ThinkLoop is a clinician-led healthcare AI consultancy. We help hospitals, health-tech startups, payers, and pharma move from raw data to deployed, clinically defensible AI — closing the loop between clinical judgment and machine intelligence, with the rigor of peer review and the discipline of production engineering.
The feedback loop is the core concept of medical AI done right. Our name carries the discipline into everything we build.
Clinical oversight is not a compliance checkbox — it is the design. Every model we ship keeps a clinician in the decision path.
PROBAST+AI and TRIPOD-aligned validation with honest performance metrics. No inflated claims, no silent leakage.
Models that stay in production: drift detection, monitoring, and runbooks. AI that ships — and stays shipped.
Deployment is where medicine begins. We design for the ward, the clinic, and the patient who comes back.
From first feasibility question to a monitored system in production — engagements are scoped, milestone-based, and always include validation as a named deliverable.
AI opportunity assessment, data-readiness audits, and regulatory path mapping (SFDA, FDA, EU AI Act). You get a strategy report and a realistic 12-month roadmap.
Model build with PROBAST+AI / TRIPOD-aligned validation, honest performance metrics, external validation design, and submission-grade documentation.
Deployment, monitoring, drift detection, CI/CD for models, and operations runbooks. The difference between a demo and a system that stays in production.
NHANES, GBD, and EMR analytics; dashboards; real-world evidence generation; epidemiological studies with manuscript-ready outputs.
Fractional Chief Data or AI Officer for health-tech startups and health systems — strategy plus hands-on execution, about 20–40 hours a month.
"Explaining ML to Clinicians," data science bootcamps, and validation-methods workshops for hospitals, universities, and health ministries.
A surgeon's discipline applied to machine learning: examine carefully, validate honestly, and only then operate.
Founded and run by an ENT surgeon and data scientist. We speak both languages fluently — the ward and the model card.
PROBAST+AI and TRIPOD discipline on every engagement. If we cannot validate it, we will tell you before you pay for it.
Models that die in notebooks help no one. We engineer for deployment, monitoring, and long-term clinical use.
Peer-review discipline in everything: reproducible pipelines, honest metrics, no pipeline comparisons, no inflated claims.
Dr Ameye is an ENT surgeon and data scientist with more than two decades in medicine and a focused second career in machine learning. He was a Reader (Associate Professor) in Otorhinolaryngology at Obafemi Awolowo University and its teaching hospital, and has published 26 peer-reviewed articles and two theses spanning clinical medicine and AI.
His research sits at the intersection that matters: audiology and hearing health, clinical fuzzy systems, and the WHO hearing-loss agenda — work that has produced a validated Fuzzy Audiometric Index and a research pipeline built on honest, reproducible validation. That is exactly the discipline he now brings to clients.
Based between the GCC and West Africa, with engagements across three continents, he leads ThinkLoop's projects personally — from the first data-readiness conversation to the deployed, monitored system.
Whether you are a hospital exploring AI feasibility, a startup that needs validated models, or a funder with an audiology agenda — the first conversation is free and it starts with the data question, not the demo.
Tell us about your data, your problem, and your timeline. We respond within two business days.
ThinkLoop