Orthopedic and Rehabilitation Device Development

The short answer. Orthopedics is one of the main areas of OVA Solutions: rehabilitation aids, range-of-motion and gait monitors, post-operative therapy devices, powered rehabilitation equipment and surgical fixation tools. We produced 100 pairs of a load-sensing rehabilitation crutch, took a cold therapy device from concept to clinical trials in 12 months, and built the electronics for a rehabilitation chair in 16 weeks. Recovery happens at home, so the device has to measure what the surgeon prescribed and keep the patient using it.

I would say that we have three main areas of expertise which are orthopedics and everything related to orthopedics bone drilling equipment implants itself at home rehabilitation sensor and so on.

What we build

Projects from our portfolio

Engineering decisions in orthopedic devices

Measure what the surgeon prescribed

A crutch measures load through the crutch. The prescription is load through the limb. The two are linked by gait pattern, body weight and the other leg, and the link changes from a three-point to a four-point gait. A 2016 comparison of three commercial biofeedback devices against a force plate found agreement of 0.76, 0.58 and 0.19 (weighted kappa) for exactly this reason. The clinical claim should be written against the measured variable, and the inference validated across gait patterns, not standing still on a force plate.

Feedback has to land inside the step

Load information after the step is a report; during the loading phase it is a correction, and only the correction changes gait. That sets the sampling rate, the latency and the actuator, so latency belongs in the requirements as a number. Patient-facing displays are ruled out during walking: looking down on a partially loaded limb is how elderly patients fall. Haptic feedback also performs better: a 2014 study measured 22.4 pounds of loading with haptic feedback against 43.8 with bathroom scale training and 60.3 with verbal instruction.

Claim change over weeks, not an absolute angle

Body-worn inertial sensors measure knee flexion with an error of 4.4 to 9.4 degrees, while the deficits after knee replacement are 4.8 to 6.6 degrees. Most of that error is a roughly constant bias from mounting and soft tissue, which largely cancels when the same device compares week two with week eight. Moving the claim to within-patient change turns the validation into a test-retest design and saves a failed accuracy study.

Keep the magnetometer out of the clinical axis

A knee prosthesis is a mass of cobalt chromium and titanium right under the sensor. In a 2023 case series with prosthesis users, one participant’s discrete-point error reached 29.21 degrees because a microprocessor-controlled prosthetic knee interfered with the magnetometer. Knee flexion is a sagittal rotation, so it can be recovered from the gyroscope bounded by the gravity vector, with magnetic heading kept out of it.

Mounting beats sensor selection

Straps over skin and muscle move relative to the bone, and every millimeter of donning variation shows up as degrees. Strap design, placement against bony landmarks and a short guided posture at the start of each session matter more than the choice between two IMU parts.

Tooling and volume decide unit cost

For an injection-molded ABS or polycarbonate medical housing, our cost modeling gives:

Annual volume	Unit cost of the housing	Resin per unit	Tooling amortization per unit
10,000	$0.51 to $2.45	$0.15 to $0.35	$0.20 to $1.50 or more
50,000	$0.33 to $1.35	$0.15 to $0.35	$0.05 to $0.50

A program that tools for 50,000 and sells 8,000 is stuck with that cost. We recommend one tool with interchangeable inserts for several configurations and, when the forecast is uncertain, a bridge process for the first production block.

Test the mechanics the way surgeons will use them

For the cranial fixation tool a robotic arm performed 100 standardized insertions under constant torque to measure durability and bit wear before design for manufacturing. Watching real use matters as much as rigs:

And we had a project where surgeons love the technical specs of the device, but the handle was wrong for how they naturally, you know, grip the instruments during like 4-hour procedure. So that’s something you can only learn by watching someone use it.

Published numbers we design against

What was measured	Published value	Study
Compliance with partial weight bearing after lower-extremity surgery	37.5%	2020
Same, patients 65 and over with femur or pelvic fractures	22%	2020
Same, after total knee arthroplasty, day 1 / before discharge	0% / 2%	2023 trial
Smart insole load measurement against a force plate	Intraclass correlation about 0.97	2025 validation
Instrumented crutch prototype, load prediction	99.3% precision	2016
Knee flexion error of body-worn inertial sensors	4.4 to 6.0 degrees slow, 5.4 to 9.4 fast movements	2026 validation
Gait deficits more than a year after knee replacement	4.8 to 6.6 degrees	2024 systematic review
Patient acceptable symptom state after knee replacement	50 degrees maximum flexion, 1.2 m/s walking speed	2020

In the clinic you get a snapshot like a single frame. The patient walks in they’re nervous. They’re on their best behavior and you measure them for a couple of minutes. But the real recovery doesn’t happen in this frame. So it happens in the other 167 hours at home when no one is watching.

FDA and reimbursement for orthopedic devices in 2026

Orthopedic devices were the second-largest 510(k) panel in 2025: 404 of 3,198 clearances (12.6%), behind radiology. Physical medicine, which covers most rehabilitation equipment, had 111 (3.5%). The most cleared orthopedic product codes were bone fixation plates (54), lumbar fusion devices (29), soft-tissue fixation fasteners (27) and pedicle screw systems (24). In physical medicine powered wheelchairs (29), motorized three-wheeled vehicles (15) and powered muscle stimulators (15) led.

• New in 2026: FDA finalized its guidance on patient-matched guides for orthopedic implants on May 7, 2026.
• New in 2026: non-invasive bone growth stimulators moved from Class III to Class II with special controls, effective May 18, 2026.
• Additive manufacturing: FDA’s 2017 technical considerations guidance is still the current one.
• Non-spinal bone screws: performance criteria under the Safety and Performance Based Pathway, revised November 2024, with testing per ASTM F543. Fracture fixation plates have their own criteria (April 2022), and FDA’s 510(k) guidance for non-spinal plates, screws and washers was finalized in November 2024.

Area	Test standard	Current version
Spinal constructs	ASTM F1717	F1717-21
Intervertebral fusion devices	ASTM F2077	F2077-24 (F2077-18 accepted by FDA until July 4, 2027)
Bone screws	ASTM F543	F543-23 (F543-17 accepted until December 20, 2026)
Bone plates	ASTM F382	F382-24 (F382-17 accepted until December 20, 2026)
Knee wear	ISO 14243-1 and -3	2009 and 2014 with 2020 amendments; revision in progress
Hip stem and neck endurance	ISO 7206-4 and -6	2010 with 2016 amendment; 2013, new edition at DIS ballot
FEA of hip stems and knee components	ASTM F2996 and F3161	F2996-24, F3161-24
Additively manufactured Ti-6Al-4V	ASTM F3001	F3001-14 (reapproved 2021)

Remote therapeutic monitoring for musculoskeletal devices

Medicare pays remote therapeutic monitoring (RTM) codes for musculoskeletal devices such as instrumented rehabilitation aids. Since 2026 a device that sends data on as few as 2 days in 30 qualifies for device supply under 98985, at the same amount as 16 to 30 days under 98977.

Code	What it pays for	2026 national amount
98975	RTM set-up and patient education (now 2 days of data)	$21.71
98977	Musculoskeletal device supply, 16 to 30 days of data in 30	$51.44
98985	Musculoskeletal device supply, 2 to 15 days (new in 2026)	$51.44
98980	RTM treatment management, first 20 minutes	$54.11
98979	RTM treatment management, first 10 minutes (new in 2026)	$26.39
98981	Each additional 20 minutes	$41.42

How an orthopedic program runs: two real examples

Rehabilitation chair, electronics and firmware only. The client’s team kept the mechanics; we took the control box, the control panel and the remote.

1. Discovery about 280 hours

   Requirements, interfaces with the client’s mechanics, architecture and the plan.

2. First prototype about 720 hours

   Two boards and a remote, firmware developed under software safety Class B.

3. Testing and iteration about 500 hours

   Bench testing and design changes from the results.

4. Compliance and documentation

   Project management, CTO engagement and the documentation package for a 510(k) submission. Total: 2,100 hours in 16 weeks, $260,000 of engineering, five pre-production units and a complete design history file.

Load-sensing crutch, full cycle. Early prototypes were 3D printed in several materials, then the housing moved to aluminum molds and passed durability tests before 100 pairs were produced. The program took two phases over eight months and about 1,660 engineering hours, with a team of eight across electronics, mechanics, industrial design, quality, business analysis and project management.

Questions

Related pages

• Rehabilitation crutch case100 pairs produced
• Rehabilitation chair case2,100 hours, 16 weeks
• Cold therapy caseConcept to clinical trials in 12 months
• Wearable devicesSensors, power and BLE
• 510(k) review timesFDA data 2026
• What development costs$125 per hour, FDA fees

Sources

Updated on October 9, 2026.