From:  Artificial intelligence in the interventional management of liver disease: a narrative review from foundational concepts to clinical applications

 The AI project lifecycle: from data to clinical deployment.

StageKey stepsActivitiesIR relevance and goal
Conception & DataDefine problemIdentify a clear clinical question (e.g., predicting TACE response)The goal is to obtain sufficient, high-quality data despite scarcity challenges in IR
AcquisitionGather multimodal data, including imaging, labs, and clinical records
Preprocessing & CurationLabelingAssign a “ground truth” to the data (e.g., manual tumor segmentation or classifying patient outcomeClinician expertise is required to perform accurate labeling and to ensure features reflect meaningful pathology
Feature extractionTransform images into quantitative data (e.g., radiomic features from HCC texture)
Validation & TestingSplit dataSeparate the dataset into training, validation, and untouched testing setsThis stage establishes rigor: models must prove accuracy on unseen patients to be considered trustworthy for clinical decision-making
External validationTest the final model on data from a different center to prove generalizability
Mitigate overfittingEnsure the model performs well on new data and does not fail due to over-memorization
Deployment & IntegrationEvaluationQuantify performance using clinical metrics such as AUC, sensitivity, and specificityThe goal is to achieve genuine clinical benefit by overcoming practical barriers and securing clinician trust before routine use
Workflow integrationEnsure the tool fits seamlessly into the IR suite and minimizes disruption to existing protocols
Return to article view

AI: artificial intelligence; TACE: transarterial chemoembolization; IR: interventional radiology; HCC: hepatocellular carcinoma; AUC: Area Under the Curve.