# ProFound AI (iCAD)

Source: https://onco.cc/drugs/icad-profound-ai/  
OnCo record `icad-profound-ai` (Treatment). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

The first AI cleared for 3D mammography in the US, now part of RadNet's imaging network.

## Summary

iCAD's ProFound AI was cleared by the FDA in December 2018 as the first artificial-intelligence software for digital breast tomosynthesis, and later versions cover 2D mammography, breast density and risk. The company's reader study reported faster reading and improved sensitivity for radiologists using the software. iCAD, based in Nashua, New Hampshire, agreed in 2025 to be acquired by RadNet, which also owns DeepHealth (and, through it, Kheiron), consolidating three of the leading mammography AI products under one imaging chain. ProFound is installed at thousands of sites in the US and Europe.

## Fields

- Kind: Treatment
- Status: approved
- Last checked: 2026-09-10
- Tags: test
- Brand: ProFound Detection
- Modality: AI mammography and tomosynthesis detection software
- Mechanism: Deep-learning detection across the slices of a tomosynthesis exam or a 2D mammogram, with case and lesion scores for radiologists.
- Approvals: US 2018: 510(k) clearance: AI detection for digital breast tomosynthesis (first for 3D mammography); EU 2019: CE mark

## Notes

- What a result means: the software supports the radiologist reading your 3D mammogram; you are unlikely to be told it was used.

## Sources

- iCAD: ProFound: https://www.icadmed.com

## Connected records

- cancers: [HR-positive / HER2-negative breast cancer](https://onco.cc/cancers/breast-hr-positive/), [Triple-negative breast cancer (TNBC)](https://onco.cc/cancers/tnbc/)
- technologies: [AI in radiology](https://onco.cc/technologies/radiology-ai-screening/), [Mammography & tomosynthesis](https://onco.cc/technologies/mammography/)
- companies: [iCAD (RadNet)](https://onco.cc/companies/icad/)

---
JSON: https://onco.cc/api/v1/entities/icad-profound-ai.json