AI Tools Advance Cancer Diagnosis at AMP 2026

Researchers presented AI applications in molecular diagnostics at the AMP 2026 Annual Meeting & Expo in Seattle.

Updated on Nov. 13, 2026 in Cancer

AI Tools Advance Cancer Diagnosis at AMP 2026

Live Poll

Would you trust an AI tool to help your doctor diagnose cancer?

Researchers presented studies on AI applications in molecular diagnostics at the AMP 2026 Annual Meeting & Expo. The work covers blood cancer classification, genetic testing reports, and lung and liver cancer analysis.

Why it matters

Accurate blood cancer classification is essential for determining prognosis and treatment, while EGFR testing can identify patients who may benefit from targeted therapies but consumes limited tumor tissue and lab resources. AI tools that streamline these steps could help clinicians make faster, better-informed decisions for patients across the United States.

ALMA 3 detected hematolymphoid tumors with 99.5% sensitivity and assigned specific subtypes with a 90.9% positive predictive value across 81 blood and lymphatic cancer subtypes, based on training with more than 110,000 patient samples. Clinical validation of the model is still under way.

The players

Association for Molecular Pathology (AMP)

The Association for Molecular Pathology is a professional organization of laboratory scientists and pathologists focused on molecular diagnostics and its clinical application.

The details

ALMA 3 uses DNA methylation patterns to identify and classify blood cancers. The Turnkey Precision Oncology platform processes DNA and RNA sequencing results and generates draft reports for pathologist review, showing about 95% agreement with a clinical laboratory across 28 reports. In a lung cancer study, six molecular pathology fellows reviewed 80 consecutive lung adenocarcinoma cases and found rapid EGFR testing could have been deferred in 26 cases, or 33%.

Friday, Nov. 13, 2026: Oral presentations and poster sessions covered blood cancer AI, reporting platforms, and lung and liver cancer studies.
Saturday, Nov. 14, 2026: Poster presentations covered ALMA 3 and gastric cancer mutation prediction.

Roadmap

Foundation models trained on DNA methylation data are emerging as a tool for classifying hematologic cancers, extending earlier single-marker diagnostic approaches. The reported results place these models in an active phase of clinical validation before routine use.

Faster, more consistent blood cancer classification could shorten the time before patients receive a diagnosis and treatment plan. Ongoing validation will determine whether these AI tools become part of routine testing in clinics across the country.

The takeaway

AI-assisted molecular testing may reduce repeated tests and conserve tumor tissue, but patients and clinicians should watch for validation results before expecting these tools in routine care.