during MR guided circulatory interventional procedures or stress testing. Heart wall motion and global cardiac function can be evaluated in real time using MR and interleaving with another type of MR acquisition during the procedure. Thus, one approach for monitoring ischemic changes is to measure the heart wall motion qualitatively at the time of image acquisition. However, cardiac function measurements that are automatically detected and provided as continuous feedback during time intervals (eg, during dobutamine dose determination or catheter placement) when the physician cannot see a functional image It is advantageous to obtain

  A major problem in patient monitoring performed on ischemia in MRI is that static magnetic fields, RF pulses, and magnetic field gradient switching all distort the ECG and no longer exhibit diagnostic features in ischemia detection. When an ischemia occurs in the heart, it first causes perfusion damage, followed by abnormal heart wall motion, followed by ECG changes, and finally chest pain. Currently, in global patient monitoring, external monitoring systems with ECG, blood pressure, pulse, oximetry, and possibly invasive blood pressure are used with MR. ECG is only used to monitor heart rate. Use real-time scan plane changes to check for functional changes while performing interventional procedures. However, even if this is done, continuous monitoring cannot be performed. Perfusion scans can also be visually viewed during acquisition to check for changes in cardiac perfusion. However, this cannot be repeated as often as ventricular function imaging during the examination. This is because it is necessary to manage the contrast medium.

  However, by converting real-time MR to m-mode (motion mode) display similar to echocardiography, (a) heart function can be displayed easily and continuously, and (b) real-time segmentation and ventricle. Automatic extraction of function parameters is simplified. The concept of interleaved MR imaging is used in the field of navigator gating, where low-resolution images are used to sample bulk motion and gating acquisitions by interpreting such images. However, the present inventor has not found any publication on interleaved physiological monitoring scans with imaging scans.

  It is an object of the present invention to provide a novel system and method in a system and method for image-based physiological monitoring.

In the present invention, the above problem is
Selecting a plane through the heart from which cardiac function data is to be acquired;
Obtaining a two-dimensional image from the selected surface, wherein the image includes a plurality of intensities defined in a domain of points on a 2D grid;
Selecting the center of the ventricle of the heart;
Projecting a plurality of beams onto the image through the center of the ventricle, wherein each beam comprises a substantially collinear subset of points in the image;
Sampling the image intensity along each point of the beam;
The step of acquiring a two-dimensional image and the step of projecting a plurality of virtual beams onto the image are repeated, and image intensity is calculated to form a sequence of temporal measurement results, where each beam is And a step of plotting the image intensity along the time dependently.

The problem is
Defining a plurality of planes through the heart from which cardiac imaging data is to be acquired;
Obtaining a sequence of two-dimensional images from the selected surface, wherein each image includes a plurality of intensities defined in a domain of points on a 2D grid;
Selecting one or more of the series of images to provide imaging data for monitoring cardiac function and adapting the remaining unselected images for diagnosis;
Repeating the steps of obtaining a sequence of two-dimensional images from the selected surface and selecting one or more for monitoring cardiac function from the sequence, wherein a time series of images is It is also solved by a method characterized in that it comprises the steps of alternately obtaining an image for monitoring the function and an image for diagnosis.

  Embodiments of the invention described herein generally include methods and systems for assessing cardiac function in real time by 2D MR imaging. By evaluating the cardiac function visually or displaying it as the m mode, the temporal history can be evaluated. The m-mode display and the resulting measurement results can be interleaved by different types of acquisition and can be included in a real-time interface for scanner control. In one embodiment of the system and method of the present invention, ischemia detection can be performed early using real-time imaging interleaved with “standard” imaging based on heart wall motion. By interleaving MR data acquisition, a portion of the data can be used continuously for direct cardiac function monitoring while the remainder of the data can be used for diagnostic or interventional purposes. Heart wall motion and ventricular function parameters are extracted from real-time imaging, and the method of detecting changes alerts the operator that the cardiac function has changed. In addition to ischemia monitoring, the method according to embodiments of the present invention is used to detect changes in the pattern of contractions (due to conduction abnormalities) and to detect changes in cardiac output (real-time flow imaging). ) Or changes in perfusion pattern (real-time myocardial perfusion imaging). Embodiments of the present invention extract relevant features from interleaved monitoring data, clinically identify relevant changes, and assist the operator in monitoring cardiac function. In a system according to an embodiment of the invention for performing real-time image interleaving, cardiac function can be quasi-continuously monitored in an MR environment where the ECG exhibits no diagnostic features. In addition, the potential for providing more sensitive feedback to changes in cardiac function than can be achieved by global ECG monitoring, regardless of its use in an MRI environment, is potentially hidden. This increase in sensitivity is based on the fact that changes in heart wall motion and myocardial perfusion precede an ECG change during ischemia. The concept of performing some other diagnostic procedure or therapeutic imaging while using direct measurement of cardiac function in the background is another imaging mode such as computed tomography (CT) or ultrasound (US). It can also be applied to.

  The validation results demonstrate the utility of a global real-time evaluation method of left ventricular function that can be performed in combination with stress testing or interventional procedures. It is feasible to incorporate into a real-time environment using one or two faults for functional monitoring and additional temporally interleaved faults for procedure guides.

  Embodiments of the invention as described herein are generally based on imaging used to guide or perform stress tests or interventional circulatory procedures using magnetic resonance imaging (MRI). Includes systems and methods for performing real-time monitoring of cardiac function by interleaving. Accordingly, various modifications and alternative constructions of the invention are possible, of which specific embodiments are shown by way of example in the drawings and will now be described in detail. However, this description is not intended to limit the invention to the specific form disclosed herein, and on the contrary, the invention is not limited to all modifications, equivalent constructions and alternatives falling within the spirit and scope of the invention. Should be included.

  The term “image” as used herein refers to multidimensional data composed of discrete image elements (eg, pixels in a 2D image and voxels in a 3D image). This image is, for example, a medical image of a subject collected by computed tomography, magnetic resonance imaging, ultrasound, or any other medical imaging system known to those skilled in the art. This image can also be supplied from a non-medical context, for example from a remote sensing system, an electron microscope or the like. Images can be viewed as a function of R3 to R, but the method of the present invention is not limited to such images and can be applied to images of any dimension, such as 2D pictures or 3D volumes, for example. Can do. For 2D or 3D images, the domain of the image is typically a 2D or 3D rectangular array, where each pixel or voxel consists of a set of 2 or 3 mutually orthogonal axes. Addressed as a reference. The terms “digital” and “digitized” as used herein are suitably obtained in digital form or digitized form obtained via a digital acquisition system, or by conversion from an analog image. Refers to an image or volume in digital form or digitized form.

  A system according to an embodiment of the present invention has imaging data acquired by interleaving, some of which is used to monitor cardiac function and the rest is used for diagnostic purposes, for example, Used for the purpose of guiding a procedure, for example in a method for extracting relevant features from an image relating to cardiac function, and in a method for detecting clinically relevant changes in said features and alerting an operator . The features include, for example, endocardial and epicardial boundaries, aortic volume flow, ventricular shape change, myocardial perfusion, and the like.

  The scan front end provides a user interface (UI), which allows control of scan plane orientation and feature extraction from previously acquired images or from images acquired from any common parameter optimization algorithm Both of the imaging parameter selections based on the feature extraction are performed.

The scan front end according to the embodiment of the present invention can be developed by modifying existing commercially available software. An example front end was developed and modified with additional C ++ code using a combination of RadBuilder available from Siemens Corporate Research. This front end does not limit the invention. The family of manipulators created with RadBuilder restricts the movement of the tomography to restricted movements in the 3D window and prevents the user from losing direction while moving the tomography. Fault operation can be performed in one of several modes:
Free rotation and translation Translation in the same plane Rotation around the fault center Translation along the fault normal

  The front-end application can run on a separate workstation connected to the scanner host computer via Ethernet, or it can run directly on the scanner host computer. Front-end applications can communicate with custom applications running on the host via socket communication using current open-running protocols. With such an application, the tomographic operation performed on the front-end application in 3D by graphics is converted into scanner coordinates and sent to a running protocol for updating the tomographic position. After image acquisition, the image reconstruction program sends the image data to a standard database and back to the front-end application via socket communication.

In addition to displaying the newly acquired slice in real time, the UI's main viewing window can be used to visualize renderings made from previously acquired images. An example UI has a window on the left side that is used for two-dimensional tomographic positioning and viewing both previous images and real-time updates, and a small window at the bottom that contains a history of acquisitions that have already taken place. These images can be moved to the left window or the main viewing / operation window as required. The front end has at least two modes of operation:
(1) Visualization without moving any scanner update by moving the imaging surface in the main window (2) Active tomographic operation to move the scanning surface

  According to embodiments of the present invention, cardiac function such as heart wall motion and intracardiac flow is monitored by locating one or more 2D images in a plane. These 2D images may not be in the same plane. FIG. 1 shows an example of an imaging / monitoring configuration, showing one monitoring tomography and two imaging tomography. These are orthogonal to each other as shown in the central 3D display surface. This configuration example does not limit the present invention. The three panels on the left are three 2D display surfaces and the monitoring display surface is at the top. The panel in the lower right corner is a graph of monitoring data derived from the monitoring tomography, showing for example heart wall motion, heart dimensions, flow, perfusion, etc. depending on time (horizontal axis).

  MR images can be acquired in real time with a temporal resolution of about 30-50 ms at present. Contrary to using partial data interleaving to form an image across multiple interleavings, a complete image can be acquired and interleaved in a timely manner. Therefore, the resolution of the final image is not reduced.

  In an example time chart, the time order of acquisition is M, I, M, I, M, I,. . . become that way. Here, M represents an image acquired for monitoring, and I represents an image acquired for standard imaging. Such an acquisition order in which every other image is used for monitoring does not limit the invention, and in general every nth image can be used for monitoring. In real time, the only difference between the monitoring image and the imaging image is the position in the heart. Monitoring images can be acquired at a pre-selected location of interest by the user or automatically acquired as defined by another algorithm and used to focus on high risk areas And the imaging tomography can be focused on another part of the heart as selected by the user or automatically selected.

  The user is free to locate the image and distinguish it as an image used for monitoring. Additional image planes can be located by the user and used for diagnostic purposes or used to guide interventional procedures. In doing so, the monitoring images are analyzed on a continuous basis during the scan to extract cardiac function parameters or at intervals determined by the user.

  FIG. 2 illustrates an example image-based monitoring configuration according to one embodiment of the present invention for performing 2D radio frequency (RF) excitation for real-time flow measurements interleaved by multiple tomographic imaging. The upper right panel is a 3D display surface showing how imaging and monitoring tomography are located in the heart, and the lower right panel is the display surface of the monitoring tomography (xy plane through the heart). Yes, the position of the 2D RF excitation beam is shown. The lower right corner graph shows the real-time discharge flow rate detected by the excitation beam. The excitation beam can also be replaced with a 2D slice by flow coding.

  In order to optimize the trade-off between the speed and accuracy of the calculated cardiac function parameter, the user of the system can define the desired measurement accuracy. Depending on this measurement accuracy, the system automatically adjusts the spatial and temporal sampling. Spatial sampling includes the number of slices in parallel, the image resolution per slice, and the number of projections in the case of the 1D scheme. Temporal resolution is defined as the number of time frames acquired per slice / projection in a cardiac cycle.

  According to an embodiment of the present invention, real-time MR is converted into an m-mode (motion mode) display similar to echocardiography, so that cardiac functions can be easily and continuously displayed, and real-time segmentation and Automatic extraction of ventricular function parameters is simplified. An m-mode display according to an embodiment of the present invention can be included in a real-time interface for scanner control.

FIG. 5 is a flowchart illustrating a method according to one embodiment of the present invention for extracting cardiac function parameters in real time using m-mode MR. Referring now to the drawings, in order to create an m-mode MR in step 51, the user manually selects the heart cross section from which monitoring data is to be acquired, and positions the center of the left ventricle in only one short axis tomography decide. When an MR image slice is acquired in step 52, four conformal projections are propagated through the left ventricle as m-mode projection beams in step 53, where they intersect at 45 ° segments. In step 54, each image slice is periodically sampled for intensity along the projection beam, thereby forming a substantially collinear subset of points. According to an embodiment of the present invention, this projection is one pixel wide. When the time series of image slices is acquired, the image intensity along each beam (M _{i, t} ) is plotted in step 55 as a function of time. Linear interpolation is performed in step 56 to increase and smooth the number of grid points. In each beam M _{i, t} , an endocardial contour is detected in step 57.

In one embodiment of the invention, the contour is detected using a corrected horizontal 1D canny filter. Smoothing is performed using a 1D Gaussian filter with σ = 1 in the y direction, and the gradient calculation is based on the second derivative of above Gaussian in the y direction. The canny threshold is evaluated depending on the histogram of M _{i, t} , and in this embodiment of the invention, it is assumed that 75% of the pixels do not belong to the heart wall.

In another embodiment of the present invention, an epicardial contour is detected using an active contour model, also known as a “snake” algorithm. Such an active contour model reduces the image energy function. The energy function E = λE _int + γE _img snake model is expressed and external constraints force _{E img} term that depends on the image, by the internal constraint forces _{E img} term which depends on the contour of the The E _int is a weighted sum of two terms based on a first snake derivative and a second snake derivative representing elasticity and flexural force acting on the contour. E _img includes a linear combination of vertical image gradient and low-pass filtered zero crossing. In the applied open snake model, the coordinate value of the horizontal axis is an integer representing the time point of m-mode projection, and this coordinate axis is fixed. Only the position value of the snake point on the vertical axis moves. In snake initialization, the endocardial contour is translated in the direction of the expected epicardial contour. By reducing the energy function, the active contour converges.

  In step 58, the distance between the detected endocardial contours is calculated. The maximum interval (time dependent) in each projection is the LV end diastolic (ED) diameter and the minimum interval is the end systolic (ES) diameter.

  According to an embodiment of the present invention, cardiac wall contour data is used to estimate the ejection fraction (EF) in a tomogram. In this embodiment, referring again to FIG. 5, for each heartbeat in step 59, the ventricular area is calculated with ED and ES, using the area formed by the eight vertices of the endocardial contour. Here, the ejection fraction EF is calculated as follows.

EF (%) = 100 × (EDV−ESV) / EDV,
Here, EDV is the end-diastolic ventricular area and ESV is the end-systolic ventricular area.

Tests of embodiments of the present invention were performed using data obtained from three healthy subjects. A short-axis real-time image midway between the mitral valve surface and papillary muscle level during free breathing was acquired using the TrueFISP (SSFP) series. Here, TE / TR / flip angle 0.87 / 1.74 / 60, FOV = 160 × 380 mm ² , 88 × 128 matrix, GRAPPA acceleration is × 2, fault thickness is 8 mm, time resolution Is 54 ms and the number of frames is 128.

  In the initial confirmation, the contour data derived from the m mode was superimposed on the real-time image at the apex and observed as a sine loop for visual confirmation. The contours were also displayed on the m-mode display for visual inspection, and the results were compared with EF based on manually drawn contours to evaluate statistically significant differences. FIG. 3A shows eight m-mode projections along with a short axis (SA) tomographic example, and FIG. 3B shows a derived contour superimposed on an m-mode display. . Here, the ED diameter and ES diameter are shown as shown. FIG. 3C shows an SA tomogram in which triangles are overlaid for calculating the ventricular area. A visual qualitative inspection of m-mode derived contours performed on an m-mode display is shown in FIG. 3B, indicating good agreement. The average area of this subject was in the range of 55-70% expected with normal EF. The change in EF for each heartbeat was in the range of 2-9%. Thus, despite using only one slice to estimate the EF to be examined, an EF estimation method according to embodiments of the present invention and an EF calculation based on manually drawn contours in a single slice No statistically significant difference was found during

  This method of embodiments of the present invention for calculating area ejection fraction can be extended for automatic measurement of local wall thickness and percentage wall hypertrophy. An automated method of assessing wall thickness in real time provides a platform for detecting changes during interventional procedures or stress tests with computer assistance. The wall thickness at each time point is calculated as the interval between the endocardial position value and the epicardial position value. The epicardial contour can be calculated using the same algorithm used for the endocardial contour.

  FIGS. 4A to 4C show outlines displayed in the MRI m mode for visual evaluation of accuracy. FIG. 4A shows a projection line, and FIG. 4B shows an MRI m-mode image having an automatically detected contour. Again, the wall thickness based on the calculated contour and the wall thickness based on the manually drawn contour were compared as an initial confirmation step. FIG. 4 (C1) shows the automatic wall thickness of the upper myocardium in the m-mode image versus the manual wall thickness, and FIG. Wall thickness versus manual wall thickness.

  In an embodiment of the present invention, the active contour model was verified with 12 MRI m-mode sequences (4 projections for 3 subjects each). Visual qualitative inspection of MRI m-mode derived contours on the MRI m-mode display showed good agreement. The average local heart wall thickness was in the range of 7.1 to 11.8 mm, which was equivalent to that expected with normal heart wall thickness. The table below shows the mean and standard deviation values of the minimum and maximum wall thicknesses of the myocardium and septum detected over the entire cardiac cycle in one m-mode train in the three collaborators. Is shown.

  Overall, there are good points between manual and automatic contours, except for some points in late dilation where manual contour identification was difficult due to rapid heart wall movement. Correlation was seen.

  According to another embodiment of the present invention, perfusion is monitored using interleaved real-time cardiac monitoring. A beam as described above in connection with ejection fraction is projected onto the heart with real-time images and analyzed for signal time characteristics over a predetermined time (typically 60 seconds) after intravenous application of contrast agent. did. Perfusion related parameters derived from signal strength time characteristics or absolute blood flow expressed in ml / min / 100 g can be compared to previously detected parameter characteristics. The perfusion-related parameters are derived from, for example, the slope or the area under the curve. Alternatively, the signal intensity time characteristics of individual pixels in the image can be analyzed to create a parameter map of semi-quantitative parameters.

  FIG. 6A shows a short axis tomographic parameter map in which the color of each pixel is associated with the peak signal intensity obtained when the contrast agent first passes at the position. Since the peak signal intensity is proportional to tissue perfusion, the normal perfusion region is distinguished from the region with abnormal perfusion, as shown by the bright edges extending from 8 o'clock to 12 o'clock in the myocardium. can do. FIG. 6B shows signal intensity time curves belonging to individual pixels at the tomographic position shown in FIG. Here, a baseline and a peak point are identified.

  According to another embodiment of the present invention, interleaved real-time cardiac monitoring is used to monitor cardiac flow. In the case of using pencil beam 2D excitation, the phase of the MRI signal is encoded with velocity and displayed in a time dependent manner. Parameters relating to ventricular function, such as time versus peak flow rate, ratio of reverse flow rate, peak flow rate, etc. can be derived. In the case where a stream-encoded 2D image is used in place of 2D pencil beam excitation, a segmentation method is applied to the portion of the image that includes the portion of the heart of interest, such as the aortic valve, for example. Parameters can be derived from the selection.

  The method according to embodiments of the present invention for detecting changes in features included in the derived data can be used to alert a physician that cardiac function is changing. For example, using trend analysis to detect whether the ejection fraction has changed significantly over time or whether it has changed relative to a specific point in time in the procedure defined by the operator For example, it can be detected whether the ejection fraction is changing relative to a baseline comparable to the time of stent placement.

  According to another embodiment of the present invention, change detection can be applied to the absolute value of the extracted physiological parameter or applied to the raw data or the change of the extracted parameter over time.

  For each feature derived from the monitoring image, a clinically relevant threshold can be determined. For example, a 5% change in the ejection fraction can be considered a significant change. A current average over the past n heartbeats can be maintained, and a change can be detected by comparing the parameter value at each heartbeat with the current average. In an alternative embodiment of the invention, each heart rate can be compared to a baseline value at the start of the test.

  Continuous signal change analysis or trend analysis is common in monitoring devices. However, it should be understood that in one embodiment of the invention, it is possible to provide raw data about the image before deriving features from the image and develop a change detection method or intermediate step for the image. For example, cross-correlation analysis can be used to detect portions of the image that have changed since a predefined time point.

  It should be understood that embodiments of the present invention can be implemented in various forms of hardware, software, firmware, special purpose processing, or combinations thereof. In one embodiment, the present invention can be embodied in software as an application program that can be implemented on a computer readable program storage device. This application program can be uploaded to a device having any suitable architecture or can be executed by such a device.

  FIG. 7 is a block diagram illustrating an example computer system for implementing a method for monitoring a heart on a real-time image basis, according to one embodiment of the present invention. Referring now to FIG. 7, a computer system 71 for implementing the present invention includes, among other things, a central processing unit (CPU) 72, a memory 73, and an input / output (I / O) interface 74. The computer system 71 is generally connected to a display 75 and various input devices 76 such as a mouse and a keyboard via an I / O interface 74. Support circuitry can include circuits such as caches, power supplies, clock circuits, and communication buses. Memory 73 may include random access memory (RAM), read only memory (ROM), disk drive, tape drive, or combinations thereof. The present invention can be implemented as a routine 77 stored in the memory 73 and executed by the CPU 72 that processes the signal from the signal source 78. Thus, the computer system 71 itself is a general-purpose computer system. However, when the routine 77 of the present invention is executed, the computer system 71 is an application-specific computer system.

  The platform of computer system 71 also has an operating system and microinstruction code. The various processes and functions described herein can be part of microinstruction code that is executed via an operating system, or part of an application program (or a combination thereof). In addition, various other peripheral devices, such as additional data storage devices and printing devices, can be connected to the computer platform.

  Further, since some of the system components and method steps shown in the figure can be implemented in software, the actual connection between system components or processing steps will depend on how the invention is programmed. It should be understood that these may vary. If the idea of the present invention disclosed in the present specification is provided, those skilled in the art can deduce similarities to the embodiments or configurations of the present invention.

  Although the invention has been described in detail with reference to advantageous embodiments, those skilled in the art can make various modifications and substitutions of the invention without departing from the spirit and scope described in the claims. I can recognize that.

2 is an example configuration of real-time cardiac function imaging / monitoring according to one embodiment of the present invention. 2 is an example image-based monitoring configuration for performing 2D radio frequency (RF) excitation for real-time flow measurement according to one embodiment of the present invention. FIG. 6 shows an example of a short axis fault with m-mode projection shown and an outline derived therefrom and superimposed on an m-mode display according to one embodiment of the present invention. Fig. 6 shows a contour displayed in MRI m mode for visual assessment of accuracy according to one embodiment of the invention. 2 is a flow chart illustrating a method according to one embodiment of the present invention for extracting cardiac function parameters in real time using m-mode MR. FIG. 4 is a parameter map of a signal strength-time curve obtained from a short axis fault and individual pixels included in the short axis fault according to one embodiment of the present invention. FIG. 2 is a block diagram illustrating an example computer system according to one embodiment of the present invention for performing a method for monitoring cardiac function in real time using magnetic resonance imaging (MRI).

Claims (33)

In a method for monitoring heart function in real time,
Selecting a plane through the heart from which cardiac function data is to be acquired;
Obtaining a two-dimensional image from the selected surface, wherein the image includes a plurality of intensities defined in a domain of points on a 2D grid;
Selecting the center of the ventricle of the heart;
Projecting a plurality of beams onto the image through the center of the ventricle, wherein each beam comprises a substantially collinear subset of points in the image;
Sampling the image intensity along each point of the beam;
The step of acquiring a two-dimensional image and the step of projecting a plurality of virtual beams onto the image can be repeated to calculate the image intensity and plot the image intensity along each of the beams as a function of time. Forming a sequence of temporal measurement results.   The method of claim 1, wherein additional intensity is linearly interpolated between sampled points along each of the beams.   The method of claim 1, wherein an endocardial contour is detected from the plurality of beams.   4. The method of claim 3, wherein the endocardial contour is detected using a horizontal 1D canny filter.   Active reducing the energy of the endocardial contour, which is a weighted sum of an external constraining force determined by the image and an internal constraining force depending on the shape of the endocardial contour 4. The method of claim 3, wherein the detection is performed using a simple contour model. Calculating the interval between the detected endocardial contours;
4. The method of claim 3, wherein the time-dependent maximum interval in each projection beam is the end-diastolic (ED) diameter and the minimum interval is the end-systolic (ES) diameter. Calculate the ventricular area from the area formed by the points of the endocardial contour,
Calculate the ejection fraction from (EDV-ESV) / EDV,
7. The method of claim 6, wherein EDV is end-diastolic ventricular area and ESV is end-systolic ventricular area.   8. The method of claim 7, wherein the ventricle is a left ventricle. Detecting an epicardial contour from the plurality of beams;
4. The method of claim 3, wherein heart wall thickness is calculated from the endocardial contour and the epicardial contour as a function of time. Use image intensity time curves along each of the beams to detect changes in values of physiological parameters over time;
Detection of a change in value includes selecting a clinically relevant change threshold, maintaining a current average of the parameter values over a predetermined number of previous heart beats, parameter values for each heart beat and the current The method of claim 1, comprising comparing to an average value or a baseline value of the parameter.   The method of claim 1, wherein the acquired image is part of an image sequence that includes an image acquired for diagnostic purposes. In a method for monitoring heart function in real time,
Defining a plurality of planes through the heart from which cardiac imaging data is to be acquired;
Obtaining a sequence of two-dimensional images from the selected surface, wherein each image includes a plurality of intensities defined in a domain of points on a 2D grid;
Selecting one or more of the series of images to provide imaging data for monitoring cardiac function and adapting the remaining unselected images for diagnosis;
Repeating the steps of obtaining a sequence of two-dimensional images from the selected surface and selecting one or more for monitoring cardiac function from the sequence, wherein a time series of images is A method comprising alternately obtaining an image for monitoring a function and an image for diagnosis. For each monitoring image, select the center of the heart's ventricle,
Projecting multiple beams onto the image through the center of the ventricle;
Each beam includes a subset of points that are substantially collinear in the image;
13. The method of claim 12, wherein the image intensity along each point of the beam is sampled to form a series of temporal measurements that can plot the image intensity along each of the beams as a function of time.   13. The method of claim 12, wherein every nth image is selected for monitoring cardiac function.   The method of claim 13, wherein the image intensity temporal characteristics are analyzed over a predetermined period after intravenous application of contrast agent. The image is a magnetic resonance (MR) phase image;
Plot the velocity-encoded MR phase as a function of time;
13. The method of claim 12, wherein functional parameters are derived from a phase-time curve, wherein one or more of time versus peak flow rate, reverse flow rate ratio, and peak flow rate are included.   17. The method of claim 16, wherein each phase image is a 1D image created using pencil beam 2D high frequency excitation to derive measurement results from a tissue column.   The method of claim 16, wherein the phase image is a 2D image created from a velocity encoded MR image. A computer readable program storage device for executing a program of instructions executable by a computer,
The instructions are in the form of instructions for performing the steps of the method for monitoring cardiac function in real time,
The method
Selecting a plane through the heart from which cardiac function data is to be acquired;
Obtaining a two-dimensional image from the selected surface, wherein the image includes a plurality of intensities defined in a domain of points on a 2D grid;
Selecting the center of the ventricle of the heart;
Projecting a plurality of beams onto the image through the center of the ventricle, wherein each beam comprises a substantially collinear subset of points in the image;
Sampling the image intensity along each point of the beam;
The step of acquiring a two-dimensional image and the step of projecting a plurality of virtual beams onto the image can be repeated to calculate the image intensity and plot the image intensity along each of the beams as a function of time. A computer-readable program storage device comprising: a step of obtaining a series of temporal measurement results.   20. The computer readable program storage device of claim 19, wherein the method linearly interpolates additional intensities between sampled points along each of the beams.   20. The computer readable program storage device of claim 19, wherein the method detects an endocardial contour from the plurality of beams.   The computer readable program storage device of claim 21, wherein the endocardial contour is detected using a horizontal 1D canny filter.   The endocardial contour is an active reducing energy that is a weighted sum of an external constraining force determined by the image and an internal constraining force that depends on the shape of the endocardial contour The computer readable program storage device of claim 21, wherein the computer readable program storage device is detected using a simple contour model. In the method, the distance between the detected endocardial contours is calculated,
The computer readable program storage device of claim 21, wherein a time-dependent maximum interval in each projection beam is an end-diastolic (ED) diameter and a minimum interval is an end-systolic (ES) diameter. In the method, the ventricular area is calculated from the region formed by the points of the endocardial contour,
Calculate the ejection fraction from (EDV-ESV) / EDV,
25. The computer readable program storage device according to claim 24, wherein EDV is a ventricular area at end diastole and ESV is a ventricular area at end systole.   26. The computer readable program storage device of claim 25, wherein the ventricle is a left ventricle. In the method, an epicardial contour is detected from the plurality of beams,
The computer readable program storage device of claim 21, wherein heart wall thickness is calculated from the endocardial contour and the epicardial contour as a function of time. The method uses an image intensity time curve along each of the beams to detect changes in values of physiological parameters over time,
Detection of a change in value includes selecting a clinically relevant change threshold, maintaining a current average of the parameter values over a predetermined number of previous heart beats, parameter values for each heart beat and the current 20. The computer readable program storage device of claim 19, comprising comparing to an average value or a baseline value of the parameter.   The computer-readable program storage device according to claim 19, wherein the acquired image is a part of an image sequence including an image acquired for diagnostic purposes.    20. The computer readable program storage device of claim 19, wherein the method analyzes the image intensity time characteristics over a predetermined period after intravenous application of a contrast agent. In the method, the image is a magnetic resonance (MR) phase image;
Plot the velocity-encoded MR phase as a function of time;
20. The computer readable program storage device of claim 19, wherein the functional parameters are derived from a phase-time curve, wherein one or more of time versus peak flow rate, reverse flow rate ratio and peak flow rate are included.   32. The computer readable program storage device of claim 31, wherein each phase image is a 1D image created using pencil beam 2D high frequency excitation to derive measurement results from a tissue column.   32. The computer-readable program storage device according to claim 31, wherein the phase image is a 2D image created from a velocity-encoded MR image.