WO2011039685A1

WO2011039685A1 - Four-dimensional roadmapping usable in x-ray guided minimally invasive cardiac interventions

Info

Publication number: WO2011039685A1
Application number: PCT/IB2010/054306
Authority: WO
Inventors: Eberhard Sébastien HANSIS; Michael Grass; Dirk Schaefer
Original assignee: Koninklijke Philips Electronics N.V.; Philips Intellectual Property & Standards
Priority date: 2009-09-30
Filing date: 2010-09-24
Publication date: 2011-04-07

Abstract

A method for four-dimensional roadmapping usable in coronary artery interventions is described. In an embodiment of the method a first cardiac data set (1) is acquired together with a cardiac phase indicating signal (3) (S1). A first three-dimensional reconstruction (7) from images of the cardiac data set (1) is generated using the cardiac phase indicating signal (3) (S3). Therein, the first three-dimensional reconstruction (7) corresponds to a first cardiac phase (b). Furthermore, a combined patient-specific model (9) of heart motion is generated based on the first three-dimensional reconstruction (7) and a motion information (5) (S5). Therein, the motion information (5) indicates at least one of a motion direction, a motion speed, a motion acceleration and extremal states of the heart. A four- dimensional road map (11) is generated based on the patient-specific model (9) of heart motion and the first three-dimensional reconstruction (7) (S7).

Description

Four-dimensional roadmapping usable in X-ray guided minimally invasive cardiac

interventions

FIELD OF THE INVENTION

The present invention relates to the field of four-dimensional roadmapping usable in minimally invasive cardiac interventions. Particularly, the present invention relates to a method and an apparatus for simultaneously and dynamically visualizing coronary information and X-Ray projections. Furthermore, the present invention also relates to a computer program element adapted for controlling such method when executed on a computer and to a computer-readable medium on which such computer program element is stored. BACKGROUND OF THE INVENTION

In many examination and intervention procedures it is important to provide a visualization of an area being examined or manipulated. Particularly, when moving objects such as a heart or a lung are examined high quality images of each motion state at any time of the procedure may be essential.

During an examination of the heart a well established method for providing images is angiography, where after an injection of a contrast agent (CA) a series of images is acquired for example with the help of X-rays.

Also catheter-based coronary artery interventions may be performed under X- ray guidance, using an interventional C-arm system. The fluoroscopy sequences show interventional devices, but coronary arteries are only visible as long as contrast agent is present therein. Even then, projections offer only a two-dimensional view of the coronaries.

Methods are known for creating a three-dimensional reconstruction of the coronary arteries from a rotational X-ray angiography sequence. But the resulting static three- dimensional view of the coronaries may be not optimal for roadmapping.

A sequence of three-dimensional reconstructions in multiple cardiac phases, to be displayed as a four-dimensional movie, may not be feasible due to long computation times and high memory requirements.

Thus, there may be a need for an improved method or apparatus for providing a four-dimensional patient specific coronary roadmap with enhanced quality.

SUMMARY OF THE INVENTION

These needs may be met by the subject-matter according to the independent claims. Advantageous embodiments of the present invention are described in the dependent claims.

According to a first aspect of the present invention a method for four- dimensional road mapping usable in minimally invasive cardiac interventions is provided. The method comprises the following steps: acquiring a first cardiac dataset together with a cardiac phase indicating signal; generating a first three-dimensional reconstruction from images of the cardiac data set using the cardiac phase indicating signal, wherein the first three-dimensional reconstruction corresponds to a first cardiac phase ; generating a combined patient-specific model of heart motion based on the first three-dimensional reconstruction and the motion information, wherein the motion information indicates at least one of a motion direction, a motion speed, a motion acceleration and extremal states of the heart; and generating a four- dimensional road map based on the patient-specific model of heart motion and the first three- dimensional reconstruction.

In other words, the first aspect of the present invention may be seen as based on the idea to provide a moving three-dimensional visualization of the coronary artery tree by combining information about the motion of the heart with a three-dimensional reconstruction generated for example from a gated rotational coronary angiography sequence at a phase of low motion of the heart. By fusing the patient-specific three-dimensional reconstruction and the information on the motion of the heart a patient-specific model is generated which may be displayed combined with further currently acquired X-ray projections in real-time. Thus, by combining general knowledge and patient specific data a patient specific model of heart motion representing the heart in different/arbitrary cardiac phases is received.

With the inventive method road mapping may be greatly improved and thereby intervention navigation may be made easier. The method according to the first aspect of the present invention may further facilitate the generation of a four-dimensional road mapping overlay, with feasible requirements with respect to computation time. The resulting combined display of the moving three dimensional coronary artery tree together with interventional devices can greatly help in navigation. The amount of contrast agent required for an intervention might be reduced and intervention times may be shortened. Four-dimensional road mapping may refer to moving three-dimensional frames. For example, a four-dimensional road map may represent a sequence in time of three- dimensional images for example showing one or several cycles of heart motion.

Road mapping may denote the visualization of geometrical data possibly acquired in the past and processed. Furthermore in the context of road mapping the geometrical data may be linked to further data currently acquired for example to provide a navigation support.

The four-dimensional road map is usable in minimally invasive cardiac interventions. The minimally invasive cardiac interventions may for example be coronary artery interventions and interventions taking place in the coronary veins like the placement of pacemaker leads. Furthermore, such interventions may for example be catheter-based procedures executed by a physician. The procedures may concern vessels of the heart. An example for a coronary artery intervention is the dilation of a narrowed vessel by inflating a balloon inside the vessel, or the implantation of a stent, a device that supports the vessel wall and keeps the vessel lumen open.

In a first step of the method a first cardiac data set is acquired together with the cardiac phase indicating signal. Therein acquiring may denote for example retrieving information from a memory database. Alternatively or additionally acquiring may denote gathering images by a technique like for example X-ray, computer tomography (CT) or magnetic resonance (MR) imaging.

The first cardiac data set may be a sequence of images of frames in which vessels, particularly vessels of the heart, are visible. Therein "cardiac" denotes that the data set is related to the heart. The first cardiac data set may for example be acquired during a contrast agent is present in the vessels. The first cardiac data set particularly may comprise information about the vessels of the heart and the heart cycle of a patient. Preferably, the first cardiac data set comprises information on more than one complete heart cycle. For example, the first cardiac data set may be a fluoroscopic sequence or a rotational coronary angiography sequence.

A cardiac phase indicating signal is acquired for example simultaneously with the first cardiac data set. The cardiac phase indicating signal may for example be an external signal like an electrocardiogram (ECG) or alternatively a signal derived from the images of the first cardiac data set. For example, the cardiac phase indicating signal may be derived from the flow rate of the blood or from the change in the geometry of the heart chambers. An ECG signal may correspond to a graphical recording of the cardiac cycle produced by an electro cardio graph.

Therein a cardiac phase may correspond to a phase of the cardiac cycle like for example the systole or the diastole. Particularly a cardiac phase may indicate the state of the systole and diastole like for example an end systole and a late diastole.

According to the inventive method a first three-dimensional reconstruction is generated. Furthermore a combined patient-specific model of heart motion is generated. Moreover, a four-dimensional road map is generated. Therein generating may denote producing or creating of a data set and possibly visualization thereof. The generating may for example be done by computation e.g. by carrying out processing steps for example applied to images of the first cardiac data set.

The first three-dimensional reconstruction may be a tomographic reconstruction from a rotational X-ray and/or angiography sequence. It may be reconstructed by gating of a first projection data set acquired during a first desired cardiac phase of a plurality of cardiac cycles. For example, only images acquired in late diastole are selected from the first cardiac data set with the help of e.g. gating with the cardiac phase indicating signal. The images selected in this way may correspond to projections from different viewing directions in the same phase of the heart cycle. Therein for example only an area of interest such as a desired coronary artery branch e.g. left or right coronary artery are reconstructed.

Therein gating may refer to selecting X-ray projections from a projection sequence that were acquired in or close to a desired phase of the heart motion cycle.

The cardiac phase indicating signal is used to gate the first cardiac data set, thereby selecting images corresponding to a low phase of motion.

A first cardiac phase may be a state of the heart corresponding to relatively low heart motion i.e. corresponding to a maximum contraction (end systole) or near maximum dilation (late diastole).

A combined patient-specific model of heart motion is generated by combining and/or fusing and/or superimposing the first three-dimensional reconstruction with motion information. The combining may be done by processing and computing steps. The combined patient-specific model of heart motion is for example generated by superimposing general knowledge and patient-specific data.

The motion information may be general knowledge about heart motion e.g. knowledge about when the heart motion speed is maximum and when it is lower, and information that enables extrapolation of the maximally dilated motion state at end systole. Motion information may also be derived from the first cardiac data set for example by direct analysis of the acquired rotational X-ray projections of the first cardiac data set. For example, the motion information may be derived by extracting a motion function by cross-correlating line integral profiles from adjacent projections of the first cardiac data set. For example, motion information indicates at least one of a motion direction, a motion speed, a motion acceleration and extremal states of the heart. Therein, the extremal state of the heart may be the heart's shape at maximum contraction and maximum dilation.

Motion direction, motion speed and motion acceleration may be represented by vectors for each pixel of the images of the first cardiac data set. They may be derived from general knowledge such as mean motion models derived independently from a large number of patients or from patient-specific data or alternatively a combination of both.

A four-dimensional road map may be generated by overlaying the combined patient-specific model of heart motion and the three-dimensional reconstruction. This may be done for example by fusing vectors of the patient-specific motion model with pixels of the three-dimensional reconstruction.

According to an embodiment of the invention the motion information is one of a general knowledge about heart motion speed and motion information derived from the first cardiac data set.

The general knowledge may be acquired for example from a database including statistics of several patients or alternatively general facts for example about anatomical features. The derived motion information may for example be acquired with the help of an algorithm which processes the data of the first cardiac data set. The algorithm may provide a direct analysis of the acquired rotational X-ray projections. Therein, for example a motion function may be extracted by cross-correlating of line integral profiles from adjacent frames of the first cardiac data set.

According to a further embodiment of the invention, the method further comprises the step of generating a second three-dimensional reconstruction from images of the first cardiac data set using the cardiac phase indicating signal. Therein, the second three- dimensional reconstruction corresponds to a second cardiac phase.

The second three-dimensional reconstruction is reconstructed from the same cardiac data set as the first three-dimensional reconstruction.

The second cardiac phase may be a phase of relatively low motion and may correspond to a near-extremal state of heart motion. The cardiac phase used for the second reconstruction differs from the cardiac phase used for the first reconstruction. For example, the first cardiac phase may correspond to an end systolic phase. Then, the second cardiac phase may correspond to the late diastolic phase.

According to a further embodiment of the invention the method further comprises the step of elastically registering the first and the second three-dimensional reconstructions and thereby generating an estimated patient-specific model of heart motion.

In other words, the invention according to this embodiment may be seen as based on the idea to combine two three-dimensional reconstructions acquired in different phases of the heart motion to receive an estimated patient-specific model of heart motion.

The two three-dimensional reconstructions are elastically registered to one another. A possible registration method consists in segmenting the vessel center lines and then finding an elastic transformation function that aligns the center lines as good as possible. This elastic registration may yield a patient-specific model of the heart motion between the two, for example near extremal, states of heart motion. An adequate algorithm may be employed for elastically registering the first and second three-dimensional reconstructions.

The estimated patient-specific model of heart motion is based only on the information comprised in the three-dimensional reconstructions which both may be generated with the help of a first cardiac data set. The estimated patient-specific model of heart motion may be generated before the combined patient-specific model of heart motion. The combined patient-specific model of heart motion may comprise more information such as additional motion information and therefore be more exact than the estimated patient-specific model of heart motion. The combined patient-specific model of heart motion may be generated by adding motion information to the estimated patient-specific model of heart motion.

By combining the estimated patient-specific model of heart motion with motion information a patient-specific model in arbitrary cardiac phases is generated.

According to a further embodiment of the invention the combined patient- specific model of heart motion may be used to generate a three-dimensional representation of at least an intermediate cardiac phase.

The combined patient-specific model of heart motion comprises information on the heart in different arbitrary cardiac phases. The intermediate cardiac phase may be a non- extremal state of heart motion between the first and second phase of heart motion.

Furthermore the combined patient-specific model may comprise information on several cardiac phases.

According to a further embodiment of the invention, the first cardiac phase is one of the following group: end-systolic phase and late diastolic phase.

The second cardiac phase may be the respective other one of the end-systolic phase and the late diastolic phase. For example, in the case that the first cardiac phase is the end-systolic phase the second cardiac phase is the late diastolic phase.

Therein, the systolic phase corresponds to the contraction of the heart chambers and the diastolic phase refers to the widening or relaxing of the chambers of the heart between two contractions, i.e. diastole is the period of time when the heart fills with blood after systole.

According to a further embodiment of the invention the first cardiac data set comprises a rotational coronary angiography sequence.

The first cardiac data set may comprise further data or may consist only of rotational coronary angiography data.

According to a further embodiment of the invention the external signal is an electrocardiogram (ECG) signal. The electrocardiogram signal is employed to gate the first cardiac data set.

According to a second aspect of the invention a method for simultaneous dynamic visualization of coronary information and X-ray projections is provided. The method comprises the following steps: generating a four-dimensional road map for coronary artery interventions according to one of the above-described embodiments; acquiring a second cardiac data set with an external signal; deteirnining the cardiac phase for each frame of the second data set using the external signal; selecting a corresponding frame from the combined patient-specific model of heart motion for each cardiac phase; providing an overlay of the second cardiac data set and the corresponding frames from the combined patient-specific model.

In other words, the second aspect of the invention may be seen as based on the idea to provide a combined display of the moving three-dimensional coronary artery tree together with further information such as interventional devices inside the artery tree that is contained in the second cardiac data set. The four-dimensional roadmapping overlay received in this way may greatly help in navigation. Furthermore the amount of contrast agent required for an intervention might be reduced and intervention times may be shortened for example for the following reasons: The first cardiac data set may be acquired during a contrast agent is present in the vessels. From these data a four-dimensional roadmap is modelled which may be used for further applications such as examinations and interventions at the same patient without the need to inject contrast agent.

The overlay of the second cardiac data set and the corresponding frames from the combined patient-specific model are for example received by simultaneous display of the combined patient-specific model and currently acquired X-ray projections of a patient.

The second cardiac data set may be acquired at the same patient and at the same area of interest as the first cardiac data set but later in time. For each frame or image of the second cardiac data set the cardiac phase is determined for example with the help of an electrocardiogram. During the display of a certain frame of the second data set a

corresponding frame from the combined patient-specific model is displayed as an overlay. The correspondence of the frames is determined by selecting the same cardiac phases.

According to a further embodiment of the invention the method further comprises the following step: adapting the orientation of the presentation of the frame from the combined patient-specific model of heart motion and the frame from the second data set.

The adaptation may be done for example by rotating the three-dimensional moving patient-specific model of heart motion and at the same time deforming it such that the viewing angle reflects the current viewing direction of the X-ray projections for example acquired with the help of a C-arm.

According to a further embodiment of the invention the method further comprises the following step: providing a spatial registration between the frames of the second cardiac data set and the corresponding frames from the combined patient specific model.

According to a further embodiment of the invention the spatial registration is based on at least one of a landmark which is consistent between the first cardiac data set and the second cardiac data set, a geometric anatomical constraint and a use of anatomical landmarks.

A spatial registration may correspond to the matching of movement and translations between the two data sets or single frames. For example, when a patient moves for example between the acquisition of the first and second cardiac data set or if he/she breathes the roadmapping overlay may be shifted by the spatial registration such that it coincides with the actual position of the heart in the frame from the second data set.

The spatial registration may be carried out using 1) landmarks like the location of catheters that were not moved between the first and second acquisition, 2) geometric constraints, e.g., the interventional devices visible in the second acquisition must lie inside the vessels visible in the 3D reconstruction, or 3) comparing the X-ray projections of the first and second acquisition and using anatomic features like ribs, the spine, or the diaphragm as references.

According to a third aspect of the invention an apparatus for acquiring data and providing a four-dimensional roadmapping for coronary artery interventions is provided. Therein the apparatus is adapted to perform the above-described methods.

Such apparatus may include an X-ray source for imaging X-rays, an X-ray detector for acquiring for example X-ray data of the heart, a contrast agent injector for introducing a contrast agent into vessels of a patient, a controlling unit for controlling at least one of the X-ray source, the X-ray detector and the contrast agent injector, a computing unit for executing some of the steps of the methods described above and at least one displaying device for visualizing the generated data.

According to a fourth aspect of the invention, a computer program element is presented which is adapted to control the method described above when executed on a computer.

According to a fifth aspect of the invention, a computer-readable medium with a computer program element as described above is presented.

It has to be noted that embodiments of the invention are described with reference to different subject-matters. In particular, some embodiments are described with reference to method type claims whereas other embodiments are described with reference to apparatus type claims. However, a person skilled in the art will gather from the above and the following description that, unless other mentioned, in addition to any combination of features belonging to one type of subject-matter also any combination between features relating to the different subject-matters, in particular between features of the apparatus type claims and features of the method type claims, is considered to be disclosed with this application.

The aspects defined above and further aspects, features and advantages of the invention can be derived from the examples of embodiments described hereinafter. The invention will be described in more detail hereinafter with reference to examples of embodiments but to which the invention is not limited.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 shows a flow diagram schematically representing a method for four- dimensional roadmapping according to an exemplary embodiment of the invention

Fig. 2 shows a flow diagram schematically representing a method for four- dimensional roadmapping according to a further exemplary embodiment of the invention

Fig. 3 shows a flow diagram schematically representing a method for simultaneous dynamic visualization of coronary information and X-ray projections according an exemplary embodiment of the invention

Fig. 4A shows a first three-dimensional reconstruction of coronary arteries in a first cardiac phase of a relatively low motion usable in an exemplary embodiment of the invention

Fig. 4B shows a second three-dimensional reconstruction of coronary arteries in a second cardiac phase of a relatively low motion usable in an exemplary embodiment of the invention

Fig. 4C shows an electrocardiogram acquired with the first cardiac data set usable in an exemplary embodiment of the invention

Fig. 5 shows elastically registered three-dimensional reconstructions usable in an exemplary embodiment of the invention

Fig. 6 A shows a fluoroscopy sequence of the second cardiac data set in different cardiac phases

Fig. 6B shows three-dimensional reconstructions corresponding to the sequence shown in Fig. 6 A

Fig. 6C shows an electrocardiogram acquired with the second cardiac data set usable in an exemplary embodiment of the invention

Fig. 7 shows schematically an apparatus according to one exemplary embodiment of the invention.

All figures are only schematical and not to scale.

DETAILED DESCRIPTION OF EMBODIMENTS

In Fig. 1 a flow diagram schematically representing a method for four- dimensional roadmapping according to an exemplary embodiment of the invention is presented. In step SI a first cardiac data set 1 is acquired together with a cardiac phase indicating signal 3. In step S3 a first three-dimensional reconstruction 7 (also denoted as representation) from images of the first cardiac data set 1 corresponding to a first cardiac phase b of low motion is generated. In step S5 motion information 5 is combined with the first three-dimensional reconstruction and thereby a combined patient-specific model of heart motion 9 is generated. Therein, the motion information 5 may indicate at least one of a motion direction, a motion speed, a motion acceleration and extremal states of the heart. In step S7 a four-dimensional road map 11 is generated by superimposing the combined patient-specific model of heart motion 9 and the first three-dimensional reconstruction 7.

In Fig. 2 a schematical flow diagram according to a further embodiment of the method is presented. In Fig. 2 steps SI to S7 are similar to the steps shown in Fig. 1. The four-dimensional visualization and the subsequent roadmapping overlay are generated as follows:

After an acquisition of a first cardiac data set 1 and a corresponding cardiac phase indicating signal 3 like for example an ECG signal in step S 1 a first three-dimensional reconstruction 7 corresponding to a first cardiac phase b of low motion is generated in step S3. Subsequently or at the same time a second three-dimensional reconstruction 13 corresponding to a second cardiac phase e of low motion is generated in step S9. The two three-dimensional reconstructions 7, 13 generated in steps S3 and S9 may be two ECG gated three-dimensional reconstructions of the desired coronary artery branch e.g. left or right coronary artery. They may be generated from a rotational coronary angiography sequence. One of the three-dimensional reconstructions 7, 13 may be generated in the end-systolic phase b as for example shown in Fig. 4. The respective other three-dimensional reconstructions 7, 13 is generated in the late diastolic phase e as shown in Fig. 4.

The two reconstruction phases of the three-dimensional reconstructions 7, 13 may represent phases of low motion in the cardiac cycle, making high quality reconstruction feasible. Moreover, being the states of maximum contraction and near maximum delation they closely represent the two extremal heart motion states.

In step S 11 the first three-dimensional reconstruction 7 and the second three- dimensional reconstruction 13 are elastically registered. Therein, one of the two three- dimensional reconstructions 7, 13 is registered elastically to the other. A possible registration method consists in segmenting the vessel center lines and then finding an elastic

transformation function that aligns the center lines as good as possible. This elastic registration yields a patient-specific model of the heart motion between the two near extremal states, here also called estimated patient-specific model of heart motion 15.

The such derived estimated patient-specific model 15 of heart motion is combined with motion information 5 such as general knowledge about heart motion, e.g. knowledge about when the heart motion speed is maximum and when it is lower and information that enables extrapolation of the maximally dilated motion state at end systole. Motion information can also be derived from the angiographic sequence. The estimated patient-specific model of heart motion 15 generated in step SI 1 is thus enhanced by combining it with further motion information 5 in step S5. The enhanced estimated patient-specific model 15 results in the combined patient-specific model 9 in step S5.

In step S7 a combined patient-specific model 9 of heart motion and one of the three-dimensional reconstructions 7 and 13 or a combination of both reconstructions 7, 13 are used to generate a four-dimensional road map 11 of the coronary arteries.

A different order of the steps of the presented method is possible. For example, step S9 may be carried out in parallel, before or after step S3.

In Fig. 3 a flow diagram schematically representing a method for simultaneous dynamic visualization of coronary information and X-ray projections according to a further exemplary embodiment of the invention is presented. The method in Fig. 3 may comprise any combination of step SI to step SI 1 resulting in a four-dimensional road map 11, in an estimated patient specific model 15 of heart motion or alternatively in a combined patient specific model 9 of heart motion.

Furthermore the method comprises a step S13 in which a second cardiac data set 17 and a cardiac phase indicating signal 19 are acquired. The second cardiac data set 17 may correspond to a fluoroscopy sequence which is acquired with an ECG signal of a patient. In step S15 the cardiac phase for each frame of the second cardiac data set 17 is determined using the cardiac phase indicating signal 19. Step S15 yields a second cardiac data set 17a with determined cardiac phases for each frame.

In step S17 a corresponding frame to each frame of the second cardiac data set

17a is selected from the combined patient-specific model 9 of heart motion for each cardiac phase. Alternatively the corresponding frames may be selected form the estimated patient specific model 15 of heart motion.

Step S17 provides frames 11a from the combined patient-specific model 9 corresponding to frames of the second cardiac data set 17a.

In step S19 an overlay 21 of the second cardiac data set 17, 17a and the corresponding frames 11a from the combined patient-specific model 9 is provided.

In other words, a possible procedure is as follows: An ECG signal of the patient is measured together with a fluoroscopy sequence. For each frame of the sequence the heart phase is determined from the ECG and the three-dimensional reconstruction is deformed according to the motion model to represent a corresponding motion state. The deformed reconstruction is displayed as an overlay over the frame, where the reconstruction is rotated such that the viewing angle reflects the current viewing direction of the C-arm. The combined display of fluoroscopy and moving reconstruction shows both the coronary artery tree and the information contained in the second cardiac data set, such as the current position of an interventional device. This visualization procedure can be performed either in real-time or after the acquisition of the fluoroscopy sequence.

The described method facilitates the generation of a four-dimensional road mapping overlay, with feasible requirements with respect to computation time. The resulting combined display of the moving three-dimensional coronary artery tree together with interventional devices can greatly help in navigation. The amount of contrast agent required for an intervention might be reduced and intervention times may be shortened.

In Fig. 4 the first and second three-dimensional reconstructions 7 and 13 are presented together with a cardiac phase indicating signal 3 acquired as an electrocardiogram signal 23. Fig. 4C shows an electrocardiogram signal 23 acquired with the first cardiac data set 1. The distinctive peaks of the ECG signal are called R waves and are denoted with R. The reference RR denotes the time interval between two consecutive R waves in an ECG. Some of the different stages of heart motion i.e. the cardiac phases are indicated below the

electrocardiogram signal 23 and denoted with the letters a, b, c, d, e and f. Therein a for example may correspond to 12.5% of the RR cycle, b may represent 25% of the RR cycle, i.e. one fourth of a heartbeat, b may correspond to the end-systolic phase, c may be 37.5% of the RR cycle, d may be 50% of the RR cycle and f may represent 87.5% of the RR cycle.

Furthermore, e may correspond to 75% of the RR cycle, which corresponds to the late diastolic phase in this example. The locations of the low motion phases in the RR cycle may vary from patient to patient and may vary with the heart rate.

In Fig. 4A the first three-dimensional reconstruction 7 is presented in phase b, i.e. in the phase of end systole. In Fig. 4B the second three-dimensional reconstruction 13 is presented in the phase e, i.e. in the late diastolic phase. In the three-dimensional

reconstructions 7, 13 the vessels of the heart 25 are represented.

The three-dimensional reconstructions shown in Fig. 4A and 4B are generated from a rotational coronary angiography sequence gated with an ECG signal. In Fig. 5 elastically registered three-dimensional reconstructions 7, 13 are represented. Therein, one reconstruction is elastically registered onto the other, yielding an estimated patient-specific model 15 of heart motion. In the representation in Fig. 5 vessel center lines were used as landmarks for the registration. Therein, a reconstruction of the late diastolic phase may be used as reference. The reconstruction representing the end-systolic phase is registered to the reconstruction representing the late diastolic phase.

The estimated patient-specific model 15 of heart motion is extrapolated to arbitrary heart phases using general knowledge about heart motion and possibly motion information derived from the first cardiac data set 1. The general knowledge and motion information fused with the estimated patient-specific model 15 of heart motion is denoted with motion information 5 in Figs. 1 and 2.

In Fig. 6 a fluoroscopy sequence of the second cardiac data set 17 is presented in correspondence with frames 11a from the combined patient-specific model 9.

In Fig. 6 A a fluoroscopy sequence of the second cardiac data set 17 is presented in different cardiac phases a, b, c, d, e and f. The cardiac phases corresponding to the frames of the second data set can be arbitrary, determined with the help of an

electrocardiogram signal 23 shown in Fig. 6C and acquired together with the second cardiac data set 17. The correlation of the second cardiac data set 17 with the electrocardiogram signal 23 provides a sequence as shown in Fig. 3 in step S15 of frames corresponding to the second cardiac data set 17a with determined cardiac phases for each frame.

In Fig. 6B a sequence of frames 11a from the combined patient-specific model 9 is shown. The frames 11a correspond to the frames of the second cardiac data set 17a in their cardiac phases and in their geometrical orientation.

The second cardiac data set 17 e.g. a fluoroscopy sequence as in Fig. 6A may be displayed together as an overlay with the sequence shown in Fig. 6B. The sequence in Fig. 6 A may be acquired subsequently to the first cardiac data set 1 and the generation of the combined patient-specific model 9. In the sequence represented in Fig. 6A for example no contrast agent is present in the vessels of the heart. For this reason only intervention instruments like catheter 26 may be visible in the representation. In the frames 11a shown in Fig. 6B which result from the first cardiac data set 1 vessels of the heart 25 are visible in good quality because they are modelled based on three-dimensional reconstructions acquired during a phase of low motion of the heart and during a presence of contrast agent in the vessels.

For each frame of the sequence in Fig. 6A the respective heart phase is determined from the ECG signal shown in Fig. 6C. The reconstruction is then displayed in the correct viewing angle and deformed to display the respective motion state with the deformation determined from the combined patient-specific motion model 9 as presented in Fig. 6B. The reconstructions of Fig. 6B may be displayed as an overlay over the fluoroscopy sequence presented in Fig. 6 A possibly with colour coding.

In Fig. 7 an apparatus for acquiring data and providing a four-dimensional road map for coronary artery interventions is presented schematically. The apparatus 32 includes an X-ray source 33 for emitting X-rays, an X-ray detector 35 for acquiring for example data about an area of interest of a patient, a controlling unit 39 for controlling the X-ray source 33 and the X-ray detector 35 and a computing unit 41 for executing the steps of the described methods. The X-ray detector 35 and the X-ray source 33 may be arranged at a C-arm 31. The apparatus 32 may further comprise several displays. For example, a first display 27 and a second display 29 may be provided. Furthermore the displays 27, 29 may be divided into several further displays.

It should be noted that the terms "comprising", "including" etc. do not exclude other elements or steps and the "a" or "an" does not exclude a plurality. Also elements described in association with different embodiments may be combined. It should also be noted that reference signs in the claims should not be construed as limiting the scope of the claims.

LIST OF REFERENCE SIGNS:

I first cardiac data set

3 cardiac phase indicating signal for first cardiac data set

5 motion information

7 first three-dimensional reconstruction corresponding to a first cardiac phase of relatively low motion

9 combined patient specific model of heart motion

I I four-dimensional roadmap

11a frame from combined patient specific model corresponding to frame of second cardiac data 17a

13 second three-dimensional reconstruction corresponding to a second cardiac phase of relatively low motion

15 estimated patient specific model of heart motion

17 second cardiac data set

17a second cardiac data set with detemiined cardiac phases for each frame

19 cardiac phase indicating signal for second cardiac data set

21 overlay of the second cardiac data set and the corresponding frames from the combined patient specific model

23 electrocardiogram (ECG) signal

25 vessels of the heart

26 catheter

27 first display

29 second display

31 C-arm

32 apparatus

33 X-ray source

35 X-ray detector

37 contrast agent injector

39 controlling unit

41 computing unit R peak of ECG

RR time interval between two consecutive R waves in an ECG

a 12,5 % of RR cycle

b 25 % of RR cycle

c 37,5 % of RR cycle

d 50 % of RR cycle

e 75 % of RR cycle

f 87,5 % of RR cycle SI acquiring first cardiac data set together with a cardiac phase indicating signal

S3 generating a first three-dimensional reconstruction from images of the first cardiac data set corresponding to a first cardiac phase of relatively low heart motion

S5 generating a combined patient specific model of heart motion based on motion information and the first three-dimensional reconstruction

S7 generating a four-dimensional roadmap based on patient specific model of heart motion and the first three-dimensional reconstruction

S9 generating a second three-dimensional reconstruction from images of the first cardiac data set corresponding to a second cardiac phase of relatively low heart motion

S 11 elastically registering the first and second three-dimensional reconstructions and generating an estimated patient specific model of heart motion S13 acquiring second cardiac data set together with a cardiac phase indicating

signal

S15 determining the cardiac phase for each frame of the second cardiac data set

S17 selecting a corresponding frame from the combined patient specific model of heart motion for each cardiac phase

S19 providing an overlay of the second cardiac data set and the corresponding

frames from the combined patient specific model

Claims

CLAIMS:

1. A method for four-dimensional roadmapping usable in minimally invasive cardiac interventions, the method comprising the following steps:

acquiring a first cardiac data set (1) together with a cardiac phase indicating signal (3) (SI);

generating a first three-dimensional reconstruction (7) from images of the first cardiac data set (1) using the cardiac phase indicating signal (3) (S3), wherein the first three- dimensional reconstruction (7) corresponds to a first cardiac phase (b);

generating a combined patient specific model (9) of heart motion based on the first three-dimensional reconstruction (7) and a motion information (5) (S5), wherein the motion information (5) indicates at least one of a motion direction, a motion speed, a motion acceleration and extremal states of the heart; and

generating a four-dimensional roadmap (11) based on the combined patient specific model (9) of heart motion and the first three-dimensional reconstruction (7) (S7). 2. The method according to claim 1,

wherein the motion information (5) is one of a general knowledge about heart motion speed and motion information derived from the first cardiac data set (1).

3. The method according to one of claims 1 and 2, further comprising:

generating a second three-dimensional reconstruction (13) from images of the first cardiac data set (1) using the cardiac phase indicating signal (3) (S9);

wherein the second three-dimensional reconstruction 13) corresponds to a second cardiac phase. 4. The method according to claim 3, further comprising:

elastically registering the first and the second three-dimensional reconstructions^, 13) (SI 1), and

thereby generating an estimated patient specific model (15) of heart motion.

5. The method according to one of claims 1 to 4,

wherein the combined patient specific model (9) of heart motion represents a three-dimensional reconstruction of at least an intermediate cardiac phase.

6. The method according to one of claims 1 to 5,

wherein the first cardiac phase is one of the following group: end-systolic phase and late diastole phase. 7. The method according to one of claims 1 to 6,

wherein the first cardiac data set (1) comprises a rotational coronary angiography sequence.

The method according to one of claims 1 to 7,

wherein the cardiac phase indicating signal (3) is an electrocardiogram signal; wherein the electrocardiogram signal is employed to gate the first cardiac data

9. A method for simultaneous dynamic visualization of coronary information and

X-Ray projections, the method comprising the following steps:

generating a four-dimensional roadmap for coronary artery interventions according to one of claims 1 to 8;

acquiring a second cardiac data set (17) with a cardiac phase indicating signal (19) (S13);

deteimining the cardiac phase for each frame of the second cardiac data set (17) using the cardiac phase indicating signal (19) (SI 5);

selecting a corresponding frame from the combined patient specific model (9) of heart motion for each cardiac phase (SI 7);

providing an overlay of the second cardiac data set (17) and the corresponding frames (11a) from the combined patient specific model (9) (S19).

10. The method according to claim 9, further comprising: adapting the orientation of the presentation of the frame from the combined patient specific model (9) of heart motion and the frame (11a) from the second data set (17).

11. The method according to one of claims 9 and 10, further comprising:

providing a spatial registration between the frames of the second cardiac data set (17) and the corresponding frames (11a) from the combined patient specific model (9).

12. The method according to claim 11,

wherein the spatial registration is based on at least one of a landmark which is consistent between the first cardiac data set (1) and the second cardiac data set (17), a geometric anatomical constraint and a use of anatomical landmarks.

13. Apparatus (32) for acquiring data and providing a four-dimensional roadmapping for coronary artery interventions, the apparatus being adapted to perform method according to one of claims 1 to 12.

14. Computer program element adapted to control the method according to one of claims 1 to 12 when executed on a computer. 15. Computer readable medium with a computer program element according to claim 14.