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

Instant Records Access Feeds Payer AI Agents
Patient-permissioned data sharing bypasses months of legacy payer exchanges to feed clinical AI on day one of coverage.

Open AI Models Match Commercial LLMs on Notes
Open-weight AI models can extract clinical comorbidities from medical text just as accurately as proprietary systems like GPT-4 and Claude.

Digital CBT Cuts Insomnia in Cancer Survivors
A tailored sleep app doubled insomnia remission rates for cancer survivors compared to basic sleep education.

AI CTA Analysis Predicts Mortality Post-TAVR
Automated tissue analysis on CTA scans predicts post-procedure mortality in minutes, addressing a major bottleneck in cardiac risk assessment.

Colorectal Cancer ICU Risk Model Fails External Validation
A machine learning model built to predict 28-day ICU mortality for severe colorectal cancer patients saw its predictive power plummet when tested...

Abu Dhabi Deploys AI Assistant Across National Records
Abu Dhabi has embedded a generative AI physician assistant directly into its centralized health information exchange.

High AUC hides calibration flaws in sleep AI
A machine learning tool accurately spots cognitive risk in sleep apnea patients, but faulty probability scoring proves why strong discrimination alone cannot...

Machine Learning Platform Optimizes Artificial MicroRNA Targeted Gene Silencing
Machine learning-guided design addresses complex sequence-structure relationships to optimize artificial microRNA efficacy and specificity for targeted gene silencing.

AI Predicts Hypertensive Crises Three Days Early
Continuous biometric tracking can alert clinicians to dangerous blood pressure spikes long before patients land in the emergency room.

AI Stratifies Tuberculosis Risk Without Costly Biomarkers
Clinicians can screen latent tuberculosis using epidemiological data instead of expensive blood work, but algorithmic risk scores cannot yet replace definitive diagnostic...

FDA Clears AI to Fix Wearable ECG Flaws
Consumer smartwatches generate millions of heart readings every day, but clinicians waste valuable hours sorting through uninterpretable noise.

CT Radiomics Classifies Vertebral Compression Fractures Accurately
A new CT-based machine learning model accurately differentiates fresh from old vertebral compression fractures, boosting junior radiologist performance without requiring MRI scans.
Also on LinkedIn
AI Health Hub posts the same research summaries on LinkedIn every day. More than 3,000 people follow it there.