/
Navigation
Chronicles
Browse all articles
Explore
Semantic exploration
Research
Entity momentum
Nexus
Correlations & relationships
Story Arc
Topic evolution
Drift Map
Semantic trajectory animation
Posts
Analysis & commentary
Pulse API
Tech news intelligence API
Browse
Entities
Companies, people, products, technologies
Domains
Browse by publication source
Handles
Browse by social media handle
Detection
Concept Search
Semantic similarity search
High Impact Stories
Top coverage by position
Sentiment Analysis
Positive/negative coverage
Anomaly Detection
Unusual coverage patterns
Analysis
Rivalry Report
Compare two entities head-to-head
Semantic Pivots
Narrative discontinuities
Crisis Response
Event recovery patterns
Connected
Search: /
Command: ⌘K
Embeddings: large
TEXXR

Chronicles

The story behind the story

days · browse · Enter similar · o open

Research from Alphabet's Verily shows that blood pressure, age, whether someone smokes, and more can be deduced by analyzing scans of patients' retinas with AI

James Vincent / The Verge :

The Verge James Vincent

Context & Ripple Effects

The retina has been quietly turning into Alphabet's favorite diagnostic surface. DeepMind's five-year project on a million anonymous NHS eye scans established the data pipeline, and this Verily result shows why the effort was worth it: the same image that screens for eye disease also encodes blood pressure, age, and smoking status.

The finding lands ahead of the regulatory and payment infrastructure that would commercialize it — an approval path that arrived weeks later with the first FDA-cleared AI diagnostic requiring no doctor interpretation, and a reimbursement one by 2020 when CMS began paying doctors to use AI eye-disease diagnostics.

First-order effects

  • Clinics already photographing retinas for diabetic screening gain cardiovascular risk signals — blood pressure and smoking status — from images they were taking anyway, at no added capture cost.

Second-order effects

  • Payers and employers get a cheap, passive biomarker source, which pressures device makers and EHR vendors to treat every retinal camera as a multi-purpose sensor rather than an eye-disease tool.

Third-order effects

The trend: Retinal imaging is becoming a general-purpose health-data platform, with Alphabet building the scan pipelines, models, and regulatory precedents across Verily, DeepMind, and Google.

Discussion

  • @reza_zadeh Reza Zadeh on x
    Surprising new predictions made from computer vision on retinal images: - Distinguish smoker vs non-smoker 71% accurate - Predict heart-attack or stroke in 5 years 70% accurate Using attention mechanisms can show model focuses on blood vessels. Paper: http://www.nature.com/... ht…
  • @skathire Sek Kathiresan MD on x
    Quite impressive potential. But please note main result = predict *age* well by looking at retinal vessels. And age is by far dominant factor in CVD risk prediction models. http://twitter.com/...
  • @sundarpichai Sundar Pichai on x
    AI offers us the potential for new, less invasive tests for heart health - predicting cardiovascular results from retinal images with computer vision - encouraging early results! https://research.googleblog.com/ ...