Transforming Cancer Care with AI Precision Medicine
Empowering physicians with advanced multi-modal tools to improve treatment selection and patient outcomes.
AI Precision Medicine
Ataraxis is developing pioneering artificial intelligence methods to integrate information from multiple digital data modalities. Through cutting-edge artificial intelligence, Ataraxis systems discover novel clinical, morphological, and molecular features of patients and their disease, which capture the heterogeneity of cancer beyond a single data type.
Our proprietary technology analyzes multi-modal patient data to assess the aggressiveness of the disease, predict risk of cancer recurrence or death, and how various treatment options can benefit different patients, for example in breast cancer through Ataraxis Breast.
All Ataraxis tests are based on AI foundation models, such as Kestrel, trained with hundreds-of-millions of pan-cancer pathology slide images, that extract novel information about tumor morphology whivch strongly correlate with patient outcomes and treatment response.
Unlike traditional molecular genomic tests, we deliver results of our analyses within 24 hours from the receipt of patient data. Additionally, physical tissue is not exhausted and no additional histology preparation is necessary.
Landmark Clinical Validation Study
Ataraxis Breast, the first AI-native prognostic/predictive test for breast cancer, has been developed and clinically validated using data of 8,161 patients from 15 institutions. Ataraxis Breast was shown to identify low- and high-risk patients more accurately than standard of care genomic assays, and it was prognostic in all breast cancer patients, including underserved groups.
In an analysis of 858 patients from three external validation cohorts, Ataraxis Breast outperformed standard-of-care 21-gene assay reducing prediction error by approximately 50%.
In an external cohort of 1,733 patients, patients with high Ataraxis score experienced breast cancer recurrence 4 times more often than low Ataraxis score patients after 5 years of follow up.
Latest News
Browse latest news and achievements from our team, updated weekly and exclusively.