{"entity":{"id":"ai-oncology-clinic","kind":"roadmap","name":"AI in the oncology clinic: from narrow cleared tools to multimodal decision support","aka":[],"tldr":"How AI is moving from single-task readers of scans and slides towards systems that weigh everything about a patient, and what regulators and evidence still require.","summary":"Hundreds of narrow AI devices are cleared, mostly in radiology triage and screening. The first predictive pathology tools (ArteraAI) prove that AI can change treatment decisions under regulation. Foundation models promise breadth, but the evidence base for deployment, the regulatory framework for updating models, and reimbursement are all unsettled.","asOf":"2026-09-08","links":[{"label":"MASAI: AI-supported mammography screening finds more cancers with half the radiologist workload (Lancet Oncology 2023)","url":"https://doi.org/10.1016/S1470-2045(23)00298-X"}],"tags":[],"related":[],"cancers":[],"sections":["ai-computation"],"technologies":["digital-pathology-ai","radiology-ai-screening","pathology-foundation-model","virchow","prov-gigapath","musk","med-gemini","ai-trial-matching","sybil","mirai"],"targets":[],"drugs":["artera-ai-prostate","artera-ai-breast"],"companies":[],"institutions":[],"pathways":[],"terms":[],"trials":[],"people":[],"bottlenecks":[],"keyPapers":[],"journals":[],"dependsOn":[],"notes":[],"steps":[{"era":"2017-2021","title":"Narrow detection tools cleared","description":"FDA clears the first AI detection aids: mammography CAD successors, Paige Prostate (2021), radiology triage for haemorrhage and embolism. Evidence is mostly reader studies.","refs":["digital-pathology-ai","radiology-ai-screening","paige"],"status":"historic"},{"era":"2022-2024","title":"Screening at scale and risk models","description":"MASAI (Sweden) shows AI-supported mammography reading finds more cancers with less workload; Sybil and Mirai predict future cancer from today's scan; whole-slide imaging becomes routine, enabling slide-level AI.","refs":["mammography","sybil","mirai","digital-pathology-ai"],"status":"current"},{"era":"2025-2026","title":"First predictive AI and foundation models in products","description":"ArteraAI Prostate (de novo 2025) predicts treatment benefit; ArteraAI Breast cleared 2026. Pathology foundation models (Virchow2, Prov-GigaPath, TITAN, Atlas) move into commercial biomarker products; multimodal models (MUSK) predict immunotherapy response retrospectively.","refs":["artera-ai-prostate","artera-ai-breast","virchow","prov-gigapath","titan","atlas-aignostics","musk"],"status":"current"},{"era":"2026-2028","title":"Prospective evidence and regulatory frameworks","description":"Randomised or pragmatic trials of AI-guided decisions (screening intervals, treatment selection); FDA predetermined change control plans for model updates; EU AI Act high-risk obligations; payment codes for AI-derived biomarkers. LLM assistants (Med-Gemini class) enter tumour boards for documentation and trial matching under human review.","refs":["med-gemini","ai-trial-matching","tempus-multimodal","foresight-ehr"],"status":"emerging"},{"era":"2029+","title":"Speculative: multimodal decision support as standard of care","description":"A single model reads slides, scans, genomics and records to recommend and monitor therapy, audited against outcomes and updated continuously. Depends on data-sharing, liability and validation questions that are open today.","refs":["idea-multimodal-foundation-model","flatiron-foundation-cgdb","imaging-data-commons"],"status":"speculative"}],"watch":[]},"route":"/roadmaps/ai-oncology-clinic/","neighbours":{"section":[{"id":"ai-computation","kind":"section","name":"AI & Computation","route":"/fronts/ai-computation/"}],"technology":[{"id":"radiology-ai-screening","kind":"technology","name":"AI in radiology","route":"/technologies/radiology-ai-screening/"},{"id":"ai-trial-matching","kind":"technology","name":"AI trial matching & clinical decision support","route":"/technologies/ai-trial-matching/"},{"id":"atlas-aignostics","kind":"technology","name":"Atlas (Aignostics, Mayo Clinic, Charité)","route":"/technologies/atlas-aignostics/"},{"id":"digital-pathology-ai","kind":"technology","name":"Digital pathology & AI","route":"/technologies/digital-pathology-ai/"},{"id":"foresight-ehr","kind":"technology","name":"Foresight (generative EHR model)","route":"/technologies/foresight-ehr/"},{"id":"mammography","kind":"technology","name":"Mammography & tomosynthesis","route":"/technologies/mammography/"},{"id":"med-gemini","kind":"technology","name":"Med-Gemini and MedLM (Google)","route":"/technologies/med-gemini/"},{"id":"mirai","kind":"technology","name":"Mirai (MIT breast cancer risk from mammograms)","route":"/technologies/mirai/"},{"id":"musk","kind":"technology","name":"MUSK (Stanford, vision-language pathology)","route":"/technologies/musk/"},{"id":"pathology-foundation-model","kind":"technology","name":"Pathology & radiology foundation models","route":"/technologies/pathology-foundation-model/"},{"id":"prov-gigapath","kind":"technology","name":"Prov-GigaPath (Microsoft, Providence)","route":"/technologies/prov-gigapath/"},{"id":"sybil","kind":"technology","name":"Sybil (MIT/MGH lung cancer risk from CT)","route":"/technologies/sybil/"},{"id":"tempus-multimodal","kind":"technology","name":"Tempus multimodal models","route":"/technologies/tempus-multimodal/"},{"id":"titan","kind":"technology","name":"TITAN (whole-slide multimodal model)","route":"/technologies/titan/"},{"id":"virchow","kind":"technology","name":"Virchow / Virchow2 (Paige, MSK)","route":"/technologies/virchow/"}],"drug":[{"id":"artera-ai-breast","kind":"drug","name":"ArteraAI Breast","route":"/drugs/artera-ai-breast/"},{"id":"artera-ai-prostate","kind":"drug","name":"ArteraAI Prostate","route":"/drugs/artera-ai-prostate/"}],"company":[{"id":"paige","kind":"company","name":"Paige AI","route":"/companies/paige/"}],"idea":[{"id":"idea-multimodal-foundation-model","kind":"idea","name":"Patient-level multimodal foundation models for treatment selection","route":"/ideas/idea-multimodal-foundation-model/"}],"collection":[{"id":"flatiron-foundation-cgdb","kind":"collection","name":"Flatiron Health and Foundation Medicine Clinico-Genomic Database","route":"/collections/flatiron-foundation-cgdb/"},{"id":"imaging-data-commons","kind":"collection","name":"NCI Imaging Data Commons (IDC)","route":"/collections/imaging-data-commons/"}],"roadmap":[{"id":"ai-oncology-roadmap","kind":"roadmap","name":"AI in oncology roadmap: pattern readers → foundation models → agents in the workflow","route":"/roadmaps/ai-oncology-roadmap/"},{"id":"diagnostics-roadmap","kind":"roadmap","name":"Diagnostics roadmap: stains → gene panels → blood tests that decide treatment","route":"/roadmaps/diagnostics-roadmap/"}]}}