AI · Quality and safety
Insurance companies · Health and car

AI Clinic to measure, correct and customize the rehabilitation on real time

Rehub transforms the patient’s camera into objective data of movement. It detects anatomical points, calculates ROM, speed, symmetry and compensation, generates immediate feedback during exercise and helps the professional adjust the therapy with clinical traceability. Without wearables. Without recording patient video. under a health product quality system.

CE Class IIa

Medical Products MDR 2017/745

ISO 13485

Quality Management System

ISO 27001

information security

GDPR · HIPAA

Privacy and data protection

El principio diferencial

Más que ver

From camera to clinical action

Each signal has to lead to a useful decision.

Rehub AI allows you to see what was previously invisible: how the patient moves, how he executes each exercise, when he compensates, how much he progresses and when he needs professional intervention.

1

Frictionless capture

The device camera records the execution without mandatory wearables

2

Analyze with context

Algorithms interpret mobility, rhythm, repetitions, symmetry, and trade-offs based on exercise

3

Activate a response

The patient receives feedback and the professional obtains information to intervene when it adds value

4

Measure the evolution

The execution is integrated with adherence, pain, questionnaires and treatment results

Detection

pts

Frequency

pts

More accurate than the human eye

Shoulders OK Symmetry

Column · Correct alignment

Right elbow Rom 88°

the technical stack

A layer of intelligence for every moment of rehabilitation

We do not use third party APIs for clinical analysis. The Estimation pose, the analysis and the decision are their own code trained with real medical data.

Recommendation engine to prescribe more accurately

ROM calculation, speed, repetitions, symmetry, motor control and compensation.

Computer Vision that interprets the execution

Detects anatomical points, calculates movement parameters and recognizes relevant deviations to offer feedback during exercise and objective data to the professional

Prioritization to manage by exception

It allows to identify patients who require attention due to high pain, low adherence, incidents or lack of progress to concentrate the clinical time where it matters most


Feedback adapted to guide without replacing the professional

Combines visual and voice instructions with the result of the execution to correct, reinforce and accompany the patient during the session



multivariate analysis to understand evolution

Relates execution, fulfillment, pain, questionnaires and time evolution to offer a more complete reading than an isolated metric






value for each actor

the same intelligence with three different results

The value proposition changes according to who uses the information and what decision you need to make

for the patient

More guidance and more confidence during recovery

The patient better understands what to do, receives feedback when he needs it, and can visualize his progress more clearly.

  • Real time correction
  • Visual and voice instructions
  • session-based reinforcement
  • Continuity from anywhere
for the professional

More useful information with less manual review

The professional has objective data and prioritized signals to adapt the treatment and manage more patients with clinical criteria.

  • Objective measurement of movement
  • Tracking by exception
  • Recommendation of exercises
  • contextualized clinical evolution
for the organization

More care capacity without losing traceability

The organization can deploy hybrid rehabilitation models, compare results on processes, users and data.

  • Clinical scalability
  • Corporate interoperability
  • Measurement of results
  • Security and operational control
European Regulation of AI

Responsible and safe artificial intelligence

Rehub® incorporates artificial intelligence under a manufacturer’s total control model, ensuring:

  • predictable and verifiable behavior at all times.
  • No automatic modifications or autonomous learning during its clinical use.
  • No autonomous decision capacity.
  • Exclusive human supervision in all therapeutic decisions.

In accordance with Regulation (EU) 2024/1689, REHUB® operates in strict regulatory compliance, since:

  • Its AI functions are strictly auxiliary and do not determine therapeutic behavior.
  • The deactivation of these modules does not affect the safety or clinical efficacy of the device.
  • It does not generate individual profiles or automatic diagnostic inferences.
  • Total control and traceability of all system operations by the manufacturer. Rehub® incorporates artificial intelligence under a manufacturer’s total control model.
Clinical validation

Scientific evidence peer-reviewed.

Every clinical decision of AI is supported by publications in indexed journals. Here the most recent studies.

+k
Patients treated with DyCare
+
Peer-Reviewed Scientific Publications
Randomized clinical studies
Hospitals and Participating Services
Sensors
· 2024

validation of computer vision-based rehabilitation in shoulder rehabilitation

Comparative study of precision vs. Manual clinical measurement in 240 sessions.

BMC Health Services
2023

Telerehabilitation Effectiveness in Chronic Low Back Pain

RCT with 312 patients — adherence and functional outcome at 12 weeks.

JMIR Rehabilitation
· 2023

Adherence patterns in ai-guided home rehabilitation

Analysis of 35,000 real sessions in the real portfolio of hospital patients.

Respiratory Medicine
· 2024

Pulmonary rehabilitation in COPD via digital platform

Prospective pilot in PDO II-III patients with 4 months of follow-up.

Annals of Oncology
· 2024

Prehabilitation and recovery in breast cancer patients

Protocol of pre-habilitation + post-mastectomy recovery. 80% adhesion.

European Stroke Journal
· 2023

Computer Vision for Post-Stroke Motor Recovery

case-control evaluating functional recovery of the affected member.

Confianza · seguridad · evidencia

Comprometidos con la seguridad, la transparencia y la excelencia

Descubre todas las certificaciones que respaldan nuestra determinación por construir una plataforma segura, de calidad y con base científica.

Security architecture

Clinical data. clinical safety.

Security is not an add-on — it’s built into every layer of the system. We built it as a health product from day one.

End-to-end encryption

TLS 1.3 in transit (all API + Apps)
AES-256 at rest (all clinical data)
Keys rotated and managed with KMS
Encrypted backups with separate key

Certified European hosting

Infrastructure in AWS EU Regions
Guaranteed data residence
AWS Healthcare Competency
Datacenters with ISO 27001 + SOC2

Default anonymization

Pseudonymized clinical data
Separate identity of usage data
We do not use patient videos outside of your session
Delete verifiable on request (GDP)

audit and traceability

immutable logs of each access
Granular Roles and Permits
Mandatory MFA for professionals
Annual External Audit Report
ethical commitment

Five commitments that we do not negotiate.

AI in health is not neutral — design decisions have clinical and ethical consequences. These are the principles that govern how we build DyCare, written in stone from day one.

We never use patient data to train external models. The data serves the treatment of the patient — point.
Transparent and auditable algorithms. Every clinical AI decision can be explained and reviewed by a professional.
Clinical decision always in the hands of the professional. AI recommends and corrects; The physical therapist decides.
Privacy by design. The patient can request its export or delete at any time — no questions asked.
No advertising tracking. Neither third party cookies, no pixels, or anything that is not at the service of rehabilitation.

Do you want to go into detail technical or regulatory?

We have Data Sheets, White Papers and regulatory documentation available to your legal, technical or clinical team. Request them.

Scroll to Top