What it is
A Dutch interdisciplinary research consortium.
A-PROOF develops artificial intelligence methods that create structured information on human functioning from written and spoken clinical text.
A-PROOF
AI-based Prediction of Recovery of Functioning.
A-PROOF is not an Avivly product. It is a Dutch interdisciplinary research consortium developing AI methods that create structured information on human functioning from written and spoken clinical text.
- Collaboration
- Consortium context
- Website contribution
Why it matters
Structured information about functioning is hard to extract, but essential in rehabilitation.
A-PROOF is trying to make rehabilitation data more structured and more useful by turning clinical language into organized information about functioning.
The project is grounded in the WHO International Classification of Functioning, Disability and Health, which gives the work a real clinical and research framework instead of a loose AI narrative.
That is why it belongs on this site as a collaboration: it sits close to evidence-based rehabilitation, clinical language, and responsible AI, even though it is not founder-led Avivly product work.
Clinical framework
Built around the WHO ICF framework.
The classifiers are grounded in the International Classification of Functioning, Disability and Health, so the output connects to a real rehabilitation language instead of generic AI labels.
Validation
Validated on clinical data from Amsterdam UMC.
The platform is positioned as evidence-based rehabilitation infrastructure, with accuracy, transparency, and responsible AI as core claims.
Demo status
The public demo is still in development.
The live site describes a real-time demonstration for spoken and written input, but it is still being tested and validated by the consortium before broader public release.
ICF scope
The current classifier set already covers a meaningful range of rehabilitation categories.
Emotional functions
Sensation of pain
Exercise tolerance
Mobility of joint functions
Muscle power functions
Walking and moving around
Work and employment
The public site lists areas such as emotional functions, pain, exercise tolerance, walking, mobility, and work participation. That reinforces that A-PROOF is about functioning in the rehabilitation sense, not just generic prediction output.
Consortium
A-PROOF is built by a broader research network.
The site positions A-PROOF as a consortium across rehabilitation medicine, text mining, ethics, and applied AI.
Amsterdam UMC, Department of Rehabilitation Medicine
Vrije Universiteit Amsterdam, CLTL Text Mining Lab
Leiden University, Department of Psychology, Ethics and Health
Amsterdam University of Applied Sciences, Centre of Expertise Applied AI
My role
My contribution sits in the public-facing layer, not in claiming ownership of the consortium.
The live A-PROOF site explicitly credits the website as developed and designed by Aviv Hidrian at Avivly.
That is the right framing here as well: I contributed through the website, design direction, and selected product and AI exploration around the public-facing experience.
A-PROOF itself should still be understood as consortium work, not as a standalone Avivly product.
If you want to understand my own core products, start with PT-CHARLIE and EBPcharlie. A-PROOF belongs here as a serious collaboration in rehabilitation AI.
Source framing on the live A-PROOF site: Dutch interdisciplinary consortium, ICF-based classifiers, Amsterdam UMC validation, and a demo still in development.