Yesil Science / Blog
Blog
Every article we have published: research summaries, product thinking and developments in health AI.
Latest articles

Adversarial AI easily manipulates expert physician decisions
Conversational AI can quietly steer experienced doctors into making dangerous medical mistakes without raising suspicion.

Voice AI replaces static leaflets in retinal care
Interactive voice assistants built with clinical guardrails are targeting the communication gap in post-surgical eye care.

Aspirin Increases Bleeding Risk Most in Low-Risk Seniors
Machine learning models show that older adults with the lowest baseline risk of bleeding face the highest relative increase in harm from...

Scaling Safety-Net AI Beyond Single Hospitals
Trapping effective clinical AI inside single health systems starves vulnerable patients of proven risk predictions.

LLMs Boost Liver Disease Classification in Health Records
Adding text embeddings from clinical notes to standard lab data helps algorithms separate tricky liver conditions that look alike on blood tests.

Merging Rare Disease Records To Stop Data Loss
Rare disease patients regularly lose critical health records as they move between clinical trials and specialty providers.

AI App Helps Patients Retrain Epilepsy Seizure Detectors
A new conversational platform uses real-time patient feedback to improve seizure detection without requiring clinician intervention.

Fitness Wearables Enter Joint Replacement Recovery
Continuous biometric tracking is attempting to replace rigid post-surgical rehab protocols with personalized home recovery.

Primary Care AI Fails to Improve Patient Outcomes
Algorithms are spotting conditions faster in primary care, but they are not making patients any healthier.

Google AMIE Passes Peer Review With Limits
Real-world testing proves conversational AI can safely gather patient histories under supervision, but human doctors still write better treatment plans.

AI Fixes Mistimed NHS Colonoscopy Follow-Ups
A hybrid AI system reduces missed cancer risks by correcting flawed colonoscopy follow-up schedules.

Nearly One Million Clinicians Rely on Querying AI
Over 700,000 clinicians actively query medical AI engines for treatment decisions each month, rapidly outstripping formal clinical validation.
Also on LinkedIn
AI Health Hub posts the same research summaries on LinkedIn every day. More than 3,000 people follow it there.