Abstract:

In laparoscopic liver surgery (LLS) image-guidednavigation systems could support the surgeon by providingsubsurface information such as the positions of tumors andvessels. For this purpose, one option is to perform a registra-tion of preoperative 3D data and 3D surface patches recon-structed from laparoscopic images. Part of an automatic 3Dregistration pipeline is the feature description, which takes intoaccount various geometric and spatial information. Since thereis no leading feature descriptor in the field of LLS, two featuredescriptors are compared in this paper: The Fast Point FeatureHistogram (FPFH) and Triple Orthogonal Local Depth Images(TOLDI). To evaluate their performance, three perturbationswere induced: varying surface patch sizes, spatial displace-ment, and Gaussian deformation. Registration was performedusing the RANSAC algorithm. FPFH outperformed TOLDIfor small surface patches and in case of Gaussian deformationsin terms of registration accuracy. In contrast, TOLDI showedlower registration errors for patches with spatial displacement.While developing a 3D-3D registration pipeline, the choice ofthe feature descriptor is of importance, consequently a carefulchoice suitable for the application in LLS is necessary.

Abstract:

Monocular depth estimation is an important topic in minimally invasive surgery, providing valuable information for downstream application, like navigation systems. Deep learning for this task requires high amount of training data for an accurate and robust model. Especially in the medical field acquiring ground truth depth information is rarely possible due to patient security and technical limitations. This problem is being tackled by many approaches including the use of syn- thetic data. This leads to the question, how well does the syn- thetic data allow the prediction of depth information on clini- cal data. To evaluate this, the synthetic data is used to train and optimize a U-Net, including hyperparameter tuning and aug- mentation. The trained model is then used to predict the depth on clinical image and analyzed in quality, consistency over the same scene, time and color. The results demonstrate that syn- thetic data sets can be used for training, with an accuracy of over 77% and a RMSE below 10 mm on the synthetic data set, do well on resembling clinical data, but also have limitations due to the complexity of clinical environments. Synthetic data sets are a promising approach allowing monocular depth esti- mation in fields with otherwise lacking data.

Abstract:

In laparoscopic liver surgery, image-guided navigation systems provide crucial support to surgeons by supply- ing information about tumor and vessel positions. For this purpose, these information from a preoperative CT or MRI scan is overlaid onto the laparoscopic video. One option is performing a registration of preoperative 3D data and 3D reconstructed laparoscopic data. A robust registration is challenging due to factors like limited field of view, liver deformations, and 3D reconstruction errors. Since in reality various influencing factors always intertwine, it is crucial to analyze their combined effects. This paper assesses registration accuracy under various synthetically simulated influences: patch size, spatial dis- placement, Gaussian deformations, holes, and downsampling. The objective is to provide insights into the required quality of the intraoperative 3D surface patches. LiverMatch serves as the feature descriptor, and registration employs the RANSAC algorithm. The results of this paper show that ensuring a large field of view of at least 15-20% of the liver surface is necessary, allowing tolerance for less accurate depth estimation.

Abstract:

The lack of labeled, intraoperative patient data in medical scenarios poses a relevant challenge for machine learning applications. Given the apparent power of machine learning, this study examines how synthetically-generated data can help to reduce the amount of clinical data needed for robust liver surface segmentation in laparoscopic images. Here, we report the results of three experiments, using 525 annotated clinical images from 5 patients alongside 20,000 synthetic photo-realistic images from 10 patient models. The effectiveness of the use of synthetic data is compared to the use of data augmentation, a traditional performance-enhancing technique. For training, a supervised approach employing the U-Net architecture was chosen. The results of these experiments show a progressive increase in accuracy. Our base experiment on clinical data yielded an F1 score of 0.72. Applying data augmentation to this model increased the F1 score to 0.76. Our model pre-trained on synthetic data and fine-tuned with augmented data achieved an F1 score of 0.80, a 4% increase. Additionally, a model evaluation involving k-fold cross validation highlighted the dependency of the result on the test set. These results demonstrate that leveraging synthetic data has the ability of limiting the need for more patient data to increase the segmentation performance.

Abstract:

The application of machine learning approachesin medical technology is gaining more and more attention.Due to the high restrictions for collecting intraoperative patientdata, synthetic data is increasingly used to support the trainingof artificial neural networks. We present a pipeline to createa statistical shape model (SSM) using 28 segmented clinicalliver CT scans. Our pipeline consists of four steps: data pre-processing, rigid alignment, template morphing, and statisti-cal modeling. We compared two different template morphingapproaches: Laplace-Beltrami-regularized projection (LBRP)and nonrigid iterative closest points translational (N-ICP-T)and evaluated both morphing approaches and their corre-sponding shape model performance using six metrics. LBRPachieved a smaller mean vertex-to-nearest-neighbor distances(2.486±0.897 mm) than N-ICP-T (5.559±2.413 mm). Gen-eralizationand specificity errors for LBRP were consistentlylower than those of N-ICP-T. The first principal componentsof the SSM showed realistic anatomical variations. The perfor-mance of the SSM was comparable to a state-of-the-art model.

Abstract:

Acute ischemic stroke is a major health problem with a high mortality rate and a high risk for permanent disabilities. Selective brain hypothermia has the neuroprotective potential to possibly lower cerebral harm. A recently developed catheter system enables to combine endovascular blood cooling and thrombectomy using the same endovascular access. By using the penumbral perfusion via leptomeningeal collaterals, the catheter aims at enabling a cold reperfusion, which mitigates the risk of a reperfusion injury. However, cerebral circulation is highly patient-specific and can vary greatly. Since direct measurement of remaining perfusion and temperature decrease induced by the catheter is not possible without additional harm to the patient, computational modeling provides an alternative to gain knowledge about resulting cerebral temperature decrease. In this work, we present a brain temperature model with a realistic division into gray and white matter and consideration of spatially resolved perfusion. Furthermore, it includes detailed anatomy of cerebral circulation with possibility of personalizing on base of real patient anatomy. For evaluation of catheter performance in terms of cold reperfusion and to analyze its general performance, we calculated the decrease in brain temperature in case of a large vessel occlusion in the middle cerebral artery (MCA) for different scenarios of cerebral arterial anatomy. Congenital arterial variations in the circle of Willis had a distinct influence on the cooling effect and the resulting spatial temperature distribution before vessel recanalization. Independent of the branching configurations, the model predicted a cold reperfusion due to a strong temperature decrease after recanalization (1.4-2.2  C after 25 min of cooling, recanalization after 20 min of cooling). Our model illustrates the effectiveness of endovascular cooling in combination with mechanical thrombectomy and its results serve as an adequate substitute for temperature measurement in a clinical setting in the absence of direct intraparenchymal temperature probes.

Abstract:

Therapeutic hypothermia (TH) is an approved neuroproctetive treatment to reduce neurological morbidity and mortality after hypoxic-ischemic damage related to cardiac arrest and neonatal asphyxia. Also in the treatment of acute ischemic stroke (AIS), which in Western countries still shows a very high mortality rate of about 25 %, selective mild TH by means of Targeted Temperature Management (TTM) could potentially decrease final infarct volume. In this respect, a novel intracarotid blood cooling catheter system has recently been developed, which allows for combined carotid blood cooling and mechanical thrombectomy (MT) and aims at selective mild TH in the affected ischemic brain (core and penumbra). Unfortunately, so far direct measurement and control of cooled cerebral temperature requires invasive or elaborate MRI-assisted measurements. Computational modeling provides unique opportunities to predict the resulting cerebral temperatures on the other hand. In this work, a simplified 3D brain model was generated and coupled with a 1D hemodynamics model to predict spatio-temporal cerebral temperature profiles using finite element modeling. Cerebral blood and tissue temperatures as well as the systemic temperature were analyzed for physiological conditions as well as for a middle cerebral artery (MCA) M1 occlusion. Furthermore, vessel recanalization and its effect on cerebral temperature was analyzed. The results show a significant influence of collateral flow on the cooling effect and are in accordance with experimental data in animals. Our model predicted a possible neuroprotective temperature decrease of 2.5 ℃ for the territory of MCA perfusion after 60 min of blood cooling, which underlines the potential of the new device and the use of TTM in case of AIS.

Abstract:

Surgical navigation is revolutionizing complex procedures by providing surgeons with real-time guidance, enhancing precision, and improving patient outcomes. Augmented Reality (AR) further advances surgical navigation by overlaying digital information onto the surgical field, enabling surgeons to visualize and locate critical anatomical landmarks, even in challenging scenarios. A key requirement for AR frameworks is deformable registration, which ensures alignment between preoperative and intraoperative data while adapting to tissue shape and positional changes. However, the unpredictable deformability and variability of anatomical structures remain significant challenges for accurate registration.In this work, we propose a non-rigid 3D-3D registration framework tailored for laparoscopic procedures, with a focus on vascular structures. Our approach integrates deep learning-based feature extraction with a multi-step registration process, including initial rigid alignment followed by deformable refinement. The framework was trained and validated using a synthetic dataset that simulates intraoperative conditions, enabling the alignment of preoperative 3D models with intraoperative data. Our framework achieved an inlier ratio of 23.26%, a root mean square error of 34.24 mm and an inference time of 0.97 s. These results demonstrate the system's potential to enhance intraoperative navigation accuracy while maintaining computational efficiency, making it suitable for real-time surgical applications.

Abstract:

In laparoscopic surgery image-guided navigation systems could support the surgeon by providing subsurface information such as the positions of tumors and vessels. For this purpose, one option is to perform a reliable registration of preoperative 3D data and a surface patch from laparo-scopic video data. A robust and automatic 3D-3D registration pipeline for the application during laparoscopic surgery has not yet been found due to application-specific challenges. To gain a better insight, we propose a framework enabling a qualitative and quantitative comparison of different registration approaches. The introduced framework is able to evaluate 3D feature descriptors and registration algorithms by generating and modifying synthetic data from clinical examples. Different confounding factors are considered and thus the reality can be reflected in any simplified or more complex way. Two exemplary experiments with a liver model, using the RANSAC algorithm, showed an increasing registration error for a decreasing size of the surface patch size and after introducing modifications. Moreover, the registration accuracy was dependent on the position and structure of the surface patch. The framework helps to quantitatively assess and optimize the registration pipeline, and hereby suggests future software improvements even with only few clinical examples. Clinical relevance - The introduced framework permits a quantitative and comprehensive comparison of different registration approaches which forms the basis for a supportive navigation tool in laparoscopic surgery.

Abstract:

Stroke is the third-most cause of death in developed countries. A new promising treatment method in case of an ischemic stroke is selective intracarotid blood cooling combined with mechanical artery recanalization. However, the control of the treatment requires invasive or MRI-assisted measurement of cerebral temperature. An auspicious alternative is the use of computational modeling. In this work, we extended an existing 1D hemodynamics model including the characteristics of the anterior, middle and posterior cerebral artery. Furthermore, seven ipsilateral anastomoses were additionally integrated for each hemisphere. A potential stenosis was placed into the M1 segment of the middle cerebral artery, due to the highest risk of occlusion there. The extended model was evaluated for various degrees of collateralization (“poor”, “partial” and “good”) and degrees of stenosis (0%, 50%, 75% and 99.9%). Moreover, cerebral autoregulation was considered in the model. The higher the degree of collateralization and the degree of stenosis, the higher was the blood flow through the collaterals. Hence, a patient with a good collateralization could compensate a higher degree of occlusion and potentially has a better outcome after an ischemic stroke. For a 99.9% stenosis, an increased summed mean blood flow through the collaterals of +97.7% was predicted in case of good collateralization. Consequently, the blood supply via the terminal branches of the middle cerebral artery could be compensated up to 44.4% to the physiological blood flow. In combination with a temperature model, our model of the cerebral collateral circulation can be used for tailored temperature prediction for patients to be treated with selective therapeutic hypothermia.

Abstract:

In Western countries, stroke is the third-most cause of death; 35- 55% of the survivors experience permanent disability. Mild therapeutic hypothermia (TH) showed neuroprotective effect in patients returning from cardiac arrest and is therefore assumed to decrease stroke induced cerebral damage. Recently, an intracarotid cooling sheath was developed to induce local TH in the penumbra using the cooling effect of cerebral blood flow via collaterals. Computational modeling provides unique opportunities to predict the resulting cerebral temperature without invasive procedures. In this work, we generated a simplified brain model to establish a cerebral temperature calculation using Pennes’ bio-heat equation and a 1D hemodynamics model of the cranial artery tree. In this context, we performed an extensive literature research to assign the terminal segments of the latter to the corresponding perfused tissue. Using the intracarotid cooling method, we simulated the treatment with TH for different degrees of stenosis in the middle cerebral artery (MCA) and analyzed the resulting temperature spatialtemporal distributions of the brain and the systemic body considering the influence of the collaterals on the effect of cooling.

Abstract:

In Western countries, stroke is the third-most widespread cause of death. 80% of all strokes are ischemic and show a mortality rate of about 25%. Furthermore, 35-55% of affected patients retain a permanent disability. Therapeutic hypothermia (TH) could decrease inflammatory processes and the stroke-induced cerebral damage. Currently, the standard technique to induce TH is cooling of the whole body, which can cause several side effects. A novel cooling sheath uses intra-carotid blood cooling to induce local TH. Unfortunately, the control of the temporal and spatial cerebral temperature course requires invasive temperature measurements. Computational modeling could be used to predict the resulting temperature courses instead. In this work, a detailed 1D hemodynamics model of the cerebral arterial system was coupled with an energetic temperature model. For physiological conditions, 50% and 100% M1-stenoses, the temperatures in the supply area of the middle cerebral artery (MCA) and of the systemic body was analyzed. A 2K temperature decrease was reached within 10min of cooling for physiological conditions and 50% stenosis. For 100% stenosis, a significant lower cooling effect was observed, resulting in a maximum cerebral temperature decrease of 0.7K after 30min of cooling. A significant influence of collateral flow rates on the cooling effect was observed. However, regardless of the stenosis degree, the temperature decrease was strongest within the first 20min of cooling, which demonstrates the fast and effective impact of intra-carotid blood cooling.