SOPHiA DDM for Radiomics

SOPHiA DDM™ for Radiomics is an AI-powered radiomics analytics platform developed by SOPHiA GENETICS, designed to unlock entirely novel insights from radiology images within existing clinical and research workflows. The platform is compatible with any 3D imaging modality — including CT, MRI, and PET scans — across any equipment type and cancer indication, making it a versatile solution for oncologists, radiologists, and researchers engaged in precision medicine initiatives.

By combining expert-driven visualization, automated segmentation, and robust feature extraction in a single user-friendly interface, SOPHiA DDM™ for Radiomics enables research teams to move efficiently from raw imaging data to actionable insights. The platform supports multicenter, collaborative research environments, empowering institutions globally to contribute to and benefit from shared multimodal data.

Core Capabilities

• Streamlined 3D segmentation: Automatically or semi-automatically segment anatomical areas of interest in seconds using proprietary algorithms, reducing manual effort and improving consistency.
• Reliable radiomic feature extraction: Leverage validated, AI-powered methods for automated extraction of radiomic features from segmented regions, ensuring reproducibility and analytical confidence.
• Advanced image visualization: Visualize 3D images from multiple angles simultaneously through an intuitive interface, facilitating thorough review and interpretation of complex imaging data.
• Collaborative multicenter environment: Participate in multicenter studies across a global network using a shared multimodal research database accessible by anyone, anywhere.
• Longitudinal and multimodal analysis: Analyze radiomic features and integrate data into multimodal cohorts and longitudinal analyses to track disease progression and treatment response over time.

Workflow: From Images to Insights

1. Data upload: Import 3D images such as CT, MRI, and PET scans into a project-specific database and research environment.
2. Data visualization: Visualize imported images in 3D from multiple angles simultaneously using the platform's user-friendly interface.
3. 3D segmentation: Automatically or semi-automatically segment anatomical areas of interest using proprietary algorithms.
4. Feature extraction: Apply robust AI-powered algorithms to confidently extract radiomic features from segments and store them in the research environment.
5. Analysis: Analyze extracted radiomic features and incorporate data into multimodal cohorts and longitudinal studies.

Clinical Research Applications

• Metastatic non-small cell lung cancer (mNSCLC): A machine learning model developed for the TRIDENT post hoc analysis of AstraZeneca's POSEIDON trial used radiomics as a key contributor to identify multimodal signatures predicting patient benefit from adding tremelimumab to durvalumab and chemotherapy.
• Triple-negative breast cancer (TNBC): A machine learning algorithm applied to baseline multimodal data was developed to predict complete pathological response to neoadjuvant chemotherapy, with radiomics serving as one of the key contributors to predictive power.
• Solid renal tumors: In the UroCCR-75 study, a machine learning model incorporating clinical, radiologic, and radiomics features from pre-operative multiphasic contrast-enhanced CT scans was developed to differentiate benign from malignant solid renal tumors.
• Meningioma: Spatial mechanistic modeling leveraging radiomics was used to predict growth of benign asymptomatic meningiomas from only two MRI examinations, outperforming naïve linear regression in predicting changes in tumor volume and shape.
• Colorectal liver metastases: A post hoc evaluation of the TRIBE2 study found that radiomics may be a valid tool for predicting prognosis in liver-limited metastatic colorectal cancer patients and for stratifying risk of relapse after curative resection.

Key Benefits for Precision Medicine

• Improve accuracy by utilizing robust AI-powered algorithms to derive deep insights from complex imaging data.
• Expedite turnaround times through automated processing and streamlined workflows.
• Accelerate the adoption of precision medicine across institutions and research networks.
• Optimize resources by reducing manual analytical burden and enabling scalable, reproducible research.

SOPHiA DDM™ for Radiomics is intended for research use and integrates within the broader SOPHiA DDM™ platform ecosystem, which also includes modules for genomics and multimodal data analysis. The platform supports a global collaborative network, enabling multicenter studies and contributing to the democratization of data-driven medicine.