AI in Oncology (medical + radiation)
In radiation oncology, auto-contouring and AI-assisted planning are the most mature uses, and professional bodies have published guidance for them. In medical oncology, AI for pathology, imaging and trial matching is growing, while outcome prediction models are mostly still research.
What AI does in Oncology (medical + radiation)
Draws organs at risk for head and neck, pelvis and other sites. Needs commissioning and checks.
Examples: Commercial contouring tools; TotalSegmentator for research
Predicts dose and speeds plan generation.
Examples: Vendor planning tools
Rapid re-contouring and re-planning on the day.
Examples: Vendor adaptive platforms
AI reads slides and scans that feed oncology decisions.
Examples: See radiology and pathology pages
Finds eligible patients from records.
Examples: Hospital and vendor tools
Models using imaging, pathology and genomics predict outcome.
Examples: Multimodal research models
routine = in everyday use in many centres · early adoption = in use at some sites, evidence still building · research = studies only
Key ideas for this specialty
New to the basics? Start with the theory lessons.
Tools
Open-source organ segmentation on CT and MRI, used in research for contouring.
Regulatory status: Not listed
Open segmentation framework behind many research contouring models.
Regulatory status: Not listed
A listing here is not an endorsement. Ask any vendor for its CDSCO licence and class before clinical use.
Where to learn
Consensus advice on selecting, commissioning and monitoring auto-contouring.
Practice data for oncology imaging projects.
Review across the cancer pathway.
Links pathology AI to treatment choice.
Combining imaging, pathology and genomics.
Covers prognosis and imaging models.
Do it yourself
Run TotalSegmentator on a public CT from TCIA. Compare the organ outlines with what you would draw and note edits needed.
Open the dataset or tool →Open a study in cBioPortal, plot survival by a mutation, and note the sample size and missing data.
Open the dataset or tool →Use public or properly de-identified data only, and get ethics approval before any project on your own patients.
Papers worth reading
- Artificial intelligence for clinical oncology Cancer Cell, 2021
- Multimodal biomedical AI Nature Medicine, 2022
- Artificial intelligence in digital pathology - new tools for diagnosis and precision oncology Nature Reviews Clinical Oncology, 2019
Societies and guidance
Auto-contouring guidance (2024).
European radiotherapy society; has AI education and guidance material.
Recent updates
Lancet: full MASAI results show AI-supported mammography screening lowered interval cancers by 12%
In the Swedish randomised trial of over 100,000 women, AI-supported screening cut interval cancer diagnoses by 12% with similar false positives (1.5% vs 1.4%). It also found 16% fewer invasive and 27% fewer aggressive-subtype cancers between screens.
Source: ecancerLinks checked 3 October 2026. “Not listed” means we could not confirm the detail, not that it is absent. Educational material, not clinical advice.