Why BI-RADS 4 Is Not a Breast Cancer Diagnosis
BI-RADS 4 means a breast imaging finding is suspicious enough for tissue sampling, not that imaging has diagnosed cancer.
Health & Evidence Library
Plain-language guides to why imaging is ordered, how structured scores work, and how benefits, limits, and incidental findings are weighed.
19 guides · Page 1 of 1
BI-RADS 4 means a breast imaging finding is suspicious enough for tissue sampling, not that imaging has diagnosed cancer.
Whole-body MRI avoids ionizing radiation, but incidental findings, false results, overdiagnosis, and missing outcome evidence limit screening use.
Learn what FDG PET/CT shows, how standardized uptake values are calculated, and why inflammation, glucose, timing, lesion size, and scanner methods matter.
The STONE trial found that starting with ultrasound lowered cumulative radiation while serious outcomes were similar to a CT-first strategy.
TI-RADS uses ultrasound features and size thresholds to find consequential thyroid cancers while limiting biopsies of small indolent nodules.
How PI-RADS v2.1 classifies prostate MRI lesions, what categories 1 through 5 mean, and why PSA density, MRI quality, history, and biopsy still matter.
How proPSMA compared PSMA PET-CT with CT plus bone scan for high-risk prostate cancer, including accuracy, management change, and limits.
A plain-language guide to CTDIvol, dose-length product, and effective dose, and what ALARA, Image Gently, and Image Wisely actually ask of a scan.
What the NLST and NELSON trials, number needed to screen, and overdiagnosis reveal about who actually benefits from low-dose CT lung screening.
A plain-language guide to the MASAI trial, the first randomized test of AI in breast screening, and what its endpoints do and do not prove.
How the always-on MRI magnet, the ACR four-zone layout, and MR Conditional labeling combine into one practical implant-safety system.
How probability, size thresholds, and ACR white papers decide which incidental imaging findings get worked up and which are safely left alone.
What a 510(k) clearance really certifies, how to read sensitivity, specificity and AUC, and the validation gaps to check before trusting radiology AI.
How an ACR imaging appropriateness rating is built: a graded literature review plus a RAND/UCLA panel vote on a 1 to 9 scale, and how to read it.
How the Lung-RADS v2022 system turns a screen-detected lung nodule into a cancer probability and a next step, and what the 2022 update changed.
A plain-language guide to how the ACR groups gadolinium contrast agents by nephrogenic systemic fibrosis cases, and why chemistry alone does not set the label.
Why the 2025 projection of 103,000 CT-linked cancers is a model output, not a tumor count, and how to weigh low-dose radiation risk for a real scan.
The MQSA final rule standardizes breast density notification. Here is what that notice proves about supplemental screening and what it does not.
Why controlled studies split real dye nephrotoxicity from coincidental kidney injury, and where the true risk still lives.