Practical Statistical Considerations for the Clinical Validation of AI/ML-enabled Medical Diagnostic Devices
It addresses validation challenges for regulatory and clinical stakeholders, but is incremental as it synthesizes existing knowledge.
The paper discusses statistical considerations for evaluating AI/ML-enabled medical diagnostic devices, providing references and good practices for clinical validation.
Artificial Intelligence (AI) and Machine-Learning (ML) models have been increasingly used in medical products, such as medical device software. General considerations on the statistical aspects for the evaluation of AI/ML-enabled medical diagnostic devices are discussed in this paper. We also provide relevant academic references and note good practices in addressing various statistical challenges in the clinical validation of AI/ML-enabled medical devices in the context of their intended use.