Initial results from an ECOG-ACRIN and Caris Life Sciences collaboration show AI-driven multimodal models can more accurately predict recurrence risk in early-stage breast cancer.
Key Details
- 1Collaboration leverages ECOG-ACRIN's clinical trial data and Caris' molecular profiling and imaging AI platforms.
- 2Over 4,000 TAILORx trial patient cases used to train and validate new multimodal deep learning models.
- 3Models integrate histopathologic slide imaging, clinical, and expanded gene expression data.
- 4AI models outperformed established recurrence risk assessment methods, especially for late recurrence (after 5 years).
- 5Potential shown for a scalable, cost-effective diagnostic test based on routine histology and clinical data rather than solely on genomic assays.
- 6Results presented at the 2023 San Antonio Breast Cancer Symposium.
Why It Matters
The study demonstrates unprecedented large-scale integration of imaging, molecular, and clinical data for recurrence prognostication, charting a path toward more personalized and precise breast cancer management. Imaging AI's role in such multimodal frameworks signifies a promising advance for the development of new diagnostic tools in oncologic pathology.
Source
EurekAlert
Related News
•EurekAlert
ML and Multimodal Imaging Power Cerebral Blood Flow Monitoring for Spaceflight
Researchers developed a machine learning model that uses ultrasound and MRI data to predict cerebral blood flow in simulated microgravity for astronaut health.
•EurekAlert
Deep Learning Model Predicts Language Outcomes After Cochlear Implants Using MRI
AI model using deep transfer learning accurately predicts spoken language outcomes in deaf children after cochlear implantation based on pre-implantation brain MRI scans.
•EurekAlert
LSTM Deep Learning Enhances Optical Sensing for Biochemical and Medical Applications
Researchers have developed an LSTM-driven interferometric sensing system that achieves both high sensitivity and wide measurement range, overcoming previous trade-offs in optical sensing.