Pose Estimation and Non-Rigid Registration for Augmented Reality During Neurosurgery

A review on organ deformation modeling approaches for reliable surgical navigation using augmented reality

Augmented Reality (AR) holds the potential to revolutionize surgical procedures by allowing surgeons to visualize critical structures within the patient's body. This is achieved through superimposing preoperative organ models onto the actual anatomy. Challenges arise from dynamic deformations of organs during surgery, making preoperative models inadequate for faithfully representing intraoperative anatomy. To enable reliable navigation in augmented surgery, modeling of intraoperative deformation to obtain an accurate alignment of the preoperative organ model with the intraoperative anatomy is indispensable. Despite the existence of various methods proposed to model intraoperative organ deformation, there are still few literature reviews that systematically categorize and summarize these approaches. This review aims to fill this gap by providing a comprehensive and technical-oriented overview of modeling methods for intraoperative organ deformation in augmented reality in surgery. Through a systematic search and screening process, 112 closely relevant papers were included in this review. By presenting the current status of organ deformation modeling methods and their clinical applications, this review seeks to enhance the understanding of organ deformation modeling in AR-guided surgery, and discuss the potential topics for future advancements.

Learning to Match 2D Keypoints Across Preoperative MR and Intraoperative Ultrasound

We propose in this paper a texture-invariant 2D keypoints descriptor specifically designed for matching preoperative Magnetic Resonance (MR) images with intraoperative Ultrasound (US) images. We introduce a matching-by-synthesis strategy, where intraoperative US images are synthesized from MR images accounting for multiple MR modalities and intraoperative US variability. We build our training set by enforcing keypoints localization over all images then train a patient-specific descriptor network that learns texture-invariant discriminant features in a supervised contrastive manner, leading to robust keypoints descriptors. Our experiments on real cases with ground truth show the effectiveness of the proposed approach, outperforming the state-of-the-art methods and achieving 80.35% matching precision on average.

2D/3D deformable registration for endoscopic camera images using self-supervised offline learning of intraoperative pneumothorax deformation

Shape registration of patient-specific organ shapes to endoscopic camera images is expected to be a key to realizing image-guided surgery, and a variety of applications of machine learning methods have been considered. Because the number of training data available from clinical cases is limited, the use of synthetic images generated from a statistical deformation model has been attempted; however, the influence on estimation caused by the difference between synthetic images and real scenes is a problem. In this study, we propose a self-supervised offline learning framework for model-based registration using image features commonly obtained from synthetic images and real camera images. Because of the limited number of endoscopic images available for training, we use a synthetic image generated from the nonlinear deformation model that represents possible intraoperative pneumothorax deformations. In order to solve the difficulty in estimating deformed shapes and viewpoints from the common image features obtained from synthetic and real images, we attempted to improve the registration error by adding the shading and distance information that can be obtained as prior knowledge in the synthetic image. Shape registration with real camera images is performed by learning the task of predicting the differential model parameters between two synthetic images. The developed framework achieved registration accuracy with a mean absolute error of less than 10 mm and a mean distance of less than 5 mm in a thoracoscopic pulmonary cancer resection, confirming improved prediction accuracy compared with conventional methods.

Augmented Reality in Neurosurgery

Augmented Reality (AR) involves superimposing digital content onto the real environment. AR has evolved into a viable tool in neurosurgery, enhancing intraoperative navigation, medical education and surgical training by integrating anatomical data with the real world. Neurosurgical AR relies on several key techniques to be successful, which includes image segmentation, model rendering, AR projection, and image-to-patient registration. For each of these technical components, different solutions exist, with each having their own advantages and limitations.Intraoperative AR applications cover diverse neurosurgical disciplines including vascular, oncological, spinal, and functional surgeries. Preliminary studies indicate that AR may improve the understanding of complex anatomical structures and offer sufficient accuracy for use as a navigational tool. Additionally, AR shows promise in enhancing surgical training and patient education through interactive 3D models, aiding in the comprehension of intricate anatomical details. Despite its potential, the widespread adoption of AR in clinical settings depends on overcoming technical limitations and validating its clinical efficacy.

Learning Expected Appearances for Intraoperative Registration during Neurosurgery

We present a novel method for intraoperative patient-to-image registration by learning Expected Appearances. Our method uses preoperative imaging to synthesize patient-specific expected views through a surgical microscope for a predicted range of transformations. Our method estimates the camera pose by minimizing the dissimilarity between the intraoperative 2D view through the optical microscope and the synthesized expected texture. In contrast to conventional methods, our approach transfers the processing tasks to the preoperative stage, reducing thereby the impact of low-resolution, distorted, and noisy intraoperative images, that often degrade the registration accuracy. We applied our method in the context of neuronavigation during brain surgery. We evaluated our approach on synthetic data and on retrospective data from 6 clinical cases. Our method outperformed state-of-the-art methods and achieved accuracies that met current clinical standards.

Intra-operative applications of augmented reality in glioma surgery: a systematic review

Background

Augmented reality (AR) is increasingly being explored in neurosurgical practice. By visualizing patient-specific, three-dimensional (3D) models in real time, surgeons can improve their spatial understanding of complex anatomy and pathology, thereby optimizing intra-operative navigation, localization, and resection. Here, we aimed to capture applications of AR in glioma surgery, their current status and future potential.

Methods

A systematic review of the literature was conducted. This adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline. PubMed, Embase, and Scopus electronic databases were queried from inception to October 10, 2022. Leveraging the Population, Intervention, Comparison, Outcomes, and Study design (PICOS) framework, study eligibility was evaluated in the qualitative synthesis. Data regarding AR workflow, surgical application, and associated outcomes were then extracted. The quality of evidence was additionally examined, using hierarchical classes of evidence in neurosurgery.

Results

The search returned 77 articles. Forty were subject to title and abstract screening, while 25 proceeded to full text screening. Of these, 22 articles met eligibility criteria and were included in the final review. During abstraction, studies were classified as "development" or "intervention" based on primary aims. Overall, AR was qualitatively advantageous, due to enhanced visualization of gliomas and critical structures, frequently aiding in maximal safe resection. Non-rigid applications were also useful in disclosing and compensating for intra-operative brain shift. Irrespective, there was high variance in registration methods and measurements, which considerably impacted projection accuracy. Most studies were of low-level evidence, yielding heterogeneous results.

Conclusions

AR has increasing potential for glioma surgery, with capacity to positively influence the onco-functional balance. However, technical and design limitations are readily apparent. The field must consider the importance of consistency and replicability, as well as the level of evidence, to effectively converge on standard approaches that maximize patient benefit.

A Tool-free Neuronavigation Method based on Single-view Hand Tracking

This work presents a novel tool-free neuronavigation method that can be used with a single RGB commodity camera. Compared with freehand craniotomy placement methods, the proposed system is more intuitive and less error prone. The proposed method also has several advantages over standard neuronavigation platforms. First, it has a much lower cost, since it doesn't require the use of an optical tracking camera or electromagnetic field generator, which are typically the most expensive parts of a neuronavigation system, making it much more accessible. Second, it requires minimal setup, meaning that it can be performed at the bedside and in circumstances where using a standard neuronavigation system is impractical. Our system relies on machine-learning-based hand pose estimation that acts as a proxy for optical tool tracking, enabling a 3D-3D pre-operative to intra-operative registration. Qualitative assessment from clinical users showed that the concept is clinically relevant. Quantitative assessment showed that on average a target registration error (TRE) of 1.3cm can be achieved. Furthermore, the system is framework-agnostic, meaning that future improvements to hand-tracking frameworks would directly translate to a higher accuracy.

Visualization, navigation, augmentation. The ever-changing perspective of the neurosurgeon

Introduction

The evolution of neurosurgery coincides with the evolution of visualization and navigation. Augmented reality technologies, with their ability to bring digital information into the real environment, have the potential to provide a new, revolutionary perspective to the neurosurgeon.

Research question

To provide an overview on the historical and technical aspects of visualization and navigation in neurosurgery, and to provide a systematic review on augmented reality (AR) applications in neurosurgery.

Material and methods

We provided an overview on the main historical milestones and technical features of visualization and navigation tools in neurosurgery. We systematically searched PubMed and Scopus databases for AR applications in neurosurgery and specifically discussed their relationship with current visualization and navigation systems, as well as main limitations.

Results

The evolution of visualization in neurosurgery is embodied by four magnification systems: surgical loupes, endoscope, surgical microscope and more recently the exoscope, each presenting independent features in terms of magnification capabilities, eye-hand coordination and the possibility to implement additional functions. In regard to navigation, two independent systems have been developed: the frame-based and the frame-less systems. The most frequent application setting for AR is brain surgery (71.6%), specifically neuro-oncology (36.2%) and microscope-based (29.2%), even though in the majority of cases AR applications presented their own visualization supports (66%).

Discussion and conclusions

The evolution of visualization and navigation in neurosurgery allowed for the development of more precise instruments; the development and clinical validation of AR applications, have the potential to be the next breakthrough, making surgeries safer, as well as improving surgical experience and reducing costs.