AI That Heals Clinical AI Infrastructure | Predictive Models | Voice & Transcription | Therapeutic Gaming ▶

The operating system for clinical AIVoice · prediction · transcription · therapy — one infrastructure for any condition

01

swiv.

Part 01 / 03

The Problem.

Care that can’t scale, and the infrastructure gap behind it.

01 The Problem 02 The Platform 03 The Proof

swiv.

The scale of the problem

0M

people need therapeutic care.
Today, it reaches almost none of them.

swiv.

The problem

The Healthcare AI Infrastructure Gap.

970M+ people need therapeutic care, but access to clinicians is limited.

Generic chatbots are filling the gap, without any clinical awareness.

There is little healthcare AI infrastructure, making solutions expensive to deploy.

Point solutions don't scale, forcing a separate tool for each condition.

02

swiv.

Part 02 / 03

The Platform.

One clinical substrate — voice, models, and care, composed on demand.

01 The Problem 02 The Platform 03 The Proof

swiv.

So, what is Swiv?

Swiv is the clinical AI infrastructure.

Not an app. Not a chatbot. Not just companions.

The companion a patient falls in love with is only the surface. Underneath is the layer that makes any AI clinically aware, safe, and measurable, and turns any condition into real care.

swiv.

The platform · Six pillars

Six pillars. One platform.

Own the clinical substrate, and every new capability is a module that plugs in, not a new product to build from scratch.

Clinical Voice

Sub-three-second therapeutic speech with barge-in and live lip-sync, character-consistent and condition-aware.

Predictive & Risk Models

Validated scales scored passively; risk re-scored continuously — with device and wearable signals folded in as live context.

Care-Pathway Builder

Describe a condition in a sentence; the platform generates the pathway, guardrails, and approval gates.

Agentic Actions

Companions that do, not just say — schedule follow-ups, draft notes, fire interventions, escalate to a clinician. Every action risk-tiered.

Generative Therapy

Companions, games, and guided practices generated on demand, personalized to each patient in real time.

Human in the Loop

Routine steps run autonomously; anything clinical-grade waits for a clinician’s yes. Every action role-scoped, auditable, and reversible.

swiv.

The platform · Who it serves

One platform. Three perspectives.

The same clinical substrate, felt three different ways.

For the patient

Care that heals, and stays.

An always-on companion, therapeutic games, and guided practices that remember you and adapt in real time, not a static app you open once.

For the clinician

Hours back, and a second brain.

Decision support, documentation, and safety checks, with every patient signal between visits flowing into one clear dashboard.

For the health system

Any condition, live in minutes.

One EMR-ready, governed, and measurable platform, instead of buying a separate vendor for every condition.

swiv.

The platform · How it works

One signal path, from conversation to care.

Live conversation

voice · text · biometrics · wearable signals

↓

Clinical Integration Layer

The neural hub

injects patient history, clinical guidelines, and live context into everything below

↓

Transcription

every word, structured

Prediction & detection

risk & disease signals

Assessment

validated scales, auto-scored

Emotional intelligence

affect & response

Memory

three-tier, per patient

↓

Patient

Therapy that adapts in the moment, and remembers.

Clinician

Alerts, drafted notes, and decision support, prioritized.

Health system

EMR-ready records, analytics, and full auditability.

swiv.

Under the hood

Real models. Real machine learning. Not just prompts.

Every conversation is processed live by a full stack of models, not a single prompt behind a mask.

Voice layer

Streaming speech-to-text and text-to-speech, sub-second, with natural barge-in.

Live transcription

Patient and clinician conversations transcribed and structured in real time.

Predictive ML

Models re-score deterioration, relapse, no-show, and crisis after every session.

Disease detection

ML surfaces ADHD, ASD, anxiety, and depression signals from natural conversation.

Assessment scoring

PHQ-9, GAD-7, ADHD-RS-5 and more, scored automatically from language.

Emotional intelligence

Sentiment and emotion models drive how every companion responds.

Clinical documentation

Auto-generated summaries, notes, and treatment-plan suggestions.

3D generation

In-house rigging, animation, and morphing, 15M+ animations across the library.

swiv.

The product

The Swiv Solution.

Core Technology
Proprietary 3D companion engine · persistent memory & emotional intelligence · cross-platform deployment.

AI Companions
Real-time emotional response · continuous therapeutic relationships · one experience across devices.

Consumer
Condition-specific companions · clinically-grounded games · physician-linked insights.

Clinician
Decision support · safety & compliance checks · automated admin workflows.

Create your companion in seconds

Play clinically-grounded therapy

Care that stays with you

Smarter support for clinicians

swiv.

Pillar · Voice

Real-time therapeutic voice, felt as conversation.

A full voice pipeline, not a bolt-on. Streamed, character-consistent speech with live lip-sync, tuned so a patient builds a real relationship with the agent.

150ms

Wake-word

→

800ms

Recognition

→

1.2s

Clinical response

→

600ms

Synthesis

→

~2.8s

End-to-end

Real-time transcription Every word, patient and clinician, is transcribed and structured live, feeding assessment, prediction, and the clinical record.

Streaming & barge-in The patient can interrupt naturally; the agent yields, listens, and resumes with no awkward pause.

Hear the voice layer

Maya · English

Therapeutic check-in with a patient

0:20

بدر · Arabic

Follow-up session · Saudi dialect

0:19

swiv.

Pillar · Clinical intelligence

Every conversation is a clinical instrument.

Background analysis correlates conversational patterns with validated clinical indicators, passive, continuous measurement no worksheet can match.

ADHD-RS-5

BRIEF-2

PHQ-9

GAD-7

PSS-10

WFIRS-P

Assessment

Four developmental domains

Social Communication, Executive Function, Emotion Regulation, and Sensory Processing, aligned to DSM-5 and CDC criteria.

Prediction & detection

Machine learning, on every signal

ML models re-score deterioration, relapse, and crisis risk, and surface early disease-detection signals, each with a confidence band.

Intervention

Signal-triggered CBT tools

Thought records, behavioral activation, breathing, and attention training, fired on clinical signal, not a fixed schedule.

Automated crisis detection with escalation pathways runs on every session.

swiv.

Pillar · Generative infrastructure

Describe a condition. The platform builds the care.

What used to take years to build for a single diagnosis, the substrate now composes from reusable, protocol-compliant parts, a companion, an assessment loop, and the therapy to support it.

Describe in plain language

“A companion for anxiety with relapse prevention”

→

Compose the pathway

steps · conditions · clinical guardrails

→

Approval gate

clinician reviews before it goes live

→

Live in the patient's hands

safe, condition-specific, measurable

Natural-language authoring Describe an operation in a sentence and the builder assembles the modules, conditions, and gates, ready to run.

Visual flow editor Every generated pathway is a diagram a clinician can inspect, adjust, and version, no black box.

under 10 minutes

from a description to a clinically-aware companion, live and in a patient's hands, not in a roadmap.

swiv.

The platform · Any condition

One infrastructure layer. Condition-specific companions on top.

Each one composed by the Care-Pathway Builder on the same substrate — a new companion per condition, not a new product per condition.

Medication Adherence

Chronic Disease Monitoring

Physical Therapy

Appointment Prep

Mental Health Screening

Stroke Rehab

Elderly Care

ADHD & ASD Support — and counting

And across devices

Phone · tablet · smart mirror · wearables · embedded hardware — biometrics stream straight into the session.

A smart mirror, not a phone — the same companion engine, wearing a different job.

swiv.

From living room to clinic

A day on Swiv.

8:14 PM · At home

A session, not an app

The patient talks with their companion. Voice is streamed, transcribed, and understood live.

→

8:31 PM · Seconds later

The models go to work

Risk re-scored, instruments updated, emotional state mapped, all from one conversation.

→

Next morning · In clinic

One prioritized alert

The clinician sees what changed overnight, with the note already drafted.

→

Next session

Care adapts

The pathway re-tunes to the new signals, approved by the clinician, felt by the patient.

swiv.

The product · Live today

AI Companion Ecosystem, live today.

Guided therapeutic programs

30–60s quick-reset interventions

Voice-first companion, listening live

Companions in any form — playful or lifelike

The clinician’s view, same platform

03

swiv.

Part 03 / 03

The Proof.

Evidence, market position, traction, and the team behind it.

01 The Problem 02 The Platform 03 The Proof

swiv.

The proof · Clinical evidence

Clinical evidence validates the need for therapeutic AI infrastructure.

Directional figures from early studies and published literature; formal IRB validation in progress.

0%

ASD improvement in social & emotional abilities

0%

ADHD improvement above baseline scores

0%

Improved arm & hand movement recovery

0%

Anxiety & depression treatment completion

swiv.

The proof · Market position

Others ship applications. Swiv ships the infrastructure behind them.

$1B+ has gone into single-purpose apps, one condition, one modality each. Swiv is the layer they could all have been built on.

Single application

Full infrastructure

Any condition

One condition

Woebot Akili Cognoa AppliedVR MindMaze DeepWell Headspace Tolan Swiv

swiv.

The proof · Traction

De-risked technology. Proven demand. Ready to scale.

Core technology, built

Proprietary 3D + AI engine · agentic application framework · production-ready stack.

Product, working

Live companion & gaming MVP · clinical context adapts the experience in real time.

Market, validated

600+ active creators & users · high-engagement community · multi-channel acquisition working.

✔ Real demand confirmed✔ Hardest technical risks eliminated

swiv.

The team

A unified team leveraging multidisciplinary expertise.

Rami al Hendi

CTO

Multiple startup exits with deep technical expertise building scalable software.

Yousef Chahien

COO

Successful exits in tech with a strong engineering background and product leadership.

Zaid Chahien

CMO

Proven tech and AI leadership delivering enterprise solutions and driving innovation.

swiv.

The team · Advisors

Advisory conversations across psychology, neurology & digital health.

Dr. Nirav Shah

Stanford, CDC & CHAI, advising on AI and clinical excellence

Dr. Per Carlbring

Advancing behavioral therapy at leading academic programs

Dr. Eirini Karyotaki

Amsterdam VU & Harvard clinical psychology professor

Dr. Pedro Estrella

Bioelectronics spanning Cambridge and Bath

Dr. Chris Ralyea

Clinical leader in neurology and brain health

Mr. Ziad Jammal

GM, Google Cloud UAE, with AI leadership

swiv.

From our advisory conversations

“Your work on a clinical AI infrastructure platform sounds both timely and important. We will soon be publishing a randomized controlled trial evaluating a fully automated AI therapy for depression, and a new study on how a chatbot can assist a therapist in delivering internet-based CBT for social anxiety.”

Professor Per Carlbring
Department of Psychology, Stockholm University

swiv.

The road ahead

From foundation to market dominance.

1

2

3

Foundation

0 – 7 Months

Clinical Integration Layer live
Companions with persistent memory
Consumer & clinician apps launched

Growth

7 – 14 Months

Clinically-capable AI companions
Therapeutic gaming frameworks
Biometric & EMR integration

Dominance

14 Months +

Enterprise-grade clinical platform
AR/VR & advanced biometrics
Regulatory pathway & first enterprise customers

swiv.

Beyond healthcare

This is just the beginning.

The same substrate — voice, memory, emotion, safety — works wherever a person talks to a character.

Education Tutors that remember every lesson, and adapt the way companions adapt to patients.

Robotics & embedded hardware The companion gets a body — same brain, new shell, printed or built in.

AR & immersive experiences Already in the room today — these moments are real, not renders.

We built this for our own kids first.
Our co-founder’s daughters with their companion, at home. Real moments — not renders.

The moment care becomes scalable

Where intelligence meets compassion.

Build with Neuro swiv.ai · neurolab.ae