Diagnostics & imaging AI · Breast cancer screening AI
Therapixel
Therapixel is a French company that makes MammoScreen, FDA-cleared AI that flags suspicious findings on screening mammograms and helps radiologists draft the report.
Strong fit
- Breast imaging centers that read 2D and 3D screening mammograms
- Radiology groups on GE or Hologic mammography machines
- Teams that want AI to pre-fill the first draft of each report
Weak fit
- Radiology groups that want one AI vendor for many body areas
- Buyers who need a posted price
- Centers that want breast ultrasound or MRI AI
Bottom line
Therapixel's MammoScreen is a strong choice for breast imaging teams that want FDA-cleared AI to flag findings on 2D and 3D mammograms and pre-fill reports. It suits US centers on GE or Hologic machines best. Expect to negotiate the price, since none is posted.
Score breakdown
Weights: Outcomes 35% · Product 30% · Implementation 20% · Pricing clarity 15%.
Olivier Clatz and Pierre Fillard, researchers at INRIA, France's national institute for computer science research, founded Therapixel in 2013. Its software team works in Nice and its research team in Paris. Matthieu Leclerc-Chalvet is chief executive.
Therapixel first sold Fluid, a touch-free image viewer for operating rooms. In 2017 it won the DREAM Digital Mammography Challenge, a contest among nearly 1,200 research teams, and in 2020 it sold its surgical unit to focus on radiology.
MammoScreen reads 2D mammograms and 3D tomosynthesis images from GE and Hologic machines. It marks soft tissue lesions and calcifications, gives each finding a suspicion score from 1 to 10, and can use prior exams and pre-fill an initial report. The radiologist makes every final call.
The FDA first cleared MammoScreen in 2020 after a reader study found radiologists detected cancers better with it than without it. It cleared MammoScreen 5 in June 2026. Mayo Clinic uses it across its sites, Onsite Women's Health uses it in its US mammography network, and Hologic distributes it in the US.
Therapixel says 25% to 30% of screening mammograms in France are read with AI help, and about 60% of those use MammoScreen. The company has about 50 staff, most of them in France. It is working with Institut Curie on using mammograms to predict a woman's cancer risk.
ScreenPoint Medical fits breast imaging teams that want another widely studied mammography AI. Koios Medical fits teams that want AI for breast and thyroid ultrasound, and Aidoc fits radiology groups that want one AI platform across many conditions.
Competitor landscape
| Vendor | Overall | Ease of implementation |
|---|---|---|
| ScreenPoint Medical | 7.9 | 7.6 |
| Aidoc | 7.9 | 7.5 |
| Therapixel | 7.5 | 7.2 |
| Koios Medical | 6.7 | 6.3 |
Pricing
| Item | Detail |
|---|---|
| Model | Therapixel sells MammoScreen by contract, directly or through partners such as Hologic. |
| What usually drives cost | For Therapixel, cost depends on study volume, whether you read 2D or 3D exams, and whether you buy through a distributor. |
| What to ask before you buy | Ask Therapixel for its price per study or per site, which features are FDA cleared for your machines, and how results reach your reading worklist. |
| Published pricing | Therapixel does not publish a price list. |
Methodology
| Weight | Factor | What it measures |
|---|---|---|
| 35% | Customer outcomes | Whether buyers get measurable operational or clinical-workflow results after go-live |
| 30% | Product | Capability depth, reliability, and fit for the job the category buys |
| 20% | Implementation | How hard it is to stand up, integrate, train, and stabilize |
| 15% | Pricing clarity | Whether a buyer can model total cost without a mystery quote |
| Label | Meaning |
|---|---|
| Highly recommend | Strong outcomes and product with manageable caveats |
| Recommend | Solid fit for the right buyer; know the tradeoffs |
| Conditional | Only with a specific use case or heavy caveats |
| Not recommended | Avoid for most buyers in this category |
Read our full methodology for how we weight scores and assign recommend labels.