A sensor slant board prototype became a connected rehabilitation platform: mobile and web apps for clinicians and researchers, an AWS back end, stable Bluetooth data and a software package ready for FDA 510(k) submission.
Rehabilitation still leans on the eye of the clinician. Range of motion is often judged visually, which makes progress hard to measure and harder to compare. A medical technology startup in Illinois, founded by a board-certified orthopedic surgeon with over 30 years of clinical experience, set out to change that with a smart slant board.
They had a working hardware prototype. What they needed was everything around it: firmware, mobile and web apps, cloud infrastructure and the regulatory paperwork. We built that connected rehabilitation platform as an MVP fit for clinical validation and investor demos.
The startup designed a slant board with four motion sensors, placed behind the knees and under the heels. Together they capture dorsiflexion range of motion, joint mobility and indicators of posterior chain tissue behavior while a patient stretches. The in-house team had built the prototype from off-the-shelf parts and lined up a third-party manufacturer to scale it.
A board alone is not a product a clinician can use. It needed custom firmware, apps that make the data readable at the point of care, a secure back end and documentation the FDA would accept. The deadline was tight, and the MVP had to work in real clinics and impress investors.
We assembled a cross-functional team: a project manager, a compliance officer, a business analyst, a UI/UX designer, iOS and Android developers, back-end and front-end developers, a C++ engineer, DevOps specialists and QA engineers. If you are staffing a similar build, the mix matters as much as the headcount.
Each phase answered one question before the next began: does the concept work, is the data reliable, can clinicians use it, and is it ready for regulators?
Phase 1
iOS was chosen first because it is common among rehabilitation clinicians. A lightweight app connected to the board over Bluetooth, captured raw mobility data and computed preliminary dorsiflexion metrics. The startup used it to validate the idea with stakeholders and investors.
Phase 2
Early tests showed dropouts. The cause was interference from nearby wireless networks and noise in the device wiring. Our engineers modified the board with filtering and electrical isolation, and sessions began to run reliably at home and in clinics.
Phase 3
The iOS app gained patient profiles, automatic device pairing and live mobility metrics. Web interfaces followed for clinicians and researchers, backed by AWS.
Phase 4
As the pilot entered real-world testing, the team added security controls, trend dashboards, guided session materials, PDF reports, an Android app and admin tools.
Phase 5
Our compliance specialist assembled the software documentation for the 510(k) pathway and guided the startup through FDA expectations for connected devices.
During 1 to 5 minute sessions, the app shows maximum, minimum and average knee and ankle angles for each leg, plus angle dynamics in real time.
Each session produces its own optimal flexibility angle, a starting point for future load planning and individual rehabilitation programs.
Clinicians manage patient lists and follow historical trend dashboards across rehabilitation sessions.
Visuals and descriptions walk patients through each stretch, so execution is consistent from one session to the next.
Clinicians review patient-specific results on the web. Researchers export aggregated, de-identified data for scientific work.
Session reports export to PDF. Admins centrally manage hospitals, users and the devices connected to the platform.
A key design question was where to process sensor data. The answer shaped cost, complexity and how the app behaves in a clinic with poor Wi-Fi.
| Option | Strengths | Drawbacks |
|---|---|---|
| On-device calculation, AWS for routing and storageRecommended | Less system complexity, lower cloud load, responsive sessions even with limited internet. Modern phones handle real-time sensor math without tradeoffs. | Not suited to large-scale retrospective analytics on the device itself. |
| Cloud-side calculation | Central processing suits heavy analytics over large historical datasets. | Adds complexity and cloud cost, and live sessions depend on a steady connection. |
Because the calculations use only real-time sensor data, the phone was the right place for them. The AWS back end handles session data routing and storage.
Connected medical devices draw scrutiny on both data security and documentation. These controls and deliverables were part of the MVP scope, not an afterthought.
As of February 2026, the startup is preparing its 510(k) pre-market notification.
The team ramped up quickly and was not afraid to challenge our early assumptions. They helped us refine the scope, focus on what mattered for an initial release, and avoid spending time and money on features that were not critical yet.
Test the radio link early. The proof of concept surfaced the Bluetooth problem before any clinical feature depended on it. The hardware fixes are now slated for the full product line.
Decompose features ruthlessly. Prioritizing what the first release truly needed kept software costs down. When testing raised new considerations, the scope moved without losing momentum.
Start compliance work with the code. Writing the cybersecurity file and risk management plan alongside development meant the FDA package was ready when the product was. The FDA's premarket notification 510(k) guidance is a useful starting point.
The web portals run on React.js and the back end on Node.js. If you need the same skills, you can hire React.js developers or work with our Node.js development company team. More builds like this one are in our case studies.
iOS is expected to be the most common platform among the target clinicians in physical rehabilitation. Starting there sped up the first delivery. The Android app followed once the core features were proven.
The root cause was interference from nearby wireless networks and noise in the device wiring. Our engineers modified the board hardware with filtering and electrical isolation, which made test sessions reliable at home and in clinics.
No stable connection is required for the calculations. Key metrics are computed on the phone, while the AWS back end handles data routing and storage. That keeps the app responsive where connectivity is limited.
Yes. On this project our compliance specialist assembled the cybersecurity file and software risk management plan for the 510(k) pathway, and guided the client on FDA documentation requirements.
Through feature decomposition and prioritization. We challenged early assumptions and focused on what the first release needed. If you have a similar product in mind, [get in touch](/contact).
Swift
Kotlin
React.js
Node.js
C++
Amazon Aurora PostgreSQL
Amazon Cognito
Amazon ElastiCache
AWS Lambda
Terraform
A US real estate firm ran its whole operation in Podio but had no single view of performance. A live analytics layer over 14 Podio apps now gives eight teams dashboards that load in under a second and update within a minute, cutting reporting time 40-50%.
A next-generation intelligent observability platform that delivers real-time visibility across IT, IoT, OT, applications, services, and telecom circuits. Scans 10 billion public IP addresses continuously, processes up to 100 billion data points simultaneously, and deploys in minutes instead of weeks.
We built a Cloud FinOps platform from the ground up that now tracks over $100 million in AWS and Azure spending. The platform gives finance teams real-time cost visibility, AI-powered optimization recommendations, and a full reseller partner center with billing, invoicing, and customer portfolio management.
Tell us where the work is stuck today. In one call you get a senior engineer's view of the approach, the risks worth planning for, and a realistic first release.
