Representation Learning
Pierwszy klinicznie zwalidowany test prognostyczny/predykcyjny, zbudowany od podstaw z wykorzystaniem sztucznej inteligencji, dla pacjentek z inwazyjnym rakiem piersi. Ataraxis Breast wykorzystuje multimodalne dane kliniczne, w tym slajdy ze standardowych preparatów histopatologicznych, aby wygenerować personalizowany profil ryzyka pacjentów, i pomóc podjąć decyzję na temat leczenia we wszystkich podtypach raka piersi.
State-of-the-art foundation models for digital pathology
Utilizing novel optimization strategies and a growing dataset of 160,000 whole slide images (2 billion image tiles) across 25 organs, Ataraxis™ Falcon foundation model outperformed the strongest public histopathology models, including GigaPath and UNI-2, across 24 pan-cancer tasks.
Built to generalize across diseases and clinical contexts
Our models learn rich representations by combining standard H&E pathology with routine clinical variables, discovering the most predictive signals on their own rather than being told what to look for. That shared foundation adapts to new clinical questions over time, enabling a growing portfolio of tests that deliver calibrated, patient-level predictions for real clinical decisions: recurrence risk, treatment benefit, therapy response, and biomarker discovery.
.png)
.png)
.png)
.png)
.png)
Three architectural choices make this possible:
Large-scale, pan-cancer models trained on unlabeled whole slide images learn reusable representations for downstream tasks, reducing reliance on scarce outcome-labeled data.
Models are fine-tuned specifically for each task, ensuring learned features are directly linked to underlying patient biology and long-term prognosis rather than intermediate proxies.
By combining the interpretability of survival analysis with the representational power of foundation models, we transform prognosis from a fixed snapshot into a dynamic, patient-specific prediction that adapts to individual biology and clinical context.