Skip to workflow
Open Source · Not For ProfitAI For Health · Care For All

Real-time Precision Monitoring of Oncology

AI — Powered

With just a drop of blood, AI can give you the insights that guide cancer prediction, continuous monitoring and treatment.

Supported by Senus Healthcare Research Institute, USA

From Sample to Clinical Insight

Five stages, one continuous liquid biopsy workflow.

Blood Draw

01

Blood Draw

A routine blood sample is collected from the patient.

Plasma Separation

02

Plasma Separation

Plasma is isolated to access circulating tumor DNA.

ctDNA + NGS

03

ctDNA + NGS

ctDNA is extracted and sequenced for a full genomic profile.

AI Analysis

04

AI Analysis

AI models identify and tier clinically relevant variants.

AI Report

05

AI Report

Findings are compiled into a structured, evidence-linked report.

Process flow, left to right

AI-Powered Mammogram Risk Analysis

We combine clinical radiology expertise with advanced deep learning to analyze mammogram images and predict individualized breast cancer risk over the next 5 years — enabling earlier intervention and better patient outcomes.

AI mammogram risk analysis process

Predictive Intelligence

Predicting Cancer Before It Occurs

  1. Submit Mammogram

    Extracts deep tissue features, density patterns, and micro-calcification signatures from input mammograms.

  2. AI Risk Analysis

    Generates a personalized 5-year risk score with explainable confidence intervals.

  3. Clinical Interpretation

    Delivers structured risk reports that integrate seamlessly into radiologist workflows.

Real-time monitoring using
Liquid Biopsy.

Non-invasive liquid biopsy combined with AI for continuous cancer progression tracking and treatment response monitoring.

Liquid biopsy visualization
Next Generation Sequencing machine

Next-generation sequencing at the core.

Ultra-sensitive genomic profiling from circulating tumor DNA (ctDNA) without invasive tissue sampling. AI-driven analysis delivers real-time insights.

  • Non-invasive
  • Real-time insights
  • AI-powered
  • ctDNA analysis

Currently in R&D, refining accuracy, scalability, and clinical integration for cancer-positive patients.

AI Lung Cancer Imaging

AI-Powered Lung Cancer Analysis

Low-dose CT screening finds lung cancer early, when treatment works best. We are developing AI that reads chest CT scans to find, measure and track lung nodules, so radiologists can focus on what matters most.

Chest CT · AxialLung window · W1500 / L−600
Axial chest CT scan in the lung window, with both lung fields outlined
Illustrative · not a diagnosisSlice 5.0 mm · 120 kV
Chest CT slice showing a cavitating tumour in the left lower lobe, circled

Example finding

Cavitating tumour, left lower lobe

Separate public case · illustrative

How it works

From CT scan to risk insight

  1. Upload a CT scan

    Low-dose screening or diagnostic chest CT, as a standard DICOM series.

  2. Detect and outline nodules

    AI reads every slice to find lung nodules and outline them in 3D.

  3. Estimate risk

    Size, density and shape feed a malignancy-risk estimate aligned with Lung-RADS® categories.

  4. Track change over time

    Comparing scans measures nodule growth and volume-doubling time.

The Problem

The Challenge in Modern Cancer Monitoring

Cancer care is still reactive, fragmented, and delayed. Risk prediction, diagnosis, and monitoring often work in separate systems, leaving critical gaps across the patient journey.

5 critical gaps identified

  • Many cancers are detected only after symptoms appear, when treatment is harder and survival is lower.

    99%→32%breast cancer 5-year survival, localized vs distant

  • Cancer is usually checked through scans every few weeks, not continuously. Important changes can happen in between.

    6–12wkcommon interval between imaging assessments

  • Tissue biopsy is invasive, burdensome, and not practical for frequent tracking over time.

    15–25%pneumothorax risk in some lung biopsies

  • After treatment, tiny amounts of cancer can remain unseen. Relapse is often found only later on imaging.

    Monthsearlier relapse signals seen with ctDNA in some studies

  • Imaging, pathology, genomics, and follow-up data often sit in separate systems, limiting a full patient view.

    10Mglobal cancer deaths in 2020
OncoTrace-AI · Open-Source · Not-for-Profit

Real-Time Precision
Monitoring in Oncology

AI-driven cancer intelligence across the entire disease lifecycle — from early risk prediction using imaging data to continuous progression monitoring powered by liquid biopsy.

  • Early Risk Detection
  • 24 / 7 Monitoring
  • Open Source
  • Non-Invasive
ACTIVE

Real-Time Monitoring

Continuous tracking across the entire disease lifecycle with actionable insights as cancer evolves.

All systems operational

Compliant

Privacy & Compliance

End-to-end de-identification, audit trails, and privacy-first data pipelines by design.

  • HIPAA
  • GDPR
  • ISO 27001
  • SOC 2

Our Mission

Making precision oncology intelligence accessible, transparent, and globally collaborative — from first risk signal to real-time progression tracking.
OncoTrace-AI

Who It's For

Built for the entire oncology ecosystem

  • Clinicians

    Predictive risk insights and real-time monitoring data for faster, more informed decisions.

  • Patients

    Clarity through early risk signals and transparent, continuous precision monitoring.

  • Researchers

    Open-source oncology tools to build, test, and collaborate across prediction and monitoring.

  • Pharma

    Longitudinal monitoring and continuous data across the full disease and treatment lifecycle.

BUILD TOGETHER

Connect With Us

Interested in a partnership? Reach out to us.