**3.6 Advanced visualization**

Recent post-processing algorithms analyse an entire digital subtraction angiography (DSA) sequence at once and represent the sequence in one single colour-coded image. In order to obtain a colour-coded image, the algorithm takes the time to maximum opacification of each individual pixel, starting with the injection and subsequently visualising the distribution of the contrast medium through the vessels. These time measurements are then represented by a colour, allowing visualisation of the complete vessel tree in one image. Thus, the colours represent the contrast agent from its initial entry into the blood vessels to its flow throughout the anatomy of interest in one image.

Such dynamic flow evaluations provide a greater understanding of the contrast flow within the pathology, greater ease in visualizing the success of a procedure, and they assist the

The Hybrid Operating Room 85

The information flow between the angiographic system and the workstation ensures that an anatomical structure in the fluoroscopic image can be related to an anatomical structure in the 3D image and vice versa: that means the images are registered with each other. Even preoperatively acquired images can be related to the patient by image-to-image registration of the pre-operative image with the intra-operative acquired DynaCT. Information in the preoperative image (e.g. a surgical plan) can be directly overlaid on top of a live fluoroscopy. Inherent registration of 3D images of the angiographic systems to the patient triggers new applications which go beyond just simple imaging, but towards image-driven guidance

Any 3D information extracted from the image in the workstation can be overlaid on the live fluoroscopic image. Firstly, the 3D image itself can be overlaid colour-coded on top of the fluoroscopic image. For example, in Fig. 8, a 3D angiography is colour-coded in orange and overlaid on the live fluoroscopy. Any change of the angulations of the C-arm will cause the workstation to re-calculate in real-time the view on the 3D image to match exactly the view of the live 2D fluoroscopy image. Without additional contrast agent injection the surgeon can observe device movements simultaneously with the 3D overlay of the vessel contours in

Fig. 7 illustrates an alternative way to add information from the workstation to the fluoroscopic image. After either manual or automatic segmentation of the anatomical structures of interest in the 3D image, the outline can be overlaid as a contour onto the fluoroscopic image. In this example, an AAA aneurysm has been segmented in the DynaCT image. The contour of the 3D segmentation is shown in the fluoroscopic view and provides additional information which is not visible in the fluoroscopic image. Overlaid landmarks do not necessarily need to be extracted from images directly, but might be added by the

Fig. 7. Linkage between a modern angio system and its workstation. The corresponding information flow guarantees the registration between the 3D DynaCT image in the workstation and a 2D fluoroscopic live image and is the prerequisite of true 2D/3D image

based on 3D information as illustrated in the next sections.

**3.7.2 Overlay of 3D information on top of 2D fluoroscopy** 

the fluoroscopy image.

fusion

Fig. 6. Advanced visualization of an entire DSA sequence (iFlow): Colour-coded pre and post-procedural results visualize the improvement of flow

clinicians in image review by showing a complete Digital Subtraction Angiography (DSA) run in a single image (Ahmed et al., 2009). For example, this technology can be used to enhance pre-procedural and post-procedural imaging of patients under treatment for stenoses of peripheral vessels (see Fig. 6). Flow deviations and the increased utilisation of collaterals can more easily be detected prior to intervention, since anomalies more readily attract the physician's attention due to their specific colours. Following the intervention, the success of a balloon dilatation or stent implantation of a stenosis is readily visible due to the improved flow.

### **3.7 Fusion imaging and 2D/3D overlay**

Modern angiographic systems are not just used for imaging, but support the surgeon also during the procedure by guiding the intervention based on 3D information acquired either pre-operatively or intra-operatively. Such guidance requires that the 3D information is registered to the patient. The next sections illustrate why 3D images acquired by an angiographic system are inherently registered with the patient and show new applications based on this fundamental feature of modern angiographic systems.

### **3.7.1 Information flow between workstation and angiographic system**

3D DynaCT images are calculated from a set of projections acquired from different angles around the patient. The volume is reconstructed on a separate workstation. Even though the workstation and the angio system can be considered as separate systems, there is a close link and a continuous information flow between these systems. For example, when the user virtually rotates the volume on the workstation to view the anatomy from a certain perspective, the parameter of this view can be transmitted to the angio system, which then drives the C-arm to the exact same perspective for fluoroscopy. In the same way, if the C-arm angulation is changed, this angulation can be transmitted to the workstation which updates the volume to the same perspective as the fluoroscopic view (see Fig. 7).

Fig. 6. Advanced visualization of an entire DSA sequence (iFlow): Colour-coded pre and

clinicians in image review by showing a complete Digital Subtraction Angiography (DSA) run in a single image (Ahmed et al., 2009). For example, this technology can be used to enhance pre-procedural and post-procedural imaging of patients under treatment for stenoses of peripheral vessels (see Fig. 6). Flow deviations and the increased utilisation of collaterals can more easily be detected prior to intervention, since anomalies more readily attract the physician's attention due to their specific colours. Following the intervention, the success of a balloon dilatation or stent implantation of a stenosis is readily visible due to the

Modern angiographic systems are not just used for imaging, but support the surgeon also during the procedure by guiding the intervention based on 3D information acquired either pre-operatively or intra-operatively. Such guidance requires that the 3D information is registered to the patient. The next sections illustrate why 3D images acquired by an angiographic system are inherently registered with the patient and show new applications

3D DynaCT images are calculated from a set of projections acquired from different angles around the patient. The volume is reconstructed on a separate workstation. Even though the workstation and the angio system can be considered as separate systems, there is a close link and a continuous information flow between these systems. For example, when the user virtually rotates the volume on the workstation to view the anatomy from a certain perspective, the parameter of this view can be transmitted to the angio system, which then drives the C-arm to the exact same perspective for fluoroscopy. In the same way, if the C-arm angulation is changed, this angulation can be transmitted to the workstation which updates the volume to the same perspective as the fluoroscopic view

post-procedural results visualize the improvement of flow

based on this fundamental feature of modern angiographic systems.

**3.7.1 Information flow between workstation and angiographic system** 

improved flow.

(see Fig. 7).

**3.7 Fusion imaging and 2D/3D overlay** 

The information flow between the angiographic system and the workstation ensures that an anatomical structure in the fluoroscopic image can be related to an anatomical structure in the 3D image and vice versa: that means the images are registered with each other. Even preoperatively acquired images can be related to the patient by image-to-image registration of the pre-operative image with the intra-operative acquired DynaCT. Information in the preoperative image (e.g. a surgical plan) can be directly overlaid on top of a live fluoroscopy.

Inherent registration of 3D images of the angiographic systems to the patient triggers new applications which go beyond just simple imaging, but towards image-driven guidance based on 3D information as illustrated in the next sections.

### **3.7.2 Overlay of 3D information on top of 2D fluoroscopy**

Any 3D information extracted from the image in the workstation can be overlaid on the live fluoroscopic image. Firstly, the 3D image itself can be overlaid colour-coded on top of the fluoroscopic image. For example, in Fig. 8, a 3D angiography is colour-coded in orange and overlaid on the live fluoroscopy. Any change of the angulations of the C-arm will cause the workstation to re-calculate in real-time the view on the 3D image to match exactly the view of the live 2D fluoroscopy image. Without additional contrast agent injection the surgeon can observe device movements simultaneously with the 3D overlay of the vessel contours in the fluoroscopy image.

Fig. 7 illustrates an alternative way to add information from the workstation to the fluoroscopic image. After either manual or automatic segmentation of the anatomical structures of interest in the 3D image, the outline can be overlaid as a contour onto the fluoroscopic image. In this example, an AAA aneurysm has been segmented in the DynaCT image. The contour of the 3D segmentation is shown in the fluoroscopic view and provides additional information which is not visible in the fluoroscopic image. Overlaid landmarks do not necessarily need to be extracted from images directly, but might be added by the

Fig. 7. Linkage between a modern angio system and its workstation. The corresponding information flow guarantees the registration between the 3D DynaCT image in the workstation and a 2D fluoroscopic live image and is the prerequisite of true 2D/3D image fusion

The Hybrid Operating Room 87

Fig. 9. Image-driven guidance during Trans-Aortic Valve Implantation (TAVI). Contours were automatically segmented from a 3D DynaCT image and the C-arm was positioned perpendicular to the aortic root for live fluroscopy based on anatomical landmarks extracted from the DynaCT image without user interaction. (Siemens AG, Forchheim, Germany)

Careful planning and professional expertise is a key factor for every hybrid room project. Before planning a hybrid operating room a clear vision for the utilization should be

Today's operating rooms require concepts that address the requirements and needs of different surgical specialties and procedures. Workflow efficiency is a key success factor for the hospital and the surgical program. Minimal turnover times and optimal processes throughout the entire surgical workflow and the actual surgical procedure are required (Tomaszewski, 2008). Therefore, a hybrid operating room should ideally be integrated into an existing OR suite. All aspects and steps starting with patient transfer from the ward to anesthesia and operating room preparation are important. Addiontional aspects for planning are material supply processes, i.e. of materials necessary for the procedure, and

Due to high cost, OR facilities are commonly shared by different disciplines. A very flexible room layout and design allow for the necessary repositioning of devices and changes of the

**4. Planning the hybrid room** 

postoperative intensive care surveillance and treatment.

established (Benjamin, 2008).

Fig. 8. Overlay of 3D DynaCT image (orange) on top of a fluoroscopic image during cardiac resynchronization therapy. Courtesy of K.-J. Gutleben, M.D., G. Nölker, M.D. A. Sinha, M.D., J. Brachmann, M.D., Department of Cardiology, Klinikum Coburg, Germany.

surgeon. For example to place fenestrated stents, a pre-operatively acquired CT image could be used to mark the ostia of the visceral arteries manually by the surgeon. By aligning the pre-operative CT image with the intra-operatively acquired DynaCT image, the ostia can be displayed (beside the contour of the arteries) in the fluoroscopic image. Notably, this truly is 3D information, i.e. any change in the C-arm position or angulations will update the view on the marks to perfectly match the live fluoroscopy image.
