OverRad AI gives radiologists superpowers

The best physicians use OverRad's prelim reports to deliver more accurate studies, replacing burnout with capacity.
How it works

Legacy platforms rely on binary classification. We built for pattern comprehension.

OverRad AI uses a novel convolutional neural network design to approach diagnosis like a physician, not a medical student with flashcards. The most critical observations are highlighted, and an experienced partner helps you deliver faster, more accurate reports.

Our Mission

Unlock Capacity to Enable Providers to Deliver Better, Faster Care More Accurately

Reduce Noise
More Support, Less Burnout
Replace Endurance with Capacity
Reduce Noise
More Support, Less Burnout
Replace Endurance with Capacity
Reduce Noise
More Support, Less Burnout
Replace Endurance with Capacity
Reduce Noise
More Support, Less Burnout
Replace Endurance with Capacity
Reduce Noise
More Support, Less Burnout
Replace Endurance with Capacity
Reduce Noise
More Support, Less Burnout
Replace Endurance with Capacity
THE DIFFERENCE

Unlike other models, OverRad AI is constantly learning and improving

Traditional classification-based models only capture part of the story and are limited in the kinds of pathologies they can correctly identify.

That’s why our models aim for >97% accuracy, speeding up radiologists’ workflows while improving accuracy and detection rates.

Meet the Team

Co-Founder · CEO

Jonathan Hastings

Career spanning Jet, Rent the Runway, Squarespace. VP Product at Salvo Health + Vale Health. 5 years in healthtech. Princeton, Public Policy & International Affairs.

Co-Founder · CMO · President

Matthew Knouse, M.D.

Practicing radiologist. Chief Medical Officer of Telerapp (teleradiology group, 12 countries). Advisor, Isono Health. Temple/Cooper residency. Licensed in 20+ states.

CONTACT

Let's talk.

Health systems, radiology groups, and investors — reach out and we'll be in touch within one business day.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.