X23D · Real-time intelligence for surgical guidance

From intraoperative data into real-time surgical intelligence

X23D develops AI software for surgical navigation and robotics, transforming intraoperative imaging already available in the operating room into workflow-native, real-time, patient-specific 3D anatomical insight. By unlocking the latent value of existing imaging systems—without adding complex hardware or disrupting established workflows—we deliver spatial understanding of the surgical site and enable accessible, intelligent, scalable surgical guidance.

See the platform in action

Intraoperative data Spatial intelligence Navigation + robotics

AI-powered surgical guidance

Sparse intraoperative imaging turned into context-aware 3D guidance.

X23D-Spine is AI-powered surgical navigation software that transforms a few routine intraoperative X-rays into patient-specific 3D anatomy, giving surgical teams spatial guidance within the existing imaging workflow.

Explore the technology Talk to our team

3–4 views minimal intraoperative overhead Patient-specific 3D morphology preserved Device-agnostic designed to work with standard C-arms

Concept visualization · 00:52

01 · Acquire A sparse set of fluoroscopy views shown sufficient for complete lumbar-spine reconstruction

02 · Reconstruct Reported reconstruction time using real fluoroscopy data

03 · Guide Reported average reconstruction and guidance error with minimal additions to the surgical workflow

Performance reported in peer-reviewed studies

From images to spatial understanding

Surgical navigation reimagined:
simple, accurate and accessible.

X23D converts commonly available sparse intraoperative images into a patient-specific 3D representation of anatomy, revealing spatial relationships that cannot be interpreted reliably from individual 2D views alone.

Explore the peer-reviewed evidence

Peer-reviewed research

Evidence,
published.

Our published peer-reviewed studies trace the platform from its reconstruction foundations to real-fluoroscopy adaptation, workflow validation and next-generation tool tracking.

Title page crop of the Medical Image Analysis paper on domain adaptation for lumbar spine reconstruction

02 Open journal

Medical Image Analysis2024

Domain adaptation strategies for 3D reconstruction of the lumbar spine using real fluoroscopy data

Contribution. Bridges the synthetic-to-real domain gap, enabling accurate whole-lumbar reconstruction from as few as three real fluoroscopy images in approximately 81 ms.

View publication

Title page crop of the Journal of Imaging paper introducing X23D

01 Open journal

Journal of Imaging2022

X23D—Intraoperative 3D Lumbar Spine Shape Reconstruction Based on Sparse Multi-View X-ray Data

Contribution. Introduces X23D and demonstrates patient-specific lumbar reconstruction from four sparse views, outperforming a state-of-the-art baseline by 22% in surface score.

View publication

Title page crop of the BMC Musculoskeletal Disorders paper evaluating X23D navigation ex vivo

03 Open journal

BMC Musculoskeletal Disorders2024

Spinal navigation with AI-driven 3D-reconstruction of fluoroscopy images: an ex-vivo feasibility study

Contribution. Tests the complete navigation prototype in a realistic surgical workflow, showing promising pedicle-screw accuracy with minimal disruption to established practice.

View publication

Title page crop of the journal paper on markerless inside-out tool tracking for endoscopic spine surgery

04 Open journal

International Journal of Computer Assisted Radiology and Surgery2026

Markerless inside-out tool tracking for endoscopic spine surgery: a benchmarking study and clinical dataset

Contribution. Establishes a first-of-its-kind stereo endoscopy dataset and benchmark, demonstrating millimetre-scale tracking potential while identifying the remaining robustness gap.

View publication

Hover, focus, or tap a milestone to bring its paper forward.

X23D team conducting a navigation experiment in a surgical imaging facility — 01Clinical-grade environmentIntegrated imaging and navigation testing

Close-up of a tracked surgical instrument during X23D testing — 02Instrument evaluationTesting under realistic constraints

Clinical collaborator examining a tracked instrument during a navigation test — 03Surgeon-in-the-loopRequirements refined through use

X23D navigation software and tracked instruments being tested with surgical imaging equipment — 04End-to-end workflowSoftware, imaging and guidance together

Clinical validation · from day zero

Designed with surgeons.
Tested in the workflow.

Clinical validation is not a final checkpoint. It begins before the first line of code and continues through development, deployment and post-market learning.

Working side by side with orthopaedic surgeons, we evaluate algorithms, interfaces and complete navigation workflows in high-quality surgical facilities. Every cycle turns clinical insight into product requirements, so accuracy, usability and workflow fit are engineered in from the ground up.

Clinical evidence is a design input.

Independent recognition

X23D on Swiss national television.

See the X23D team demonstrate its approach to intraoperative 3D reconstruction on GESUNDHEITHEUTE, and hear the clinical vision behind the technology.

SRF MedTech Surgical innovation

Watch the full programme on SRF Play

Clinical · Academic · Innovation