AI Medicine Glossary
Plain-language definitions for the clinical AI, medical imaging AI, and healthcare AI terms covered across AI Medicine Now.
A
-
AI Diagnostics
AI software used to detect, classify, or flag disease findings from clinical data such as images or lab results.
-
AI Scribe
A shorthand term for ambient AI documentation software that drafts clinical notes from a patient visit.
-
Ambient Clinical Documentation
AI technology that listens to a patient-clinician conversation and drafts a clinical note for physician review.
B
-
Bias in Clinical AI
Systematic differences in an AI model's accuracy across patient subgroups, often traced to unrepresentative training data.
C
-
Clinical Decision Support (CDS)
Software that gives clinicians patient-specific guidance or alerts at the point of care.
-
Clinical Validation
The process of testing an AI tool's performance against real clinical data and outcomes before and after deployment.
-
Computer-Aided Detection (CAD)
Software that flags regions of a medical image that may warrant closer review by a clinician.
E
-
Explainable AI (XAI)
AI systems designed to make their outputs and reasoning understandable to a human clinician.
F
-
FDA Clearance (510(k))
The most common FDA regulatory pathway used to authorize marketing of moderate-risk medical devices, including many AI diagnostic tools, in the U.S.
H
-
Human-in-the-Loop
A design principle where a clinician reviews and can override an AI system's output before it affects care.
L
-
Large Language Model (LLM) in Medicine
A large language model applied to healthcare tasks such as documentation, summarization, or clinical Q&A.
M
-
Model Drift
The gradual decline in an AI model's accuracy as real-world data changes from what it was trained on.
P
-
Prior Authorization AI
AI software used to prepare, submit, or process prior authorization requests to insurers.
R
-
Radiology AI
AI software applied to radiology images, such as X-ray, CT, and MRI, to assist detection, triage, or quantification.
-
Real-World Evidence (RWE)
Evidence about a clinical AI tool's performance drawn from routine clinical use rather than a controlled study.
S
-
Sensitivity and Specificity
Two standard statistics used to measure a diagnostic test's ability to correctly identify positive and negative cases.
-
Software as a Medical Device (SaMD)
Software intended for a medical purpose that performs that purpose without being part of a hardware medical device.