# AI in radiology

Source: https://onco.cc/technologies/radiology-ai-screening/  
OnCo record `radiology-ai-screening` (Technology). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

Software that reads scans alongside radiologists, catching cancers earlier and predicting who is at risk.

## Summary

Hundreds of FDA-cleared radiology AI devices exist; oncology use cases include mammography reading (Transpara, Lunit INSIGHT, MASAI trial in Sweden showed 29% more cancers detected with 44% less workload), lung nodule detection and malignancy scoring (Sybil, Optellum), prostate MRI, and risk models (Mirai). Foundation models linking images with text are emerging.

## Fields

- Kind: Technology
- Status: established
- Last checked: 2026-09-04
- Principle: Deep convolutional and transformer networks trained on labelled imaging; increasingly self-supervised on large unlabelled corpora.
- Strengths: Scales expert reading; Reduces workload and inter-reader variability
- Limitations: Dataset shift across scanners and populations; Regulatory lag for adaptive models

## Sources

- Wikipedia: https://en.wikipedia.org/wiki/Artificial_intelligence_in_healthcare
- Wikipedia: https://en.wikipedia.org/wiki/Artificial_intelligence_in_healthcare

## Connected records

- roadmaps: [AI in oncology roadmap: pattern readers → foundation models → agents in the workflow](https://onco.cc/roadmaps/ai-oncology-roadmap/), [AI in the oncology clinic: from narrow cleared tools to multimodal decision support](https://onco.cc/roadmaps/ai-oncology-clinic/), [Early detection roadmap: organ screening → blood tests for many cancers](https://onco.cc/roadmaps/early-detection-roadmap/), [Molecular imaging roadmap: FDG → PSMA → FAP → antigen and immune PET](https://onco.cc/roadmaps/molecular-imaging-roadmap/)
- technologies: [AI compute and model platforms for oncology](https://onco.cc/technologies/ai-compute-platforms/), [Aidoc CARE (clinical radiology foundation model)](https://onco.cc/technologies/aidoc-care/), [Breast MRI coils and abbreviated breast MRI](https://onco.cc/technologies/breast-mri-coils-abbreviated-mri/), [Colposcopes and digital cervical screening devices (DYSIS, EVA System, AVE)](https://onco.cc/technologies/colposcopes-digital-cervical-screening/), [Contrast-enhanced mammography](https://onco.cc/technologies/contrast-enhanced-mammography/), [CT (computed tomography)](https://onco.cc/technologies/ct/), [CT body composition and sarcopenia measurement](https://onco.cc/technologies/ct-body-composition-sarcopenia/), [CT-FM (whole-body CT foundation model)](https://onco.cc/technologies/ct-fm/), [CT, MRI and PET scanner manufacturing](https://onco.cc/technologies/medical-imaging-scanner-manufacturing/), [Dermoscopy, total-body photography & AI skin analysis](https://onco.cc/technologies/dermoscopy-ai/), [Federated learning and privacy-preserving AI](https://onco.cc/technologies/federated-learning-medical-ai/), [Hand-held and point-of-care ultrasound](https://onco.cc/technologies/hand-held-ultrasound/), [Low-dose CT lung screening](https://onco.cc/technologies/low-dose-ct-screening/), [Mammography & tomosynthesis](https://onco.cc/technologies/mammography/), [MedSAM / SAM-Med3D (segment anything for medicine)](https://onco.cc/technologies/medsam/), [Merlin (Stanford abdominal CT vision-language model)](https://onco.cc/technologies/merlin-ct/), [Mirai (MIT breast cancer risk from mammograms)](https://onco.cc/technologies/mirai/), [MRI](https://onco.cc/technologies/mri/), [Multiparametric prostate MRI (PI-RADS)](https://onco.cc/technologies/mp-mri/), [NHS Targeted Lung Health Check (lung cancer screening programme)](https://onco.cc/technologies/nhs-targeted-lung-health-check/), [Pathology & radiology foundation models](https://onco.cc/technologies/pathology-foundation-model/), [Quantitative imaging biomarkers (RECIST, PERCIST, SUV, ADC)](https://onco.cc/technologies/quantitative-imaging-biomarkers/), [RadFM (generalist radiology foundation model)](https://onco.cc/technologies/radfm/), [Radiogenomics: predicting radiation sensitivity from genes](https://onco.cc/technologies/radiogenomics/), [Sybil (MIT/MGH lung cancer risk from CT)](https://onco.cc/technologies/sybil/), [Tumour volume doubling time](https://onco.cc/technologies/tumour-doubling-time/)
- fronts: [AI & Computation](https://onco.cc/fronts/ai-computation/), [Imaging](https://onco.cc/fronts/imaging/)
- companies: [Aidoc](https://onco.cc/companies/aidoc/), [Google (Health, DeepMind, Verily)](https://onco.cc/companies/google-health/), [iCAD (RadNet)](https://onco.cc/companies/icad/), [iSono Health](https://onco.cc/companies/isono-health/), [Kheiron Medical Technologies](https://onco.cc/companies/kheiron-medical-technologies/), [Lunit](https://onco.cc/companies/lunit/), [Nucleo](https://onco.cc/companies/nucleo-research/), [Optellum](https://onco.cc/companies/optellum/), [ScreenPoint Medical](https://onco.cc/companies/screenpoint-medical/), [Therapixel](https://onco.cc/companies/therapixel/), [Vara](https://onco.cc/companies/vara/), [Volpara Health](https://onco.cc/companies/volpara-health/)
- cancers: [Ductal carcinoma in situ (DCIS)](https://onco.cc/cancers/ductal-carcinoma-in-situ/), [HR-positive / HER2-negative breast cancer](https://onco.cc/cancers/breast-hr-positive/), [Non-small-cell lung cancer](https://onco.cc/cancers/nsclc/)
- terms: [Body composition (lean mass, fat mass, visceral fat)](https://onco.cc/terms/body-composition/), [Early detection](https://onco.cc/terms/early-detection-term/), [Lung-RADS and nodule reporting](https://onco.cc/terms/lung-rads/), [Oncology workforce](https://onco.cc/terms/oncology-workforce/), [Pulmonary nodule](https://onco.cc/terms/pulmonary-nodule/), [Radiology imaging as a data modality (CT, MRI, TCIA)](https://onco.cc/terms/radiology-imaging-modality/), [Screening](https://onco.cc/terms/screening/)
- people: [Constance D. Lehman](https://onco.cc/people/constance-lehman/), [Denise R. Aberle](https://onco.cc/people/denise-aberle/), [Kristina Lång](https://onco.cc/people/kristina-lang/), [Laura J. Esserman](https://onco.cc/people/laura-esserman/), [Lecia V. Sequist](https://onco.cc/people/lecia-sequist/), [Mark Schiffman](https://onco.cc/people/mark-schiffman/), [Per Hall](https://onco.cc/people/per-hall/), [Regina Barzilay](https://onco.cc/people/regina-barzilay/), [Weimin Li](https://onco.cc/people/li-weimin/)
- bottlenecks: [AI that is built but not validated or deployed](https://onco.cc/bottlenecks/b-ai-validation/), [Not enough oncologists, nurses, pathologists, physicists](https://onco.cc/bottlenecks/b-workforce/), [The hardest cancers are found late](https://onco.cc/bottlenecks/b-early-detection/)
- ideas: [A dedicated fund for randomised trials of cancer AI with patient outcomes](https://onco.cc/ideas/idea-data-prospective-ai-trials-fund/), [A delivery-science moonshot for prevention we already own](https://onco.cc/ideas/idea-fund-prevention-moonshot/), [A federated learning consortium of cancer centres that jointly own the models](https://onco.cc/ideas/idea-fund-federated-learning-consortium/), [AI clears the normal lung screening scans so radiologists read only the suspicious ones](https://onco.cc/ideas/idea-prev-ldct-ai-negative-triage/), [AI malignancy scores to end repeat scans and biopsies for benign lung nodules](https://onco.cc/ideas/idea-prev-lung-nodule-ai-discharge/), [AI that spots pancreatic cancer on scans taken a year before diagnosis](https://onco.cc/ideas/idea-prev-pancreas-ai-prediagnostic-ct/), [AI-assisted central imaging reads to cut endpoint cost and variability](https://onco.cc/ideas/idea-tr1-ai-central-imaging-reads/), [Continuous prospective validation for every oncology AI tool after deployment](https://onco.cc/ideas/idea-moon-continuous-ai-validation-registry/), [Dose chemotherapy by muscle mass, not body surface area](https://onco.cc/ideas/idea-bio2-lean-mass-dosing/), [Every routine CT scan checked by AI for early cancer signs, with a tracked follow-up pathway](https://onco.cc/ideas/idea-prev-opportunistic-ct-ai-registry/), [Federated training of pathology and radiology models across hospitals](https://onco.cc/ideas/idea-data-federated-learning-imaging/), [Molecular indolence classifiers bundled with every screening programme](https://onco.cc/ideas/idea-moon-indolence-classifiers-with-screening/), [Read muscle loss automatically from scans patients already have](https://onco.cc/ideas/idea-bio2-ai-sarcopenia-from-ct/), [Require stage-shift or interval-cancer endpoints for AI in cancer screening](https://onco.cc/ideas/idea-data-ai-screening-endpoints/), [Roll out AI-supported mammography nationally as a stepped-wedge trial](https://onco.cc/ideas/idea-prev-mammography-ai-stepped-wedge/), [Sequestered, prospectively collected benchmark datasets that no one can train on](https://onco.cc/ideas/idea-data-sequestered-prospective-benchmarks/), [Set each woman's mammogram interval from her last mammogram, using AI risk](https://onco.cc/ideas/idea-prev-ai-mammogram-risk-intervals/), [Social impact bonds for cancer prevention, repaid from avoided treatment costs](https://onco.cc/ideas/idea-fund-social-impact-bonds-prevention/), [Validate and reimburse AI contouring and planning to expand radiotherapy capacity](https://onco.cc/ideas/idea-fund-rt-planning-ai-capacity/)
- key papers: [Artificial intelligence for breast cancer detection in screening mammography in Sweden: a prospective, population-based, paired-reader, non-inferiority study (ScreenTrustCAD)](https://onco.cc/key-papers/paper-screentrustcad-dembrower-lancet-digit-health-2023/), [MASAI: AI-supported mammography screening finds more cancers with half the radiologist workload](https://onco.cc/key-papers/paper-masai-lancet-oncol-2023/), [NELSON: volume-based CT screening reduces lung cancer deaths with fewer false alarms](https://onco.cc/key-papers/paper-nelson-nejm-2020/)
- institutions: [International Association for the Study of Lung Cancer](https://onco.cc/institutions/iaslc/), [Medicines and Healthcare products Regulatory Agency](https://onco.cc/institutions/mhra/)
- collections: [The Cancer Imaging Archive (TCIA)](https://onco.cc/collections/tcia/)
- drugs: [Lunit INSIGHT MMG](https://onco.cc/drugs/lunit-insight-mmg/), [Mia (Mammography Intelligent Assessment)](https://onco.cc/drugs/kheiron-mia/), [Optellum Virtual Nodule Clinic](https://onco.cc/drugs/optellum-virtual-nodule-clinic/), [ProFound AI (iCAD)](https://onco.cc/drugs/icad-profound-ai/), [Transpara](https://onco.cc/drugs/transpara/)
- trials: [MASAI (Mammography Screening with Artificial Intelligence)](https://onco.cc/trials/masai/), [ScreenTrustCAD](https://onco.cc/trials/screentrustcad/)

---
JSON: https://onco.cc/api/v1/entities/radiology-ai-screening.json