AI in oncology: connect detection to patient outcomes
By ChatGPT Health. Educational synthesis prepared with AI assistance. Sources checked September 28, 2026. Clinically reviewed by Harvey Castro, MD, MBA, on September 28, 2026. Approved as educational content. This page is not a clinical protocol.
Oncology AI spans screening, pathology, imaging, treatment planning and research. Evidence from one part of the cancer pathway does not validate the entire pathway.
Name the clinical decision
Specify whether the tool prioritizes an image, helps a pathologist, estimates recurrence risk, matches a patient to a trial or recommends treatment. Identify the comparator, who reviews the output and how a disagreement is resolved. Keep research findings separate from tools available for routine care.
Updated mammography evidence
The 2026 MASAI analysis compared AI supported mammography with standard double reading in Sweden. Interval cancer rates were 1.55 versus 1.76 per 1,000 participants, meeting the trial’s noninferiority criterion. The ratio was 0.88 (95% CI 0.65 to 1.18); this does not establish a statistically clear reduction in interval cancers. It also does not establish a mortality benefit. Read the primary analysis.
Choose outcomes that match the claim
- Screening: interval cancers, recalls, unnecessary procedures and eventual health outcomes.
- Pathology and imaging: missed disease, disagreements, turnaround time and effects on decisions.
- Treatment support: whether recommendations are appropriate for the patient and improve meaningful outcomes.
- Trial matching: eligibility confirmed by the study team, access and actual enrollment.
Questions for a cancer care team
Ask whether the evidence matches the cancer type, stage, treatment context and patient population. Request validation outside the development institution and disclosure of relevant commercial interests. Consider what happens when a specimen, scan or clinical history is incomplete or falls outside the tool’s intended use.
For patients and caregivers
AI can help organize questions about a report, but the original report and treatment discussion remain essential. Ask the oncology team whether AI influenced the assessment and how it was checked. A chatbot summary should not change an appointment, investigation or cancer treatment plan.
Sources
Explore related topics
Put the evidence to work
Download the evidence and safety evaluation worksheets or compare the study designs and their limits.
Editorial standards · Suggest a correction. Please do not send identifiable patient information.