Go Beyond What You See™
iBiopsy® reveals novel insights into previously unreachable knowledge. The iBiopsy® platform exploits multimodal Artificial Intelligence technologies and Data Science to meet the promise of precision medicine through cutting edge medical imaging analysis to improve patient care and treatment efficacy.
Harnessing the true power of medical images is key to accelerate therapeutic innovations and foster precision medicine. iBiopsy® extracts the utmost meaningful visual signatures from standard medical images (CT, MRI etc.) allowing learning and knowledge discovery through cutting edge Artificial Intelligence approaches.
Medical imaging reveals the disease through multiple signals, at every stage. Performed non-invasively for the benefit of patients, mining these signals allows for the most accurate detection, monitoring, diagnosis and prediction. Our strategy is to extract proprietary statistical image signatures with a lesion-agnostic paradigm, considering the organ as a whole. This approach enhances tissue coverage, which provides more detailed information at the pixel level, and improves image interpretability when compared to a standard lesion segmentation approach:
- Advanced proprietary multimodal AI algorithms
- Enrichment of image description capabilities beyond traditional radiomics
- No prior lesion segmentation
- Image signatures coverage of the whole organ
- Full web DICOM compliant environment
Median’s iBiopsy® platform supports translational research through AI powered imaging biomarker identification, and informs clinical decisions using CT Scan similarity search and analytics to compare with real-world evidence for patient profiling.
Median Technologies has developed the iBiopsy® platform to:
- Empower translational research with AI powered Biomarker identification tools, and
- Inform clinical decisions using CT scan Similarity Search for patient profiling services.
Multiple-modal AI for Biomarkers identification
iBiopsy® combines multiple AI approaches allowing high risk patient prediction and identification. An image signature mining approach enables the description and modeling of tumor host environment, tissue heterogeneity and intratumor heterogeneity together with the tumor micro-environment (TME) quantification.
iBiopsy® considers the organ as a whole, not just a subsection, and does not require prior segmentation. Our proprietary platform analyses tissue heterogeneity to reveal patterns/signatures to that are foundational to biomarker identification. Acting as surrogate endpoints, these signatures are particularly useful to:
- Predict response to therapy
- Stratify patients
- Monitor early drug response
- Call for further objective analysis & quantifiable measurements prior to formal FDA approval as Biomarkers.
Comparison with Real-World Evidence
A similarity search engine allows for the comparison of a patient’s CT to the CT Scans (and EHR) of other patients, identifying a population with similar phenotypes. Clinical records of this population are then analyzed to perform a statistical profiling of the patient, revealing not only odds of clinical outcome, but also different likelihood rates across multiple clinical indicators (e.g. clinical and histological features, treatment, outcome…). This is particularly helpful to inform clinical decisions to adapt the course of treatments and facilitate patient profiling.
Technology Application Portfolio
- Early detection and characterization of disease
- Drug Response monitoring
- High risk patient stratification
- Informed clinical decisions based on real-world evidence
iBiopsy® Biomarker identification is performed through strategic partnerships with clinical and pharmaceutical stakeholders (ex: joint laboratories). It is our hope to identify biomarker signatures across multiple pathologies. To that end, Median Technologies is developing strategic collaborations worldwide, tailored around mutual translational research agendas.
iBiopsy® is currently focusing on Liver Diseases (HCC, NASH)