Response

We acknowledge that the Clinical Benefits and Performance Claims section of the IFU lacked sufficient context for a user to fully understand the applicability of the reported metrics. Specifically, the 3-character clinical benefit identifiers were not defined in a glossary, the study reference codes were not accompanied by full bibliographic references, and the relationship between the reported metrics and the device's clinical scope was not explicitly stated. The IFU has been updated to address each of these gaps.

a. Tag definitions (clinical benefit identifiers). A new "Clinical Benefit Identifiers" section has been added to the IFU, Clinical User Manual, Clinical Benefits and Performance Claims page. This section provides a glossary table that maps each 3-character code (7GH, 3KX, 8PL, 1QF, 9VW, 5RB, 0ZC) to its full clinical benefit description. The codes are unique, non-sequential identifiers used consistently across the IFU, the technical file, clinical evaluation, and risk management documentation to ensure unambiguous traceability.

b. Study references. A new Clinical Validation Studies section has been added to the IFU, Clinical User Manual, immediately following the Clinical Benefits and Performance Claims section. This section provides full bibliographic references for all clinical validation studies cited alongside performance metrics, including: study title, principal investigator(s), investigational site(s), sample size, study period (start to completion), device version tested, and publication status. Each study code that appears as a badge in the performance claims (e.g., IDEI_2023, BI_2024) is now defined with its complete study metadata in this reference section. Additionally, a study-to-benefit cross-reference table has been included so that a user can see at a glance which clinical benefits each study supports and the number of performance claims it contributes.

c. ICD conditions and diagnostic features. A new "How to Read the Performance Claims" subsection has been added to the IFU, Clinical User Manual, Clinical Benefits and Performance Claims page. This subsection clarifies the following:

• The device always outputs a probability distribution across all validated ICD-11 categories for every image processed. It does not diagnose specific conditions; it provides an interpretive distribution representation of possible ICD categories to support clinical decision-making. The full list of ICD-11 categories covered by the device is specified in IFU, General Information, Intended Use.
• The performance metrics measure the improvement in healthcare professional diagnostic accuracy, referral precision, or severity assessment when using the device's distributional output, compared to performance without the device.
• The "study population" shown for each metric (e.g., "Multiple conditions", "Rare diseases", "Melanoma") indicates the clinical context in which the validation study was conducted — the composition of images used in the study.

d. Device function vs clinical benefit. A point of clarification on the relationship between the device's output and the condition-specific clinical benefits claimed: the device output mechanism is uniform — every image receives the same full ICD-11 probability distribution, regardless of the condition presented. There is no condition-specific mode or algorithm pathway. However, the clinical benefits listed (e.g., 9VW: improved accuracy for rare diseases; 1QF: improved accuracy for lesions suspicious for skin cancer) are condition-specific because healthcare professionals' baseline diagnostic accuracy differs across condition categories. For example, HCPs have lower baseline accuracy for rare dermatological conditions, so the improvement attributable to the device is proportionally greater in that clinical context. The device's function is uniform; the clinical benefit realised by the user is context-dependent. The updated IFU makes this distinction explicit in the "How to Read the Performance Claims" subsection.

GSPR compliance. The corrective actions described above address each of the cited requirements as follows:

• GSPR 15.1: The updated IFU clearly communicates the accuracy and precision of the decision-support information provided by the device. Performance metrics are presented with their clinical context, study provenance, full bibliographic traceability, and a cross-reference between studies and clinical benefits. A healthcare professional reading the IFU can now understand what each metric measures, which study generated it, and the clinical context in which it was validated.
• GSPR 23.4(c): Performance limitations and residual risks are now explicitly communicated per indication category. Each performance metric specifies the study population, user group, and achieved values with confidence intervals, enabling users to understand the conditions under which the stated performance applies.
• GSPR 23.4(e): The accuracy limitations per condition category serve as precautions for clinical decision-making. The "How to Read the Performance Claims" subsection explicitly states that the study population labels indicate the clinical context of the validation, and that baseline HCP accuracy differs across condition categories — meaning the device's contribution to clinical accuracy varies accordingly.
• GSPR 23.4(h): Each performance metric now identifies the study that generated it (with full bibliographic reference), the user group for whom it was validated (primary care practitioners, dermatologists, or both), and the clinical context of the study population. The study-to-benefit cross-reference table provides an additional layer of traceability between studies and the clinical benefits they substantiate.

Relevant commits

• a0a52d704 (2026-03-01, Taig Mac Carthy): Added ClinicalBenefitsGlossary component (glossary table mapping 7 clinical benefit codes to descriptions) and ClinicalStudiesReference component (bibliography-style reference list for completed studies). Created new IFU page clinical-validation-studies.mdx. Updated IFU clinical-benefits-and-performance-claims.mdx with "Clinical Benefit Identifiers" glossary section and "How to Read the Performance Claims" interpretive subsection explaining the distributional nature of the device output.
• 426db5a4f (2026-03-01, Taig Mac Carthy): Added StudyBenefitCrossReference component (cross-reference table mapping each study to the clinical benefits it supports). Reworded the "study population" bullet in IFU clinical-benefits-and-performance-claims.mdx to distinguish uniform device output mechanism from context-dependent clinical benefit. Updated es/pt localisations with glossary, interpretive text, and cross-reference sections.