AI-Powered Seating Assessment Tools Are Here — But Can They Replace a Skilled ATP?

I've been doing seating assessments for over two decades. I've sat across from clients with ALS, spinal cord injuries, cerebral palsy, MS, and everything in between. I've watched technology change this field in genuinely useful ways — pressure mapping, tilt-in-space power chairs, adjustable-tension backs. So when AI-powered seating assessment tools started showing up at trade shows and in my inbox, I didn't dismiss them. I went and looked closely.

Here's my honest read on where these tools are right now — what they can do, what they can't, and what you should watch out for.

What These Tools Actually Do

The current generation of AI seating tools generally falls into two categories.

The first is image-based posture analysis. You take a photo or video of the client sitting, and the software identifies landmarks — shoulder position, pelvic tilt, head position — and flags deviations. Some systems, like those being piloted in a few seating clinics in the UK and Canada, use depth cameras similar to what's in a Microsoft Kinect to build a 3D posture map. That data then gets cross-referenced against a database to suggest seating configurations.

The second is pressure mapping with AI interpretation. This one's further along. Companies like Tekscan have been doing pressure mapping for years, but newer software layers machine learning on top of that raw data to identify high-risk pressure zones and suggest cushion adjustments. Instead of just showing you a heat map, it says: "This distribution suggests posterior pelvic tilt and elevated ischial pressure — consider a contoured cushion with rear loading."

Those are genuinely useful prompts. I won't pretend otherwise.

Where the Tools Add Real Value

Let me be specific about what I think actually works.

In high-volume clinical settings — think a large VA hospital seeing 15 clients a week — AI-assisted pressure mapping can help a less experienced clinician catch things they might miss and document findings faster. That's not nothing. Documentation is a real bottleneck in this field, and if the software auto-generates a structured report from assessment data, that's an hour back in someone's day.

For remote or telehealth assessments, image-based posture screening has genuine potential. If a client lives four hours from the nearest seating clinic, having a caregiver take a calibrated video that gets analyzed before the appointment means the ATP walks in with a head start rather than starting cold.

Where I'd Pump the Brakes

Here's where I get concerned — and I want to be direct with you about this.

A seating assessment isn't just a posture snapshot. When I'm evaluating a client, I'm feeling for tone. I'm watching how their spasticity changes when they shift position. I'm asking about pain that gets worse after two hours in the chair, not ten minutes. I'm noticing that they flinch when I touch their left hip, and asking why. None of that is in a photograph.

I evaluated a client last year — a 34-year-old man with incomplete C5 SCI — who presented with what looked like a simple posterior pelvic tilt on visual assessment. An AI tool probably would have flagged it as a cushion fit issue. What it actually was: a developing hip flexion contracture on his right side that was rotating his whole pelvis. The fix wasn't cushion adjustment. It was a seating system with more aggressive lateral pelvic support and a referral back to his physio. You only find that with hands-on evaluation.

I'm also wary of how these tools handle complexity. AI systems trained on datasets tend to do well on common presentations and poorly on outliers. A client with mixed tone, asymmetrical positioning needs, and fragile skin who's also a full-time power chair user? That person is exactly who needs expert judgment, and they're exactly the person these tools are least equipped to help.

My Practical Advice If Your Clinic Is Considering These Tools

Use them as a second set of eyes, not a first opinion. AI pressure mapping analysis is a useful prompt, not a prescription. Treat its output the way you'd treat a note from a student — worth reading, requires verification.

Don't let them compress assessment time. I've already heard from clinics where administrators are eyeing these tools as justification for shorter appointments. Push back on that hard. The tool should make your assessment richer, not faster for the wrong reasons.

Ask vendors hard questions about their training data. What populations were used to build the model? How many power chair users versus manual? What diagnostic groups? If they can't answer clearly, be skeptical.

For clients with complex needs, there is no shortcut. If someone has significant spasticity, skin integrity concerns, progressive neurological conditions, or a history of pressure injuries, they need a qualified ATP doing a full hands-on evaluation. Full stop.

The Bottom Line

AI seating tools are real, some of them are useful, and this technology will get better. But right now, in 2024, they're assistants — not assessors. The gap between pattern recognition and clinical judgment is still wide, and the clients who fall into that gap tend to be the ones who can least afford a seating mistake.

If you're a client or caregiver, the right question isn't whether your clinic uses AI tools. It's whether you're being seen by someone who actually puts their hands on you, listens to your history, and treats your situation as individual. That part hasn't changed — and I don't think it should.