Shape Sensing for Pre-Case Planning and Tool Selection

Picture a surgeon preparing for a complex case: multiple tool options on the table, an uncertain anatomy ahead. What if, instead of relying solely on intuition, planning could be guided by thousands of prior procedures? That’s the promise of shape sensing for pre-case planning and tool selection.

Imagine a future where procedural data not only reflects what happened in the past but actively guides the choices made before each case. By capturing how tools move, which ones succeed, and how different techniques play out across patient anatomies, we can begin to see patterns that guide smarter planning. While this is already being explored in certain areas with rigid instruments and 2D imaging, the real leap forward will come when we extend it to flexible tools with full 3D shape sensing.

This post is the first in our new series, “Every Flexible Medical Device Will Have Shape Sensing.” In our opener, we explained why flexible tools are the missing link in procedural data capture. Now, we’re diving into the first of four key use cases: how shape sensing enables smarter pre-case planning and tool selection.

What Procedural Data Means for Planning

At its core, procedural data combines two things:

How tools behave in real procedures — their movements, success rates, challenges, and outcomes.
The context of those procedures — anatomy, imaging, patient history, and intraoperative conditions.

Shape sensing is the enabler here: without capturing device movement continuously, flexible tools remain a blind spot in procedural data.

When captured over thousands of cases, this data forms a learning system. AI can mine it to answer questions like:

Which tool has the highest success rate for this anatomy?
What insertion pathway or maneuver typically works best?
How do certain devices perform across different patient subgroups?

Instead of guessing, planning becomes evidence-based, guided by the collective experience of an entire dataset.

How It’s Already Being Done (with Rigid Tools and 2D Imaging)

This concept isn’t theoretical — it’s already being applied in surgeries where instruments are rigid or imaging is limited. Here are some concrete examples of how procedural data is shaping pre-case planning today:

Spine surgery: Platforms analyze thousands of prior spine deformity cases to design patient-specific rods and surgical plans. These AI-driven systems continuously improve as more outcomes are fed back, helping surgeons select the right implants and alignment strategies.
Orthopedics: Tools use registries of past shoulder replacements to predict outcomes and recommend implant choices. Machine learning has also been applied to thousands of knee replacement cases, identifying which techniques and implants yield the best results for different patient groups.
2D imaging: Even with just plain X-rays, AI templating systems are learning from prior cases to automatically size and align implants. This allows pre-op planning with better accuracy than manual methods — even without CT or MRI.
Instrument optimization: Hospitals are analyzing intraoperative data to streamline surgical trays. By tracking which instruments are actually used, some have cut tray sizes by more than 50%, saving cost and time while improving efficiency.

These examples show the power of procedural data in rigid tool settings. They prove that if we can capture enough cases, AI can surface insights that make surgery safer, faster, and more consistent.

Shape Sensing Unlocks Flexible Tools

This is where shape sensing changes everything. By embedding a hair-thin optical fiber into a device, we capture its full 3D shape continuously, in real time. Over thousands of procedures, this creates a dataset that reveals:

How different tools behave in different anatomies
Which maneuvers and paths succeed most often
Where certain devices tend to struggle
Patterns in tool handling that correlate with better outcomes

Instead of relying solely on physician intuition, AI can learn from these datasets to recommend optimal tools and strategies for the next patient.

Shape Sensing for Pre-Case Planning

With a growing library of procedural data, AI can begin to inform surgeons on which tools are most successful in specific anatomies, what pathways minimize complications, and what time or efficiency benchmarks to expect. Some concrete examples include:

Endovascular Access: Shape-sensing data could reveal which guidewires and catheter shapes most reliably reach target vessels without prolapse.
Gastrointestinal Endoscopy: By analyzing loop formation patterns, datasets could highlight tools and techniques that reduce insertion time and patient discomfort.
Central Line Placement (PICC lines): Shape data can flag when a device tends toward the wrong venous path, helping AI recommend strategies before misplacement occurs.

Each of these examples demonstrates how procedural data feeds directly into pre-case decision making, providing physicians with a clearer picture before they begin.

Why It Matters

Surgical planning today often depends on individual experience. A highly skilled operator develops intuition over hundreds of cases, but that wisdom isn’t easily shared or scaled. By contrast, procedural datasets democratize experience. Every surgeon can benefit from the learnings of thousands of cases, not just their own.

For patients, this means more personalized, predictable, and efficient procedures. For clinicians, it means fewer failed attempts, reduced radiation exposure, and more confidence in tool selection. And for healthcare systems, it means fewer complications, shorter procedure times, and lower costs.

The Path Forward

We’re at the start of this journey. Rigid tool surgeries have proven the concept: data-driven planning works. Now, with shape sensing, flexible devices can join the equation. By combining device shape data with the broader context of each case (anatomy, imaging, physiology), we can build the next generation of procedural datasets.

These datasets won’t just inform tool selection and pre-case planning — they’ll form the foundation for AI-guided and eventually autonomous interventions. But it starts with the first step: capturing how our tools move, case by case, and learning from it.

The future of pre-case planning won’t be about guessing which device might work. It will be about knowing, with data-backed confidence, the best tool and approach for each patient — before the procedure even begins.

Looking Ahead

Pre-case planning and intelligent tool selection is just the beginning. With shape sensing, we finally have a scalable way to capture flexible device behavior and learn from it across thousands of procedures. That data doesn’t just make planning smarter — it sets the stage for everything that follows: real-time AI guidance, autonomous robotic interventions, and automated documentation.

This is the first of four transformative use cases we’ll explore in our “Every Flexible Medical Device Will Have Shape Sensing” series. Next up: how continuous shape sensing powers AI-guided procedures, making intraprocedural navigation smarter and safer.