How AI Planning Improves Implant Sizing
How vendor-neutral CT-based AI planning supports implant sizing and selection, why the plan must stay independent of who pays for the software, and how OEMs license it.
Key takeaways
For an implant manufacturer, implant sizing accuracy in AI planning is not a clinical feature to admire from a distance, it is a commercial lever. Modern CT-based planning turns a stack of imaging into a 3D model, computes alignment and rotation on the patient's own anatomy, and fits that plan to a specific implant library, which supports more consistent size and offset selection before the patient is on the table. This is a workflow benefit, and it has a boundary that matters: the plan is computed on the patient's anatomy and stays independent of who paid for the software. Component fitting happens afterwards, from the library the surgeon selected. Manufacturers without a robot or an in-house planner still want to be present where planning happens, and a vendor-neutral, implant-agnostic planner offers a way in: the OEM licenses the CT-to-plan software as a customer and keeps its own implant relationship. Salnus builds exactly this software for manufacturers to license, and it is currently Research Use Only (RUO), not a cleared device, so engagements are framed as pilot, co-development, or licensing rather than the purchase of a finished cleared product.
Why sizing is where planning creates value
In arthroplasty, choosing an implant size and position from 2D radiographs and templates leaves room for intra-operative surprises: a size revised on the field, a component that does not match the anatomy as expected, or trays opened that are never used. CT-based 3D planning narrows that uncertainty. By segmenting the bone and reconstructing true dimensions, the software can support a size and offset recommendation grounded in the patient's actual geometry rather than a projected film. For a manufacturer, that is directly relevant to how its catalog is selected and used, and it is a natural extension of the workflow described in how CT-based 3D preoperative planning works.
The value is a decision-support value, not an outcome claim. Better inputs to a sizing decision are not the same as a proven clinical result, and any such claim would require the appropriate evidence and clearance that RUO software does not carry.
Why the plan increasingly drives the implant
The center of gravity in arthroplasty is moving from the hardware in the operating room to the plan made before it. When a surgeon reviews and accepts a digital plan, three things are quietly decided at once: the alignment philosophy, the rotation and sizing, and the implant family the plan is expressed on. Whoever controls the planning surface therefore has meaningful influence over which implant is selected.
Robotic ecosystems understood this early, which is why their planning is bound to their own robot and implant. That design is powerful for the vendor that owns the robot and a structural disadvantage for everyone else. A manufacturer without that stack risks becoming a commodity component that a rival's plan simply slots in or leaves out. This is the same dynamic examined in robotic versus software-based TKA planning.
What a vendor-neutral planner offers an OEM
An implant-agnostic planner does the anatomy and plan work independently of any single implant, then applies the chosen manufacturer's component library at the end. For an OEM, that separation is the point:
- A modern digital planning offer without building a robot or a planning platform from scratch.
- Sizing and alignment support that fits the surgeon's existing DICOM workflow rather than forcing a new hardware purchase.
- CPAK phenotyping and alignment options presented on the anatomy, so the plan can be expressed on the OEM's catalog across cases.
- Domestic-production relevance: in markets pushing local manufacturing, the planning software is often the last imported layer of the value chain, and a licensed planner closes it.
The mechanics of this separation, and why it is a positioning choice rather than a technical trick, are covered in implant-agnostic surgical planning. For a fuller map of who currently offers what, see which companies offer AI knee planning software and the wider orthopedic 3D planning software comparison for 2026.
How this is paid for, and where the boundaries sit
Preoperative planning time is not separately reimbursed, so the surgeon is a poor payer. That does not mean an implant maker can simply fund surgeon access on a platform that presents itself as neutral: that specific structure runs into anti-kickback rules wherever a federal healthcare programme or a public-sector physician is in scope, and it is difficult in Turkey for separate reasons. The two structures that hold are white-label licensing, where the manufacturer is the customer and the product carries its own brand, and institution-pays, where a hospital or surgical facility pays per case for the neutral platform. The reasoning is set out in the manufacturer-pays model for planning.
Engagement typically starts as a pilot or co-development program, not a shrink-wrapped product sale. The OEM keeps its implant relationship and brand, while Salnus provides the vendor-neutral CT-to-plan layer: segmentation, CPAK phenotyping, alignment options, and sizing support. IP is contract-defined, so field-of-use, exclusivity, and data rights are negotiated explicitly rather than assumed. The commercial framing of a licensed, vendor-neutral planner is described in more depth on the Salnus for manufacturers page.
The honest caveats
Two boundaries matter and should be stated plainly. First, regulatory status: Salnus is Research Use Only today, not a cleared device, so any clinical deployment requires the appropriate clearance and quality processes, and nothing here should be read as a claim of clinical superiority or a regulatory-approved outcome. Second, vendor-neutral is not uncontrolled: data governance, validation, and the regulatory pathway still apply, and implant-agnostic positioning does not remove the need for evidence and compliance. A serious OEM engagement plans for that pathway from the start rather than treating it as an afterthought.
Bottom line
Sizing and alignment accuracy is where CT-based AI planning creates concrete workflow value, because size and position are worked out before the patient is on the table rather than during the case. A vendor-neutral, implant-agnostic planner lets an OEM, especially a domestic or mid-size maker without a robot, offer modern digital planning on its own catalog under a white-label licence, with IP defined by contract and the work structured as a pilot or co-development rather than the purchase of a finished cleared product. Salnus builds this software for manufacturers to license, as Research Use Only today. If you are evaluating how to offer digital planning with your own implants, start with the Salnus for manufacturers overview.
Reviewed by the Salnus biomedical engineering team.