When Dr. Parth Desai was a medical student, he kept running into the same problem: a patient presents for revision surgery from an outside surgeon; there are no implant records; and nobody knows what company made the hardware sitting in their body.
A Spine Surgeon Built an AI Tool to Solve One of the OR’s Most Frustrating Problems. 50,000 Users Later, It’s Just Getting Started
3 min read Premium comments
AIMost PopularCompany NewsimplantImplant IdentifierParth Desai
The old solution was inefficient: search for an operative note, call device reps, text colleagues, or flip through catalogs. He thought there had to be a better way.
So, he built one.
Implant Identifier is a free mobile app that uses AI to identify orthopedic and spine implants from X-ray images. Users can upload a film, and the platform cross-references it against a growing library of real implant X-rays to return the closest matches in seconds.
“For commonly encountered implants, the system is now performing at greater than 99% accuracy in our internal testing,” said Dr. Desai. “As the dataset grows, we expect performance to continue to improve for rarer implants as well.”
Dr. Desai’s path to medicine began with entrepreneurship. Before starting medical school, he founded Devise Health, a healthcare IT consulting firm that built mobile applications for major health systems including Massachusetts General Hospital and Mount Sinai. Once he started medical school and decided to become an orthopedic surgeon, he wanted to focus on solving real problems faced by practicing surgeons. What started as a research project has grown into a community of more than 50,000 surgeons and device reps worldwide.
“As we’ve grown, this has become more than just implant identification,” Dr. Desai said. “Our goal is to unite orthopedic surgeons and device representatives under a single AI-driven pre-operative planning platform.”
That community-driven model is central to how the platform works. Implant Identifier combines AI-powered image matching with an expanding implant library and a collaborative community feed, allowing surgeons and reps to solve difficult implant identification cases together.
The growth has accelerated sharply. After years of beta development and data collection, Desai released the hip identification AI model in Q4 2025. Shoulder followed, then knee, and now ACDF plates for spine — all added within the last three months. Posterior spinal constructs are currently in development, with the goal of expanding AI identification across the full spectrum of orthopedic and spine implants.
“As we get closer to solving implant identification, we’re beginning to see the true power of AI — the possibilities are endless when a massive dataset is paired with a global community of surgeons and device reps.”
Desai is a practicing orthopedic spine surgeon based outside Atlanta, and he credits that clinical perspective with keeping the product grounded in real surgical needs.
“Building this app is a passion project,” said Dr. Desai. “I wanted it to remain free because the goal is to benefit the entire orthopedic community, and ultimately, the patients we treat.”
The app has remained free since day one, driven by a mission to make implant identification faster, easier, and more accessible for the orthopedic community.
Looking ahead, Implant Identifier is continuing to improve its AI models, expand its AI-powered pre-operative planning tools, and create new ways for surgeons and device reps to collaborate around revision surgery.
“We’re rolling out a new optional feature that allows surgeons to connect directly with the relevant device reps in their region,” said Dr. Desai. “This helps ensure the correct materials are available on surgery day, while also creating direct sales leads for device reps.”
More than 50,000 users later, the problem that once required phone calls, guesswork, and implant catalogs is being solved in seconds. The library keeps growing, the models keep improving, and the community is finding answers they could not find anywhere else.
You can learn more about Implant Identifier and download the app by going to https://implantidentifier.app/.
React:
Discussion
This is a fascinating development. In my practice we've seen similar outcomes with the revised protocol. The key differentiator seems to be patient selection criteria. Has anyone else noticed the correlation with BMI thresholds?
Great point. I'd push back slightly on the conclusion, the sample size in the cited study is too small to draw population-level inferences. That said, the directional signal is compelling and worth a larger RCT.
We implemented a similar approach last year. Early results are promising but we're still gathering 12-month follow-up data. Happy to share our protocol if anyone is interested.
Join the conversation
Orthopedic professionals are discussing this. Sign in and upgrade to read every comment and add your voice.