2026 Poster Presentations
P453: DIGITAL 3D "SCULPTING" MULTILAYER MICROSURGICAL ANATOMY OF THE CAVERNOUS SINUS: A NOVEL APPROACH TO VISUALIZE AND UNDERSTAND SKULL BASE ANATOMY
Hao Tang, MD; Chiara Angelini, MD; Marco Obersnel, MD; Roberto Rubio, MD; Neurological surgery department of UCSF
Background: The application of 3D reconstruction techniques for neuroanatomical education has become a relevant tool in enabling the transition from 2D to 3D and allowing multilayer microsurgical anatomy to be demonstrated within a single visual platform. Traditional neurosurgical anatomy training typically involves progressive dissection of intact cadaveric specimens, carefully peeling away each layer while preserving deeper structures. Even though cadaveric dissections provide a good visuospatial and haptic experience of anatomy, the process largely remains “passive” when it comes to creating a deeper personal 3D reconstruction of structures. We therefore asked whether an alternative, reverse strategy might be feasible—namely, starting from the deepest layer and sequentially constructing each anatomical level in an “active procreational” manner, thereby fostering a detailed constructive understanding of spatial relationships.
Methods: Five embalmed and latex-injected cadaveric specimens were explored using an endoscopic endonasal approach (EEA) through the cavernous sinus to the mesial temporal lobe. Two axial middle cranial fossa specimens and two coronal sections across the orbital apex were dissected microscopically, and key anatomical landmarks were correlated with corresponding endoscopic views. These observations provided 2D and 3D anatomical evidence that served as references for sculptural modeling. Using Nomad Sculpt 2.2 for iPad (Hexanomad, Les Lilas, France), stepwise modeling of osseous structures, membranes, ligaments, vessels, venous sinuses, cranial nerves, and muscles was performed from the basal layer upward, with selective transparency applied as needed. The completed 3D model was subsequently rendered in Blender 4.5 (Blender Foundation, Amsterdam, Netherlands) and uploaded to a web-based 3D model viewer (Sketchfab, Sketchfab Inc., New York, USA).
Results: A volumetric model of the cavernous sinus and adjacent middle cranial fossa structures was successfully constructed, enabling free rotation, scaling, perspective viewing, and selective isolation of individual anatomical components. Figures 1–4 demonstrate the correspondence between cadaveric dissections and the sculptural models. The final model was deployable for both local interaction (Blender) and online platforms (Sketchfab).
Conclusion: This study demonstrates a novel method for constructing complex 3D anatomical structures from the basal layer upward that was entirely made by neurosurgeons. Compared with specimen-based volumetric reconstructions, the sculptural approach provides enhanced structural clarity, superior transparency, and intuitive visualization of spatial relationships through selective hiding and rotation of elements. For the first time, the complete 360-degree configuration of the cavernous sinus—including all membranes and walls—was presented within a single model. More importantly, the 3D “sculpting” modeling framework fosters an active, constructive, and detailed mode that reinforces anatomical understanding for the surgeon exploring anatomy physically and creating it on a 3D digital scene, underscoring its potential value in neuroanatomical education and structural visualization.
