In partnership with
We don’t do “business as usual”
The world moves fast, but understanding it shouldn’t be hard.
That’s why we created Morning Brew: a free, five-minute daily newsletter that makes business and finance news approachable—and even enjoyable. Whether it’s Wall Street, Silicon Valley, or what’s trending at the water cooler, the Brew serves up the context you need in plain English, with a side of humor to keep things interesting.
There’s a reason over 4 million professionals read the newsletter daily—and you can try it for free by clicking below!
Hi!
According to leaked data, Abridge is now the #1 user of OpenAI in healthcare — no other company processes as much clinical language through AI.
Abridge is an AI medical scribe that listens to millions of medical–patient conversations, and generates notes (other companies on the list are Duolingo, Indeed, and Shopify).
In the US, AI Scribes are being integrated into the clinical workflow. What’s your opinion on this topic?
Let’s dive into today’s issue.
🤖 AIBytes
Researchers from the Technical University of Munich tested whether a robotic system could safely support early mobilization in patients recovering from severe neurosurgical conditions in critical care.
🔬 Methods
Design: Randomized controlled pilot study.
Participants: 18 adults (median age 81 years) requiring intermediate neurocritical care who could follow simple commands.
Intervention:
Control: Standard physiotherapy (15 min/day × 14 days).
Study group: Standard physiotherapy + daily 15 min robotic therapy using VEMOTION®—a system that verticalizes the bed (up to 70°) and moves the legs passively or with assist-as-needed support.
Primary outcome: Adverse or serious adverse events.
Secondary outcomes: Physical and respiratory recovery measured with the Chelsea Critical Care Physical Assessment Tool (CPAx) at baseline and after 14 days.
Follow-up: Barthel Index at discharge and 3 months post-discharge.
📊 Results
No adverse or serious events related to robotic therapy.
Participants tolerated gradual verticalization.
Physical Function (CPAx):
Study group median improvement of 15 points.
Control group: 4 points.
p=0.0002
Largest improvements in study group: moving within the bed, dynamic sitting, and grip strength.
Activities of Daily Living (Barthel Index):
Study group:
Discharge: 10.
3 months: 55.
p=0.0156
Control group:
Discharge: 15.
3 months: 15.
Click here to watch in action👉 @VEMOTION
🔑 Key Takeaways
VEMOTION® robotic therapy is safe and well-tolerated, even in fragile neurocritical patients with monitoring lines and drains.
Adding robotic mobilization to standard therapy produced clinically significant gains in mobility and respiratory function.
Early robotic assistance may enhance long-term independence after discharge.
💡Larger, multicenter trials are warranted to validate and integrate this approach into standard neurocritical rehabilitation.
🔗 Joerger AK, Peper KK, Jensen ER, et al. Safety and efficacy of add-on robotic therapy for early mobilization in intermediate neurocritical care: a pilot study. J NeuroEngineering Rehabil. 2025; 22:208. DOI: 10.1186/s12984-025-01750-5
Researchers from the Cereneo Neurorehabilitation Clinic in Switzerland. investigated how accurately wearable sensors can count steps during activities of daily living (ADL) in patients with neurological conditions.
🔬 Methods
Participants: 7 adults (mean age = 72.7 years): 4 stroke, 1 Parkinson's, 1 spinal cord injury, 1 hydrocephalus.
Functional Ambulation Category scores: 2–5.
Protocol: 30 minutes of scripted activities of daily living.
Variables tested:
Sensor location: 7 triaxial accelerometers worn at wrists, chest, lower back, waist, thigh, and ankle.
Filter range: wide-band (≤15 Hz), medium-band (0.2–5 Hz), narrow-band (0.5–3 Hz).
Window length: 0.5 s, 2 s, 5 s.
5 machine learning algorithms.
📊 Results
Neurological population-specific algorithm detected 86.4% of total steps.
Step counters built for healthy users miss most steps in neurological gait. Detected only 11–47 %.
Best Sensor location: Waist. Worst: Wrist.
Filter: 0.5–3 Hz (captures true gait frequency, reduces noise). Window: 5 seconds
Ensemble models outperformed simpler ones.
🔑 Key Takeaways
Population-specific algorithms are essential — able-bodied models missed steps in patients with neurologic disease.
Waist-mounted sensors with optimized frequency filtering and longer analysis windows provide the most reliable step detection during real-world activities.
All algorithms underestimated steps, likely due to atypical gait.
💡Custom models are necessary to monitor steps in patient with neurological conditions.
🔗 Crozat F, Pohl J, Easthope C, Bauer CM, Kuster RP.Every Step Counts—How Can We Accurately Count Steps with Wearable Sensors During Activities of Daily Living in Individuals with Neurological Conditions? Sensors. 2025; 25(18):5657. DOI: 10.3390/s25185657
🦾TechTool
Turns your text or paper abstract into clean, readable slides.
The designs generated can be edited and exported.
Free tier available; paid plans start low.
Tracks human movement using computer vision — no sensors, no wearables, only a smartphone.
Used by clinics and coaches for rehab, posture, and movement retraining.
Gives you real-time feedback to improve gait, range of motion, or exercise form.
A clinical documentation assistant made for rehab professionals.
Built specifically for PTs, OTs, and SLPs. It listens, summarizes, and drafts notes instantly.
Works inside most Electronic Records, designed to match therapy workflows.
🧬AIMedily Snaps
An AI credential program for healthcare professionals Google Cloud and Adtalem Global Education (Link)
AI-powered Drug innovation to accelerate early-stage drug discovery (Link).
A paper on AI-driven neural implants for vision and hearing (Link).
ACMR Conference and Expo, 27-30 October in Chicago (Link).
🧩TriviaRX
In what year did the first robot-assisted surgery take place?
A) 1985
B) 1992
C) 1999
D) 2003
Now, let’s see if last week you had the correct answer: B) 1956
The term “artificial intelligence” was first introduced in 1956 by John McCarthy during the Dartmouth Conference, which is considered the founding event of AI as a scientific discipline.
That’s all for today.
As always, thank you for taking the time to read 🙂.
Every forward-thinking mind you invite makes this community stronger.
Share AIMedily with a colleague. Knowledge grows faster when it’s shared.
Thank you!
Until LLMs Friday.
Itzel Fer, MD PM&R
Forwarded this email? Sign up here
P.S. Which section is your favourite? Hit reply with your answer.
How did you like today's newsletter?
Seeking impartial news? Meet 1440.
Every day, 3.5 million readers turn to 1440 for their factual news. We sift through 100+ sources to bring you a complete summary of politics, global events, business, and culture, all in a brief 5-minute email. Enjoy an impartial news experience.
