Research area · translational

Computational Medicine

Computational medicine translates molecular, cellular, and clinical data into predictive models of disease progression. We combine medical knowledge graphs with computer vision for digital pathology and polygenic risk scoring. These models personalize therapeutic interventions, tailoring care to each patient's risk profile.

0
focus areas
0
model families
0
unified pipeline
Overview

What this area is.

Computational medicine turns molecular and clinical data into predictions that change decisions: who is at risk, what a tumour will do next, which therapy fits which patient.

We combine predictive analytics, computational pathology and knowledge graphs with explainable AI, keeping clinicians in the loop and models interpretable.

Tools & technologies

PyTorchscikit-learnKnowledge graphsSurvival modelsDigital pathologySHAPFoundation models
Knowledge graphGenes, drugs and diseases, connected.
Risk stratificationOutcome split by predicted risk group.
Capabilities

What we do.

Core methods we apply in computational medicine.

Disease-risk prediction

Polygenic and multi-omic models estimating individual risk.

Predictive analytics & outcomes

Forecasting progression, response and outcome.

Clinical decision-support systems

Interpretable recommendations grounded in evidence.

Medical knowledge graphs

Linking genes, drugs, diseases and phenotypes for reasoning.

Computational pathology

Deep learning over histology and digital pathology.

Biomedical foundation models

Large models adapted to clinical and molecular data.

Workflow

From data to insight.

01

Data

molecular · clinical · imaging

02

Model

ML · foundation models

03

Predict

risk · outcome · diagnosis

04

Validate

external cohorts

05

Explain

attributions & evidence

06

Clinic

decision support

Visual analytics

Publication-grade figures.

Knowledge graphGenes, drugs and diseases, connected.
Risk stratificationOutcome split by predicted risk group.
Predictive featuresModel-ranked signals of impact.
Patient matrixFeatures across a patient cohort.
Focus

Where we go deep.

Precision therapeutics

Matching patients to the therapy most likely to work.

Virtual clinical trials

In-silico modelling to design and de-risk trials.

AI-assisted diagnosis

Decision support that augments clinical judgement.

Insights

Questions we answer.

A few common questions about computational medicine. Ask CGB-AI for more.

What is computational medicine?

The discipline of predicting and personalizing care with computational models — from disease risk to diagnosis to therapy selection.

Does AI replace the clinician?

No — models predict and prioritize; clinicians interpret and decide. We keep humans in the loop and predictions explainable.

Selected research

Publications in Computational Medicine.

Drawn from our record of 170 papers, filtered to this area.

Browse all publications →

Start a computational medicine project.

Tell us the clinical question and the data you have — we will map out an approach.

Collaborate with us →
Hi! Ask me about CGB 👋CGB mascot