The field's claimRehabilitation is becoming a closed-loop signal system
Zen Koh (MotusAcademy) argues that neurorehabilitation is shifting from compensation to precision plasticity — from measuring the task to retraining the signal, from clinic-bound sessions to home-scaled dose, from passive repetition to adaptive neuroplasticity. His framing: “Rehabilitation is becoming a closed-loop signal system … therapy redesigned as a learning architecture.”
He names the architecture the field needs — a Neuroplasticity Operating System with six layers. He describes it. The point of this page is simple: the portfolio below already instantiates all six.
“The moat is not the game. The moat is the recovery loop.” — and that loop is exactly the closed loop disclosed in EP 2 364 686 B1 / PCT EP2024/076634.
The mapping · six layersWhat the OS requires — and what already exists
| # | Layer (Zen Koh) | What it requires | Our component | Status |
|---|---|---|---|---|
| 1 | Impairment map | Identify the specific neuromotor problem — co-activation, spasticity, timing, coordination, fatigue, compensation. | HCWM / Predictor — reconstructs the hidden neuromotor state from sensor signals and computes counterfactuals (“what if I adjust this lever?”). Δ-parameter hypotheses (over-/under-tracking as asymmetry/spasticity marker, KT-041A). | 🟡 v0 |
| 2 | Signal interface | Wearables, EMG, motion/force sensors, robotic systems convert hidden physiology into measurable data. | craPos™ — force-aware robotic interface: 4× EPOS4, 100 Hz hard-real-time, encoder + torque, the “runder Tritt” captured as a signal, not a score. | ✅ built |
| 3 | Adaptive challenge | Tasks neither too easy nor too hard, progressing as the patient improves. | craPos gaits + Δ-parameters (progressive difficulty) and ECP-UGC™ composer (authorable adaptive content). | ✅ built |
| 4 | Feedback engine | Immediate, understandable, meaningful feedback that helps the patient discover better control. | ECP-UGC™ game feedback + craPos live visualization (Runder-Tritt-View) — the screen as a mirror for the nervous system. | ✅ built |
| 5 | Dose engine | Enough repetitions, intensity and continuity — moving practice beyond scarce clinic hours. | NAMRU (mobile/decentralized unit) + ECP-UGC™ home scaling — the clinic as a network, not a place. | 🟡 concept |
| 6 | Translation layer | Improvements must connect to real-world function — reaching, walking, dressing, participating. | ascros™ medical pathway (Class IIb stroke) + clinical Δ-outcome measures complementary to EDSS. | 🟡 reg. |
Zen Koh describes the operating system. craPos / HCWM builds all six layers — and patented the loop that holds them together.
The white space he namesPrecision × engagement × access — all three
Zen Koh's sharpest warning is “engagement is not recovery.” The breakthrough, he writes, lies in combining three capabilities that most systems hold only two of:
And his long-term picture — a “Recovery Grid” spanning hospital, clinic, home and community, unified by “closed-loop human learning, not the disease category” — is the cross-domain logic the passerelle already encodes (Sport → Medical → Space, one substrate).