Findings
The document introduces a risk-based credibility assessment framework for establishing and evaluating the credibility of an Artificial Intelligence (AI) model's output when used to support regulatory decisions regarding drug safety, effectiveness, or quality. The framework outlines a 7-step process beginning with defining the question of interest and the Context of Use (COU). Credibility is defined as trust, established through evidence, in the AI model's performance for a particular COU. The credibility assessment is tailored to the AI model risk, which is a combination of model influence (the AI model's evidence contribution relative to other evidence) and decision consequence (the significance of an adverse outcome from an incorrect decision). The document highlights challenges with AI use, including variability in development datasets (training/tuning), the need for methodological transparency due to model complexity, difficulty in quantifying and interpreting uncertainty in model output, and the potential for performance change over time (data drift), which necessitates life cycle maintenance.

Recommendations
Sponsors and interested parties should define the question of interest and clearly define the COU, detailing the AI model's specific role and scope and whether other information will be used. They should assess the AI model risk (low, medium, or high) to ensure that subsequent credibility assessment activities (Step 4) are commensurate with that risk and tailored to the COU. For Step 4, the credibility assessment plan should include a description of the model, model development process (including inputs, architecture, feature selection, and rationale), and data used (training and tuning data). Development data must be deemed fit for use (relevant and reliable) to mitigate issues like algorithmic bias. The plan should also detail the model evaluation process using independent test data and include performance metrics with confidence intervals, an estimate of uncertainty, and a description of model limitations. Early engagement with the FDA is strongly encouraged to discuss model risk and the adequacy of the credibility assessment plan.

Regulatory Considerations
The risk-based credibility assessment framework is intended to help organize and document information for regulatory submissions. The required stringency of assessment activities and the level of documentation should be commensurate with the AI model risk. For AI models whose performance can change over time (e.g., in pharmaceutical manufacturing or postmarketing), sponsors must implement life cycle maintenance plans to monitor performance and manage changes in a risk-based manner. Changes to AI models should be evaluated through the manufacturer's change management system and may require re-execution of parts of the credibility assessment plan. Early engagement can be facilitated through formal meetings (e.g., Pre-IND) or other specialized programs listed in the guidance, such as the Center for Clinical Trial Innovation (C3TI), the Model-Informed Drug Development (MIDD) Paired Meeting Program, and the Emerging Technology Program (ETP) or Advanced Technologies Team (CATT).

Findings
The use of Artificial Intelligence (AI) and Machine Learning (ML) is being applied to a broad range of drug development activities with the potential to accelerate the process and make clinical trials safer and more efficient. The inclusion of AI/ML is most common in the clinical development/research phase of regulatory submissions. Concerns exist that AI/ML algorithms could amplify errors and preexisting biases in underlying data sources, which raises issues related to generalizability and ethical considerations. Other challenges include limited explainability due to model complexity and proprietary reasons, as well as managing risks related to data quality, reliability, and representativeness. The FDA recognizes that a careful, risk-based assessment of the specific context of use (COU) is needed when evaluating AI/ML.

Recommendations
Stakeholders should adhere to practices in three key areas: human-led governance, accountability, and transparency; quality, reliability, and representativeness of data; and model development, performance, monitoring, and validation. A risk management plan should be applied to identify and mitigate risks based on the COU, guiding the level of documentation and transparency. Practices are needed to ensure the integrity of AI/ML and address issues like bias and missing data. For models, developers should use pre-specification steps and clear documentation for development and assessment criteria. Models must be monitored over time for reliability and consistency, and Real-World Data (RWD) performance can provide valuable feedback, including for potential re-training.

Regulatory Considerations
The FDA encourages early engagement through mechanisms like the Critical Path Innovation Meetings (CPIM), ISTAND Pilot Program, and Emerging Technology Program to discuss relevant AI/ML methodologies or technologies. The Verification and Validation (V&V 40) risk-informed credibility assessment framework and the principles for Good Machine Learning Practices (GMLP), while not specific to drug development, are helpful guides for evaluating models. The industry is exploring the use of a Predetermined Change Control Plan (PCCP) mechanism for AI/ML-based devices to proactively specify and manage modifications, enhancing adaptability. In general, a risk-based approach should guide the level of evidence and record keeping needed for the verification and validation of AI/ML models for a specific COU.

Findings
CADe devices must meet classification requirements under 21 CFR 892.2050, including general and special controls, and require FDA clearance through 510(k) submissions.
Each new CADe device or significant modification must demonstrate substantial equivalence to a predicate device in terms of safety and effectiveness.
Robust testing and validation are necessary, including standalone and clinical performance assessments, to evaluate detection accuracy and false positive rates.
Devices with substantive technological differences or new intended uses may require clinical performance assessments.
Enrichment strategies for study populations (e.g., including challenging cases) are encouraged but should not bias performance evaluations.

Recommendations
Clearly describe the CADe algorithm, training datasets, scoring methodologies, and intended use in premarket submissions.
Conduct standalone performance assessments to measure detection accuracy and generalizability.
Compare new devices to predicate devices whenever possible, using consistent datasets and methodologies.
Develop and submit user training materials that address expected device performance, limitations, and appropriate usage scenarios.
Provide comprehensive labeling, including indications for use, directions, warnings, precautions, and performance metrics, to ensure clinician understanding and appropriate application.

Regulatory Considerations
All CADe devices under 21 CFR 892.2050 must comply with 510(k) premarket notification requirements, including general and special controls.
Changes to CADe algorithms or device characteristics must be evaluated for significant impact on safety and effectiveness, potentially requiring new submissions.
Devices with altered indications for use or significant technological differences may need additional clinical performance studies to demonstrate substantial equivalence.
Labeling must comply with 21 CFR Part 801 and provide sufficient information to describe the device, its intended use, and directions for use.
Manufacturers should consult FDA for guidance on substantial modifications or unique device characteristics.

Findings
Patient experience data encompass a range of inputs, including symptom burdens, treatment impacts, patient preferences, and views on unmet medical needs.
These data inform all stages of medical product development, from discovery to post-market use.
Quantitative methods (e.g., surveys) provide numerical insights, while qualitative methods (e.g., interviews) offer in-depth understanding. Mixed methods combine both for a fuller perspective.Social media and verified patient communities present novel data collection opportunities but require consideration of verification and representativeness challenges.
Probability sampling (e.g., stratified random sampling) is emphasized for generalizability, while non-probability methods (e.g., convenience sampling) are useful for exploratory research. Representativeness ensures that patient input reflects the diversity and heterogeneity of the target population.
Data collection should adhere to good clinical practices and regulatory standards.
Research protocols should address missing data, quality assurance, and confidentiality.
Early collaboration with the FDA is recommended to align on study designs and regulatory requirements.

Recommendations
Define clear research objectives and determine specific research questions before selecting data collection methods.
Use probability sampling methods whenever feasible to ensure representativeness of the target population.
Address data quality through rigorous planning, data management, and adherence to FDA-supported standards.
Incorporate diverse perspectives by including underrepresented patient populations, tailoring methods to specific subgroups as needed.
Leverage existing data sources, such as patient registries and literature, to complement primary data collection efforts.

Regulatory Considerations
Data submitted to FDA should include clear documentation of the study protocol, intended use, and data collection methodologies.
Researchers must comply with human subject protection regulations (e.g., 21 CFR Parts 50 and 56) and good clinical practice guidelines.
For data intended to support regulatory submissions, adherence to FDA-supported data standards (e.g., CDISC) is strongly encouraged.
Missing data should be addressed through pre-planned strategies and summarized in the study report.
Patient experience data must meet methodological rigor to ensure their reliability and relevance for regulatory decision-making.