Singapore · Healthcare AI · Est. 2020

Clinical AI,
Built for Real Healthcare.

InsytAI helps healthcare and enterprise teams design, validate, govern, and deploy reliable AI systems — so your clinical teams can focus on what matters. The work is grounded in live clinical deployments, peer-reviewed publications, and Singapore health-system implementation experience.

See Engagement Models Discuss an AI Use Case
30 Publications 212 Scholar citations 8 h-index
14.5SGD M+
Funded Research
15+
Systems Live
30
Publications
2Clusters
Hospital Clusters
15+ yrs
Experience
14.5M+
SGD Research Grants
15+
Systems Live
30
Publications
2
Hospital Clusters
15+ yrs
Research Exp.
What We Do

End-to-End
AI Services

Every engagement led by a PhD researcher who has shipped AI into live hospital systems — from data strategy to clinical deployment.

01
AI Development
Custom ML and deep learning models — end-to-end from data pipeline to production on AWS or on-premise clinical infrastructure.
ML · DL · CV
02
Healthcare AI
Clinical decision support, medical imaging AI, predictive models from EMR data. Clinical workflows and patient safety, not just model accuracy.
Clinical · EMR
03
LLM & Generative AI
Fine-tuning, RAG pipelines, voice AI transcription, and intelligent chatbots — deployed at scale across Singapore's major hospital clusters.
LLM · RAG
04
Multi-Agent Platforms
Autonomous agentic workflows, multi-agent orchestration, and tool-calling systems for complex enterprise and clinical tasks.
Agents
05
Data Science Consulting
Strategic advisory — team building, project evaluation, data infrastructure design, and analytics roadmaps.
Strategy
06
AI Strategy & Governance
AI strategy papers and LLM governance frameworks — authored for major Singapore hospital cluster-wide adoption.
Policy
07
Research & Development
Academic-industry collaboration, grant support, and R&D in multimodal medical AI, clinical foundation models, and healthcare optimisation algorithms.
R&D
Why InsytAI

Clinical AI That
Actually Ships

Most AI consultants prototype. We deploy — into live hospital systems, with governance frameworks, ethics approvals, and clinical validation.

Production Deployed — Not Just POC15+ AI systems in live production across Singapore's major hospital clusters. Not demos — real clinical tools used by clinicians daily.
Published Science Behind Every System30 publications, 200+ Google Scholar citations, and peer-reviewed work in The Lancet Regional Health, Annals AMS, Electronics, LNCS, The Spine Journal, and IEEE EMBC.
Governance-First ApproachLed hospital cluster-wide AI Strategy Paper and LLM Governance Framework. We build AI that satisfies legal, ethics, and compliance requirements.
Multi-Modal Data ExpertiseEMR, radiology, genomics, proteomics, voice — the full clinical data spectrum, not just text.
$14.5MResearch Grants
15+Live Systems
30Publications
2Hospital Clusters
Selected Work

Live AI Systems

Production AI systems built across Singapore's major hospital clusters — from clinical LLMs to medical imaging and mixed-reality tools. Deployed and used by clinicians daily.

Loading projects...

View All Projects →

The Team Behind The Work

Research Depth.
Operational Reach.

InsytAI was founded by Mohammad Shaheryar Furqan, a clinical AI practitioner with hands-on deployment experience inside Singapore's major hospital clusters. The work is grounded in peer-reviewed science, governed implementation, and production-proven systems, not prototypes.

2024–
Principal Data Scientist — Singapore National Health ClusterDigital Innovation Office. AI Strategy, LLM Governance, end-to-end clinical deployment.
2021–24
Senior Research Fellow — Singapore Medical SchoolClinical informatics. Teaching, research, clinical AI systems across major hospital clusters.
2021–24
Lead Data Scientist — Singapore University Hospital SystemData science team lead. Co-designed production healthcare AI platform used across clinical departments.
2017
PhD — Top Singapore UniversityBioinformatics, biological network prediction, machine learning.
Healthcare AI Leadership Award 2024
Medical School Mentor Award 2022
IEEE Senior Member
2× Gold Medals
$8M
Multi-Modal AI Platform
$3.97M
Clinical Decision Support
$1.98M
Surgical AI System
$14.5M
Total
The Team

Founder-Led.
Partner-Supported.

InsytAI operates through a focused founder-led model, supported by a trusted network of AI engineers, clinical collaborators, research partners, and implementation specialists across Singapore and the region.

F
Mohammad Shaheryar Furqan
Founder · Principal Data Scientist
PhD. Clinical AI researcher and principal data scientist with published work across clinical LLMs, medical imaging, mixed reality, and healthcare AI platforms in Singapore.
Full Profile → Publications →
Clinical Collaborators

InsytAI works with a network of clinical co-investigators, surgeons, radiologists, and medical informaticians across Singapore's hospital clusters.

Enquire About Collaboration →
Technical Partners

Engagements are supported by specialist AI engineers, MLOps practitioners, and LLM deployment experts experienced in clinical-grade production environments.

Start a Conversation →

Founder Experience Across Singapore's Leading Institutions

Process

How We Work

A proven four-phase process — from first conversation to live clinical deployment.

01
Discovery & Feasibility
Define problem, data availability, workflow fit, risks, and success metrics. Typically 1–2 weeks. Outcome: a clear build-or-not recommendation.
02
Prototype & Validation
Build model or agent workflow, test against real-world data, evaluate safety and clinical usability. Outcome: validated proof-of-value with performance benchmarks.
03
Governance & Deployment
Add monitoring, audit logs, model documentation, ethics review, and production integration. Outcome: a governed, traceable system ready for clinical use.
04
Operate & Improve
Monitor drift, usage, safety events, and user feedback. Continuous improvement loop. Outcome: a system that gets better over time, not worse.
Engagement Models

How to Engage

Fixed-scope engagements designed for healthcare and enterprise AI timelines.

Discuss an AI Use Case →
2–4 Weeks
AI Strategy Sprint
For leaders deciding what to build, buy, or govern. Problem scoping, data audit, feasibility assessment, and prioritised AI roadmap.
Enquire →
6–10 Weeks
Clinical AI Prototype
For teams validating a high-value AI use case. Model development, clinical data evaluation, safety assessment, and proof-of-value report.
Enquire →
8–12 Weeks
LLM / RAG Deployment
For secure knowledge assistants, clinical documentation tools, and workflow agents. Fine-tuning, RAG architecture, governance, and production deployment.
Enquire →
Ongoing
AI Governance & Review
For healthcare teams needing risk assessment, LLM governance frameworks, validation protocols, and monitoring strategies for existing or planned AI systems.
Enquire →
Research & Press

Published & Recognised

The Straits Times
Supercomputer at NUHS powers AI projects that save time and give patients better care

SMILE AI programme featured as part of Singapore's national supercomputing initiative.

2023
Zaobao 联合早报
AI helps doctors quickly organise medical records, expected to shorten waiting time

CalSense project — AI-assisted clinical record management reducing patient waiting times.

2023
Dental Asia Magazine
SMILE AI: Can AI help a nation prevent oral health diseases?

International feature on SMILE AI's impact on population-level oral health screening.

2023
The Lancet — Western Pacific
Integration of customised LLM for discharge summary generation in real-world clinical settings: RUSSELL GPT
Chua, Clara et al. — 2024
The Spine Journal
MRI spine request form enhancement and auto protocoling using a secure institutional LLM
Hallinan, Leow et al. — 2025
Annals AMS
Automated Cobb angle measurement in scoliosis radiographs: A deep learning approach for screening
Low, Makmur et al. — 2024
LNCS / ASMUS
HoloPOCUS: Portable Mixed-Reality 3D Ultrasound Tracking, Reconstruction and Overlay
Ng, Gao, Furqan et al. — 2023
Electronics
HoloVein — Mixed-Reality Venipuncture Aid via CNN and Semi-Supervised Learning
Ng, Furqan, Gao et al. — 2023
CMPB
Momentary dietary lapse prediction for obesity management: eBLISS and ML prediction model
Chew, Shridhar, Furqan et al. — 2025
Enterprise-Grade

Built for Real-World Deployment

Every InsytAI engagement is scoped, governed, and delivered to production standards — not proof-of-concept standards. Healthcare AI that can't be audited, monitored, or explained isn't safe to ship.

Governance
Data governance, access control, and audit trails on every inference
Privacy & Compliance
PDPA-compliant. Supports on-premises, private cloud, or hybrid deployment. No patient data leaves institutional boundaries without explicit consent.
Human Oversight
Human-in-the-loop review built into every clinical workflow. Frictionless clinician override. Zero set-and-forget deployments.
Data
Governance & access control documented before any model training begins
Workflow
Clinical workflow mapping completed before deployment scoping
Validation
Model validation against real-world institutional data, not benchmark datasets
Audit
Full inference logging with model version, input, output, and clinician action
Infrastructure
Cloud, on-prem, or hybrid. Containerised for portability across hospital IT environments
Documentation
Ethics review, compliance brief, and leadership-ready summary for every deployment
Insights

AI Research & Intelligence

All Posts →
Loading...
Let's Build

Ready to deploy real AI?

PhD-level expertise for your most complex data and AI challenges. Hospital, healthtech, or enterprise — we ship.

LocationSingapore
Emailinfo@insytai.com
ServicesHealthcare AI · LLMs · Multi-Agent · Data Science · Governance
ExperienceMajor Singapore hospital clusters · MOH-funded · A*STAR-linked projects

By submitting, you agree InsytAI may contact you regarding your enquiry. We do not share your information with third parties. Privacy Policy