Sebastian Eulenstein

Augmenting Intraoperative 3D Ultrasound with Preoperative Models for Navigation in Liver Surgery

Organ deformation between preoperative image data and the patient in the OR is the main obstacle for using surgical navigation systems in liver surgery. Our approach is to provide accurate navigation via intraoperative 3D ultrasound. These ultrasound data are augmented with preoperative anatomical models and planning data as an important additional orientation aid for the surgeon. We present an overview of the whole ultrasound navigation system as well as an approach for fast intraoperative non-rigid registration of the preoperative models to the ultrasound volume. The registration method is based on the vessel center lines and consists of a combination of the Iterative Closest Point algorithm and multilevel B-Splines. Quantitative results for three different patients are presented.

Feasibility of navigated resection of liver tumors using multiplanar visualization of intraoperative 3-dimensional ultrasound data.

Background:Intraoperative ultrasound is widely used in liver surgery, but primarily for diagnostic purposes. We have developed and evaluated a system for navigated liver resections using on intraoperatively acquired 3-dimensional (3D) ultrasound data. Methods:Navigation technique based on 3D ultrasound and an optical tracking system. Accuracy of the system was validated experimentally in a tumor model. Subsequently, clinical application was evaluated in 54 patients for resection of central liver tumors. Clinical feasibility and accuracy of the navigation technique were assessed with respect to practicability, adequacy of visualization, and precision of navigated resection (free margin). Results:Evaluation of the system in the tumor model showed a significant increase of the accuracy of navigated resections compared with conventional resection (P < 0.05). Clinical application of 3D ultrasound-based navigation was feasible in 52 of 54 patients. Sufficient visualization was obtained with 2 orthogonal section planes. This navigation strategy provided complete anatomic orientation and accurate position control of surgical instruments. Mean histologic resection margin was 9 mm with a maximum deviation of 8 mm from the planned virtual resection margins. Conclusions:Optoelectronic navigation with section mode visualization in 2 orthogonal planes does sufficiently display intraoperative 3D data and enables accurate ultrasound-based navigation of liver resections.

Image-guided surgery of liver metastases by three-dimensional ultrasound-based optoelectronic navigation

Vessel‐oriented surgery and tumour‐free resection margins are essential for resection of liver metastases to preserve liver parenchyma and improve oncological outcome. Preoperative three‐dimensional models reconstructed from imaging data could facilitate surgical planning with the use of navigation technology.

Computer-assisted pelvic tumor resection: fields of application, limits, and perspectives.

The treatment of malignant tumors involving the pelvic area is a challenging problem in musculoskeletal oncology due to the complex pelvic anatomy and the often large tumor size at presentation. The use of navigation systems has effectively increased surgical precision aiming at optimal preservation of pelvic structures without compromising oncologic outcome by means of improved visibility of the surgical field, and enabling intraoperative display and 3D reproduction of preoperatively determined pelvic osteotomy and resection levels. In the following sections, current developments in computer-assisted pelvic surgery are reviewed and possible fields of application, as well as limitations of navigation systems, are discussed.

Development of navigation systems for image-guided laparoscopic tumor resections in liver surgery

Minimally invasive surgery has become a viable alternative to conventional surgery. The technical advantages of minimally invasive surgery can be translated into clinical benefits for the patients, i.e., less postoperative pain and impairment of lung function, better cosmetic results, shorter hospitalization, and earlier convalescence. Laparoscopic operations have replaced a significant proportion of open surgical procedures and are now routinely used. While the role of laparoscopic surgery has been generally accepted for the management of benign disorders, there is an ongoing debate regarding the adequacy of this technique in surgical oncology. There is evidence that minimally invasive surgery can reduce perioperative morbidity in cancer patients. However, definite validation of these procedures for tumor surgery is not yet available due to the lack of prospective randomized trials providing reliable long-term data on disease-free survival and overall survival. It seems likely that minimally invasive procedures will play an important role for the treatment of preneoplastic lesions and tumors of limited size. There are some technical limitations to laparoscopic surgery. The degrees of freedom for the instruments are limited because of the minimal access via the trocars. The absence of direct organ palpation and the lack of the third dimension are still limits of laparoscopy. The surgeon’s orientation and the location of anatomical and pathological structures is therefore more difficult than in open surgery. Modern image-guided surgery (IGS) systems have the potential to compensate these limitations. Preoperative computer-based intervention planning and intraoperative navigation systems are today routinely used for certain indications in neurosurgery, ENT and orthopedic surgery. Nevertheless major research efforts are still necessary to examine the clinical impact of image-guided surgery on current indications, to adapt the systems to other indications and to solve some open problems like intraoperative deformations. In liver surgery, computer-assisted 3D modeling and planning based on preoperative CT or MRI data allows much more accurate planning of surgery and may help to select the optimal procedure [1]. Surgeons can obtain a better steric vision of the individual anatomy, the location of the tumor in relation to the vascular structures and the individual vascular territories defining the resection planes. The use of such planning systems in oncological surgery and liver transplantation is in routine use at some clinics and the number of users is increasing. But at the end of this preparatory process, surgeons have to carry out the surgical plan on the patient in the OR. Intraoperative navigation systems support the surgeon to transfer preoperative plans precisely and safely. In neurosurgery, ENT surgery, orthopedic and trauma surgery navigation systems are well established for certain indications and several commercial systems are available. The main problem in liver surgery or in general in visceral surgery is the deforma2 Development of Navigation Systems for Image-Guided Laparoscopic Tumor Resections in Liver Surgery

A new measure to assess the difficulty of liver resection.

BACKGROUND There is no valid measure to assess surgical difficulty and feasibility of a planned liver resection. It is the objective of this study to evaluate a mathematical measure from a 3D graphical analysis. METHODS Eleven different 3D models of hepatic tumours were evaluated by experts for resectability and analysed with Amira graphic software taking into consideration the portal and hepatic venous vascular relationships. Virtual resection volumes with increasing resection margins from 1 to 30 mm were determined separately for portal veins, hepatic veins, their intersections and volume unions. The integral of the increasing resection volumes was defined as risk coefficient. The risk coefficients from this volumetric analysis were compared with the expert opinion. RESULTS The risk coefficient based on the integral of portal venous and hepatic venous volume unions reproduced the expert opinion highly significantly (correlation coefficient 0.9, p<0.05) and more accurately than volumetric analysis of the planned resection margin. CONCLUSION With automated volumetric analysis, anatomically problematic situations in liver surgery can be reproduced and scaled. The risk coefficient obtained is a suitable objective measure for defining risk areas in liver surgery.

Validation Metrics for Non-rigid Registration of Medical Images Containing Vessel Trees

Validation of non-rigid registration methods is still a challenging task. Different evaluation criteria were published, yet no widely accepted gold standard exists. The aim of this paper is to provide quantitative evaluation metrics suited for clinical 3D images containing vessel trees, such as liver or brain data. We present a method to identify corresponding points on different vessel trees by extracting consistent graph minors interactively. Four different metrics based on these correspondencies are proposed.

Automatic Calibration of 3D Ultrasound Probes

Navigation systems based on intra-operative ultrasound have been introduced for different surgical procedures and interventions. The accuracy of the ultrasound probe calibration is a main contribution to the overall accuracy of the navigation system. Different calibration methods based on phantoms have been introduced. The challenge of the calibration procedure is to identify the phantom structures accurately and possibly automatically in ultrasound images. In the majority of cases 2D ultrasound probes have been calibrated. The advantage of 3D probes is the acquisition of 3D ultrasound volumes out of the box enabling fast calibration approaches based on only one image acquisition. We introduce a new inexpensive and easily to manufacture planes phantom and automatic calibration algorithm for 3D ultrasound probes.

A New Class of Distance Measures for Registration of Tubular Models to Image Data

In some registration applications additional user knowledge is available, which can improve and accelerate the registration process, especially for non-rigid registration. This is particularly important in the transfer of pre-operative plans to the operating room, e.g. for navigation. In case of tubular structures, such as vessels, a geometric representation can be extracted via segmentation and skeletonization. We present a new class of distance measures based on global filter kernels to compare such models efficiently with image data. The approach is validated in a non-rigid registration application with Powerdoppler ultrasound data.

Computerunterstützte Prothesenkonstruktion mittels statistischen Formmodells bei Beckenresektionen

Die chirurgische Entfernung von Tumoren aus dem Beckenknochen erfordert oft die Implantation einer fur die individuelle Beckengeometrie angepassten Prothese, um den fehlenden Knochenanteil zu ersetzen. Die computergestutzte Planung einer solchen Prothese ermoglicht zum einen eine Beschleunigung der Prothesenkonstruktion und ist zum anderen eine wichtige Voraussetzung fur die prazise operative Umsetzung mittels eines Navigationssystems. Es wird ein Verfahren prasentiert, mit dem anatomische Landmarken und Kontaktflachen von Prothesen von einem statistischen Formmodell auf die individuelle Beckengeometrie ubertragen werden konnen. An die gewunschte Kontaktflache wird dann automatisch die Form der Prothese adaptiert.