RISMO
Turn pixels into predictions and see what others miss. Transform small patient cohorts into breakthrough insights and unlock the next frontier in precision medicine and biomarker discovery. RISMO stands for Rapid Image Segmentation and Modeling of Outcomes. It builds models to estimate risk, predict drug responses, or highlight mutation burdens. RISMO also helps you to segment tumor regions, cellular boundaries, inflammatory regions, or collagen fibers.
Rapid Image Segmentation
Build custom segmentation models for your cohort using our few-shot learning approach. Based on just a few representative examples of the target features, the model reliably learns to recognize similar patterns across the entire dataset. Cover many applications, including quantifying tumor areas, measuring biomarker expression, and comparing tissue characteristics between patient groups.
Example Study: Overcoming data scarcity in biomedical imaging with a foundational multi-task model
Modeling of Outcomes – learn from patient-level labels
Analyze patient data based on your clinical labels to stratify patients and identify relevant patterns. Regions associated with outcomes or morphological features are highlighted to gain a better understanding of the underlying disease mechanisms.
Example Study: Whole Slide Concepts: A Supervised Foundation Model For Pathological Images
Founded in science enabled through expertise
The underlying algorithm is our award-winning foundation model1,2 that was ranked first in the UNICORN challenge.
Fraunhofer MEVIS’ expertise in quality management according to EN ISO 13485 and IEC 62304 allows the application in clinical studies and further development towards clinical use.
[1] Overcoming data scarcity in biomedical imaging with a foundational multi-task model
[2] Tissue concepts: Supervised foundation models for computational pathology
Partner with RISMO
RISMO is under active research and development by Fraunhofer MEVIS. We’re seeking research partners with specific analysis challenges and access to clinical datasets. Whether you’re testing a novel hypothesis, developing diagnostic workflows, or creating laboratory-developed tests, we’d welcome the opportunity to explore how RISMO could accelerate your research goals.
Get in Touch
The images shown here are in whole or part based upon data generated by the TCGA Research Network: https://www.cancer.gov/tcga.