We combine cutting-edge machine learning and biology to identify novel biomarkers.

Biomarkers

arrow-desktoparrow-mobile

Interpretable AI

Our interpretable AI allows us to standardize the detection of existing biomarkers. We can also identify novel biomarkers by analyzing multimodal data to realize the full potential of precision medicine.

Multimodal

to capture the full complex picture of the disease.

Interpretable

to extract key biological insights from heat maps.

Clinically validated

to ensure clinical utility by leaders in the field.

Multimodal

to capture the full complex picture of the disease.

Interpretable

to extract key biological insights from heat maps.

Clinically validated

to ensure clinical utility by leaders in the field.

Our approach

We pioneer histogenomics. A novel approach that trains machine learning models on histology and genomic data to detect genomic alterations from routine digital pathology images. These biomarkers can be used in drug target identification, clinical trial optimization or be developed into diagnostic tools that fit directly into the pathologist’s workflow to identify the right treatment for each patient.

Contact us

Why Owkin?

Network of 50+ hospitals, 3 federated networks, 8 consortia, access to 45 live datasets, high dimensional data with longitudinal follow-up

We have the largest European data network focused on oncology and supported by our unique federated learning technology. This allows us to train our models on high quality, expertly curated, and fit-for-AI datasets.

Radiology images, clinical data, EHRs, histology and genomic data

Owkin goes beyond just lab and omics data by utilizing multimodality for a precise and complete picture of the patient.

Leader in federated learning, 30 Engineers who have produced over 70 models, 68+ KOLs (44 of which have a H-index > 40)

Established expertise across medicine, machine learning and engineering. Collaboration with top KOLs and institutions for critical understanding.

24 publications in top academic journals

A proven track record in leveraging deep learning multimodal models to identify new biomarkers for leading biopharma companies and academic research labs.

Network of 50+ hospitals, 3 federated networks, 8 consortia, access to 45 live datasets, high dimensional data with longitudinal follow-up

We have the largest European data network focused on oncology and supported by our unique federated learning technology. This allows us to train our models on high quality, expertly curated, and fit-for-AI datasets.

Radiology images, clinical data, EHRs, histology and genomic data

Owkin goes beyond just lab and omics data by utilizing multimodality for a precise and complete picture of the patient.

Leader in federated learning, 30 Engineers who have produced over 70 models, 68+ KOLs (44 of which have a H-index > 40)

Established expertise across medicine, machine learning and engineering. Collaboration with top KOLs and institutions for critical understanding.

24 publications in top academic journals

A proven track record in leveraging deep learning multimodal models to identify new biomarkers for leading biopharma companies and academic research labs.

Stratification Biomarker

Risk Score Biomarker

Case study

desktop-arrowmobile-arrow
model
arrow-mobile

Impact

Impact

arrow-desktop

Biopharma

High-value subgroups

BioPharma companies can use this model to select high-value subgroups of patients that are most likely to respond to the ICI being tested. This improves the statistical power of the trial.

Trial optimization

This also results in better selection of trial participants, success rates across trial phases and ultimately regulatory approval and more precise marketing.