CIP field extract — MAN_2025 → CEP row

Internal traceability artefact for task-3b13. The three fields pulled from R-TF-015-004 (MAN_2025 instance) and R-TF-015-006 (MAN_2025 instance) and summarised into the CEP Confirmatory Phase row at R-TF-015-001 line 1212. Written 2026-04-21 as part of the CEP↔CIP/CIR factual-alignment pull-through.

Source files and line references

• CIP (source of methodology): apps/qms/docs/legit-health-plus-version-1-1-0-0/product-verification-and-validation/clinical/Investigation/man-2025/r-tf-015-004.mdx
• CIR (source of actual execution facts): apps/qms/docs/legit-health-plus-version-1-1-0-0/product-verification-and-validation/clinical/Investigation/man-2025/r-tf-015-006.mdx

Field 1 — Sample-size rationale

CIP anchor: R-TF-015-004 MAN_2025 instance, §Sample size / §Power calculation (lines 152–166).

• Test: two-sided McNemar paired-binary test
• Baseline (unassisted) accuracy: 60 %
• Target (assisted) accuracy: 70 %
• Pre-specified effect size (MCID): +10 percentage points
• Significance level: α = 0.05 (two-sided)
• Power: 1 − β = 0.80
• Discordant-proportion assumption: 25 %
• Required paired observations: ≈ 200 (Lachin, 1992)
• Conservative floor: 5 readers × 149 images = 745 paired observations
• Effective-sample-size at ICC ≈ 0.15: ≈ 460 independent observations
• Robustness: conclusion preserved across ICC 0.05 – 0.30

CEP summary sentence used: "Sample-size rationale: two-sided McNemar paired-binary power calculation (baseline 60 %, target 70 %, α = 0.05, 1 − β = 0.80, 25 % discordant proportion) requiring approximately 200 paired observations; the conservative 5-reader × 149-image floor provides 745 paired observations (effective N ≈ 460 at ICC 0.15). Full derivation in R-TF-015-004 §Sample size."

Field 2 — Reference standard (ground truth)

CIP anchor: R-TF-015-004 MAN_2025 instance, §Reference standard (ground truth) (lines 136–140).

• Scheme: published atlas diagnosis, encoded as ICD-11
• Independence: established prior to and independently of the investigation
• Methodology consistency: identical to the reference-standard methodology applied in SAN_2024, BI_2024 and PH_2024
• Declared limitations: atlas diagnoses may not all be histopathologically confirmed
• Limitation mitigations: (i) 149-image set size limits influence of any individual misclassification, (ii) self-controlled design exposes both arms to any reference-standard error equally, (iii) methodology identical to the three source MRMC investigations

CEP summary sentence used: "Reference standard: published atlas diagnosis encoded as ICD-11, established prior to and independently of the investigation, consistent with the methodology of SAN_2024, BI_2024 and PH_2024; see R-TF-015-004 §Reference standard (ground truth)."

Field 3 — Committed timeline

CIP anchor: R-TF-015-004 MAN_2025 instance, §Duration (lines 205–207). CIR anchor: R-TF-015-006 MAN_2025 instance, §Results — Initiation and completion (lines 228–230).

• Planning phase: began October 2025
• Data-collection phase: 21 January 2026 to 17 April 2026
• Database closure / data lock: 17 April 2026
• Study-report date: recorded dynamically on the CIR's <Signature /> block (no fixed calendar date in the MDX source)

CEP row values used:

• Initiation date: January 21, 2026
• Inclusion period: January 21, 2026 to April 17, 2026
• Completion date: April 17, 2026
• Date of the study report: as recorded on the signature block of the signed Clinical Investigation Report (R-TF-015-006 MAN_2025 instance).

Why initiation is set to 21 January 2026 rather than October 2025: every neighbouring completed row in the CEP's Confirmatory Phase table (SAN_2024, BI_2024, PH_2024, MC_EVCDAO_2019, IDEI_2023) uses the data-collection-start date as the initiation date. Using the planning-phase start would break convention.

Reader cohort

CIR §Study population characteristics (lines 240–244).

• Enrolled: 19 HCPs
• Screen failures: 3 (R-03, R-11, R-15)
• Primary analysis cohort: 16 HCPs

The CIP-time aspirational "≥ 5 dermatologists" phrasing in the pre-edit CEP row is replaced with the CIR's actual enrolment figures.

Integrity checks applied in the edit

• Pillar 3 §4.4 at Rank 11 framing preserved verbatim — no shift to Pillar 2 or to a different rank.
• Primary / secondary objective and acceptance-criteria cells left unchanged — the study's claim and thresholds are CIP-level and do not move under the state change.
• All cross-references in the CEP row point to audit-visible identifiers (R-TF-015-004 MAN_2025 instance, R-TF-015-006 MAN_2025 instance) — no internal paths, no task-folder references, no CLAUDE.md names.