Interactive microsurgical anatomy education using photogrammetry 3D models and an augmented reality cube

OBJECTIVE

This study sought to assess the use of an augmented reality (AR) tool for neurosurgical anatomical education.

METHODS

Three-dimensional models were created using advanced photogrammetry and registered onto a handheld AR foam cube imprinted with scannable quick response codes. A perspective analysis of the cube anatomical system was performed by loading a 3D photogrammetry model over a motorized turntable to analyze changes in the surgical window area according to the horizontal rotation. The use of the cube as an intraoperative reference guide for surgical trainees was tested during cadaveric dissection exercises. Neurosurgery trainees from international programs located in Ankara, Turkey; San Salvador, El Salvador; and Moshi, Tanzania, interacted with and assessed the 3D models and AR cube system and then completed a 17-item graded user experience survey.

RESULTS

Seven photogrammetry 3D models were created and imported to the cube. Horizontal turntable rotation of the cube translated to measurable and realistic perspective changes in the surgical window area. The combined 3D models and cube system were used to engage trainees during cadaveric dissections, with satisfactory user experience. Thirty-five individuals (20 from Turkey, 10 from El Salvador, and 5 from Tanzania) agreed that the cube system could enhance the learning experience for neurosurgical anatomy.

CONCLUSIONS

The AR cube combines tactile and visual sensations with high-resolution 3D models of cadaveric dissections. Inexpensive and lightweight, the cube can be effectively implemented to allow independent co-visualization of anatomical dissection and can potentially supplement neurosurgical education.

Immersive Surgical Anatomy of the Far-Lateral Approach

The far-lateral (FL) approach is a classic neurosurgical technique that enables access to the craniocervical junction, which includes the lower clivus, the anterior foramen magnum, and the first two cervical vertebrae. The FL approach also provides access to the inferior cranial nerves (i.e., CN IX, CN X, CN XI, and CN XII), distal portions of the vertebral artery (VA), and inferior basilar trunk. Recent advances in three-dimensional (3D) technology as well as dissections allow for a better understanding of the spatial relationships between anatomical landmarks and neurovascular structures encountered during neurosurgical procedures. This study aims to create a collection of volumetric models (VMs) obtained from cadaveric dissections that depict the FL approach's relevant anatomy and surgical techniques. We describe the relevant multilayer anatomy involved in the FL approach and discuss modifications of this approach as well. Five embalmed heads and two dry skulls were used to record and simulate the FL approach. Relevant steps and anatomy of the FL approach were recorded using 3D scanning technology (e.g., photogrammetry and structured light scanning) to construct high-resolution VMs. Images and VMs were generated to demonstrate major anatomical landmarks for the FL approach. The interactive models allow for clear visualization of the surgical anatomy and windows in 3D and extended reality, rendering a closer look at the nuances of the topography experienced in the laboratory. VMs can be valuable resources for surgical planning and anatomical education by accurately depicting important landmarks.

Immersive Surgical Anatomy of the Retrosigmoid Approach

The retrosigmoid approach (RS) approach is the workhorse of the posterolateral neurosurgical techniques to access various posterior fossa structures and even extends into the middle fossa. Many studies have detailed two-dimensional (2D) descriptions of the RS technique from either the lateral or posterior view. This study is the first to provide a comprehensive analysis of the RS technique, soft tissue, extracranial landmarks, and intracranial structures of the posterolateral region using interactive three-dimensional (3D) volumetric models (VMs). The visuospatial understanding of the neuroanatomical structures and landmarks of the RS approach is critical for successful surgeries with minimal complications. This study aims to create a collection of VMs and stereoscopic media for the relevant layer-by-layer soft tissue anatomy and step-by-step surgical technique of the RS approach using cadaveric dissections. Five embalmed heads and one dry skull were used to generate stereoscopic images and VMs using 3D scanning technology (i.e., photogrammetry and structured light scanning) to illustrate and simulate the RS approach. The extracranial structures were divided into myofascial, superficial vascular, superficial nerve, and bony anatomy. The RS approach was divided into seven major steps: patient positioning, incision of the skin, dissection of the scalp flap, dissection of the muscles, craniotomy, dural opening, and closure. Additionally, we described an anatomical classification of surgical corridors based on the cisternal segments of the cranial nerves exposed during the RS approach. We discussed the nuances of the keyhole variations of the RS approach and intradural modifications of the RS approach using 3D VMs to illustrate the surgical corridors and the intradural structures accessed. These interactive VMs allow for clear visualization and dynamically immersive experience for neuroanatomical studies of the RS approach in 360-degrees and virtual reality (VR). Computer graphics can be implemented in neurosurgery to facilitate our topographic knowledge, which is crucial for anatomical education, surgical planning, intraoperative decision making, and postoperative care.

Immersive Surgical Anatomy of the Frontotemporal-Orbitozygomatic Approach

The frontotemporal-orbitozygomatic (FTOZ) approach is widely used for accessing anterolateral lesions in skull base surgery. Many studies have described the technique and quantified the surgical exposure and freedom provided by the FTOZ approach. However, few studies have provided a detailed analysis of the technique and surgical landmarks using three-dimensional (3D) models. In this study, we aimed to create a collection of volumetric models (VMs) and stereoscopic media on the step-by-step surgical technique of the FTOZ approach using cadaveric dissections. The FTOZ approach was divided into eight major steps: positioning, incision of the skin, dissection of scalp flap, mobilization of the temporalis muscle, dissection of periorbita, craniotomy, drilling of basal structures, and dural opening. The MacCarty keyhole and inferior orbital fissure are major surgical landmarks that were referenced for the six bony cuts. Photogrammetry and structured light scanning were used to construct high-resolution VMs. We illustrated the two-piece FTOZ craniotomy, followed by the one-piece and three-piece FTOZ craniotomies. Stereoscopic images, videos, and VMs were produced for each step of the surgical procedure. In addition, the mini-orbitozygomatic (MOz) and orbitopterional (OPt) approaches were considered and described as possible alternatives to the FTOZ approach. Recent advances in 3D technology can be implemented in neurosurgical practice to further enhance our spatial understanding of neurovascular structures. Surgical approaches should be carefully selected and tailored according to the patient's unique pathology and needs.

Immersive Surgical Anatomy of the Pterional Approach

The pterional approach (PA) is a versatile anterolateral neurosurgical technique that enables access to reach different structures contained in the cranial fossae. It is essential for neurosurgical practice to dominate and be familiarized with its multilayer anatomy. Recent advances in three-dimensional (3D) technology can be combined with dissections to better understand the spatial relationships between anatomical landmarks and neurovascular structures that are encountered during the surgical procedure. The present study aims to create a stereoscopic collection of volumetric models (VM) obtained from cadaveric dissections that depict the relevant anatomy and surgical techniques of the PA. Five embalmed heads and two dry skulls were used to record and simulate the PA. Relevant steps and anatomy of the PA were recorded using 3D scanning technology (e.g. photogrammetry, structured light scanner) to construct high-resolution VM. Stereoscopic images, videos, and VM were generated to demonstrate major anatomical landmarks for PA. Modifications of the standard PA, including the mini-pterional and two-part pterional approaches, were also described. The PA was divided into seven major steps: positioning, incision of the skin, dissection of skin flap, dissection of temporal fascia, craniotomy, drilling of basal structures, and dural opening. Emphasis was placed on preserving the temporal branches of the facial nerve and carefully dissecting the temporalis muscle. The interactive models presented in this article allow for clear visualization of the surgical anatomy and windows in 360-degrees and VR. This new modality of recording neuroanatomical dissections renders a closer look at every nuance of the topography experienced by our team in the laboratory. By accurately depicting essential landmarks, stereoscopy and VM can be valuable resources for anatomical education and surgical planning.