An integrated, non-invasive platform that detects early-stage lung cancer and other diseases from exhaled breath — delivering a clinical probability score in minutes.
Current detection methods are expensive, invasive, or arrive too late. The result: most patients are diagnosed only when treatment options are limited.
"CT scans are expensive and expose patients to radiation. Biopsies are invasive and come too late. Blood tests are still experimental. There is a clear gap: a fast, non-invasive, primary care–ready screening tool does not yet exist."
Our integrated system takes a patient from walk-in to clinical probability score in a single, non-invasive workflow that can run in any clinical setting.
Two proven scientific principles — breath biomarker chemistry and probabilistic inference — combined into a single diagnostic pipeline.
When cancer cells grow, they produce metabolic byproducts — small chemical compounds that enter the bloodstream and are eventually exhaled. Scientists call these Volatile Organic Compounds (VOCs).
Our proprietary system detects a validated panel of several proprietary VOC biomarkers, for example:
* Specific biomarker identities are proprietary and protected under proprietary IP. Chemical identities available to qualified collaborators and licensees under NDA.
Our system uses an Advanced AI model for clinical diagnostic reasoning by combining multiple biomarker measurements into a single, reliable probability estimate.
Rather than a simple threshold, the system uses an advanced model to process the collected data.
Representative data from peer-reviewed clinical studies · Evidence quality: 90/100
| Biomarker A* Concentration (Relative Units) | P(C | Lung Cancer) | P(C | Healthy) | Likelihood Score | Interpretation |
|---|---|---|---|---|
| 0 – 5,000 | 0.05 | 0.25 | 0.20 | Strong evidence AGAINST LC |
| 5,000 – 10,000 | 0.10 | 0.35 | 0.29 | Moderate evidence against |
| 10,000 – 15,000 | 0.15 | 0.25 | 0.60 | Weak evidence against |
| 15,000 – 20,000 | 0.22 | 0.10 | 2.20 | Moderate evidence FOR LC |
| 20,000 – 30,000 | 0.28 | 0.04 | 7.00 | Strong evidence FOR LC |
| 30,000 – 50,000 | 0.12 | 0.01 | 15.0 | Very strong evidence FOR |
| > 50,000 | 0.05 | 0.001 | 50.0 | Definitive evidence FOR |
* Biomarker A identity is proprietary and protected under proprietary IP. Likelihood values derived from peer-reviewed published clinical studies. Full biomarker panel and methodology available to qualified collaborators under NDA.
Lung cancer screening is dramatically underutilized. Only a fraction of eligible patients receive recommended annual screening — because current tools are not practical for primary care settings.
The integrated Detection Technology + AI diagnostic system has been protected, including the full system architecture, detection methodology, and AI inference pipeline. Seeking NIAID/NHLBI SBIR/STTR Phase I funding and external funding.
The same breath-based VOC detection and AI platform is applicable to any condition where metabolic biomarkers are present in exhaled breath. Lung cancer is the lead application — but the platform is disease-agnostic by design.
AATG's core expertise in advanced analytical instrumentation extends to detection systems for security, pharmaceutical, and defense applications.
A revolutionary Explosive Trace Detector leveraging advanced spectrometry for high-sensitivity identification of explosive materials. Designed for airports, public venues, and critical infrastructure with significantly reduced operating costs and no radioactive ionization source.
Explosive Trace DetectionNext-generation Chemical Warfare Agent Detector offering broad-spectrum detection of CWAs and Toxic Industrial Components (TICs). Wearable, battery-powered, and designed for continuous monitoring in battlefield and environmental applications.
Chemical Warfare DetectionA fast, on-site cleaning validation analyzer for pharmaceutical manufacturing. Eliminates the need for HPLC/MS with real-time contaminant detection in under 1 minute, dramatically reducing production downtime and validation costs.
Cleaning ValidationWe welcome inquiries from clinical researchers, potential co-founders, investors, and SBIR/STTR program officers. Whether you're interested in the breath diagnostics platform or our other detection technologies, we'd love to hear from you.