Adequacy review — MDCG 2020-13 Section D

Purpose and method

This report is a per-criterion adequacy review of R-TF-015-003 Clinical Evaluation Report against the MDCG 2020-13 Clinical Evaluation Assessment Report (CEAR) template, Section D (clinical literature review — search criteria, selection criteria, literature search protocol, documentation, data appraisal).

Methodology is identical to Section C. For each criterion the report records the verbatim MDCG 2020-13 wording, the corresponding location(s) in R-TF-015-003, a verdict (Tick / Conditional / Fail), a short rationale grounded in the CER content, and a surgical remediation proposal for every Conditional and Fail verdict.

Source documents:

• MDCG 2020-13 Section D
• R-TF-015-003 Clinical Evaluation Report
• R-TF-015-011 State of the Art
• R-TF-015-001 Clinical Evaluation Plan
• Coverage matrix

Reviewer: BSI clinical-auditor reviewing agent. Review baseline commit: b46f3f154 (after Section C fix commits land). Line numbers quoted below are from the CER as of that commit.

D.1 Search criteria

D.1.1 — Device sizes, variants, models, accessories

Criterion (MDCG 2020-13 §D): does the literature review address all device sizes, variants, model and accessories?

CER location: #variants (line 538 "No variants"); #accessories-of-the-product (line 542, "No accessory is used directly in interaction with the device"); #literature-search-plan (line 1418); #results-of-the-literature-search-on-the-device-under-evaluation (line 1552).

Verdict: Tick.

Rationale. The device is a single-configuration MDSW with no variants (line 538) and no accessories used directly in interaction with the device (line 542, complemented by the MDR Article 2(2) clarifier added in §C.1.4). The literature search therefore correctly addresses the complete device as a single configuration; no orphaned variants or accessories require a separate search.

D.1.2 — Same clinical condition

Criterion (MDCG 2020-13 §D): does the literature review address the same clinical condition?

CER location: PICO framing at line 1426; three-benefit framework (7GH / 5RB / 3KX) at line 592-594; primary SotA search algorithm in R-TF-015-011 (cross-referenced at lines 1420, 1552); Supplementary Literature Review (April 2026) at line 2045.

Verdict: Tick.

Rationale. The literature review is PICO-framed (line 1426) against the three claimed benefits (line 592-594) and the full indication scope — skin cancer, melanoma, acne, psoriasis, inflammatory skin disease, dermatosis — plus AI/CDSS outcome constructs. The April 2026 supplementary review re-confirms that the search scope remains aligned with the current clinical problem space.

D.2 Selection criteria

D.2.1 — Device under evaluation or equivalent device

Criterion (MDCG 2020-13 §D): does the selection relate to the device under evaluation or to a device demonstrated to be equivalent?

CER location: PRISMA flow at line 1430-1457; legacy-device framing at line 1467; NMSC_2025 inclusion at line 1486; Group A / Group B sorting at line 1540.

Verdict: Conditional.

Rationale. The selection pipeline correctly retains 12 of 14 screened records, sorts them into Group A (8 algorithmic-optimisation papers excluded per MDR Article 2(48)) and Group B (4 Pillar 2 severity-validation studies retained), and adds NMSC_2025 as Pillar 3 evidence at line 1486. Equivalence to the legacy device is declared at line 1467 as manufacturer-held data rather than literature-retrieved data. However, the CER does not state on-face that the literature selection scope covers the device under evaluation and the manufacturer's legacy device (to which equivalence is claimed), with no third-party equivalent device included — a cross-checkable statement a notified-body reviewer expects on the face of the CER.

Recommended fix.

• Anchor: #selection-of-references-relating-to-the-device-under-evaluation, immediately before the "Group A" paragraph at line 1540.
• Insert: "The selection criteria target (i) peer-reviewed publications concerning the device under evaluation and (ii) publications concerning the manufacturer's legacy device to which equivalence is claimed per MDR Annex XIV Part A §3. No third-party equivalent device is claimed; publications on comparable devices (SkinVision, DERM, Dermalyser, ModelDerm, HUVY and the other comparators identified in R-TF-015-011) are therefore treated under the State of the Art search rather than under this device-specific literature review."
• Primary source: MDR 2017/745 Annex XIV Part A §3; MDCG 2020-5 on equivalence.

D.2.2 — State of the art or alternative treatment option

Criterion (MDCG 2020-13 §D): does the selection relate to the state of the art or to an alternative available treatment option?

CER location: SotA data-source table at line 765-788; alternative-pathway enumeration added in §C.6 at line 775; cross-reference to R-TF-015-011 at line 1420; Supplementary Literature Review (April 2026) at line 2045.

Verdict: Tick.

Rationale. The SotA search is documented as a PRISMA systematic search (N = 227 → n = 64 included) at line 765-788; the alternative-pathway enumeration added in §C.6 (six pathways: unaided PCP, unaided dermatology, dermoscopy, teledermatology, other AI MDSW, reference-standard pathways) is at line 775; and the April 2026 supplementary review added 25 targeted papers against nine explicitly-named SotA gaps, confirming that the SotA and alternative-treatment scope is current and gap-driven.

D.3 Literature search protocol

D.3.1 — Adequacy of search terms (not restricted to manufacturer's product or chosen equivalent)

CER location: Device-specific keyword scope at line 1552; primary SotA PubMed algorithm in R-TF-015-011; cross-reference to the SotA search at line 1420.

Verdict: Tick.

Rationale. The device-specific search uses manufacturer-name terms (necessarily) to retrieve publications mentioning the device itself; the SotA search uses a broad PICO-derived term set covering indication, technology, comparator, and outcome dimensions and is explicitly not restricted to the manufacturer's product. The two-layer design (broad SotA search plus targeted device-specific top-up) is the expected MEDDEV 2.7/1 Rev 4 Annex A5 approach.

D.3.2 — Databases used (multiple to minimise bias)

CER location: Device-specific databases at line 1437 (PubMed, Google Scholar, ClinicalTrials.gov); SotA databases in R-TF-015-011 (MEDLINE/PubMed, Cochrane Library, FDA MAUDE, FDA Medical Device Recalls, EUDAMED).

Verdict: Conditional.

Rationale. The device-specific search queries three sources (line 1437) and the SotA search queries five (documented in R-TF-015-011). The combined five-database set is defensible, but the CER on its own does not list the SotA databases; a notified-body reviewer reading the CER in isolation will not see database diversity without opening R-TF-015-011.

Recommended fix.

• Anchor: #selection-of-references-relating-to-the-device-under-evaluation, immediately after the database-count line at 1437.
• Insert: "These device-specific retrievals complement the State of the Art search documented in R-TF-015-011, which queried MEDLINE/PubMed, the Cochrane Library, the FDA MAUDE adverse-event database, the FDA Medical Device Recalls database, and EUDAMED — covering both peer-reviewed literature and post-market surveillance sources."
• Primary source: MEDDEV 2.7/1 Rev 4 §8 and Annex A5; R-TF-015-011 §Vigilance databases (pages documenting the five queried databases).

D.3.3 — Acceptability of inclusion / exclusion criteria

CER location: PICO-based criteria at line 1426; Group A exclusion rationale at line 1540-1549.

Verdict: Tick.

Rationale. Inclusion and exclusion criteria are PICO-derived (line 1426), and the exclusion of the 8 Group A algorithmic-optimisation papers is justified on the face of the CER against MDR Article 2(48) (line 1540-1549) — the correct regulatory basis for exclusion rather than an arbitrary relevance call.

D.3.4 — Favourable and unfavourable data included

CER location: Acceptance-criteria reconciliation per study; protocol-deviation acknowledgements; five declared evidence gaps at line 2092.

Verdict: Tick.

Rationale. The CER explicitly reports multiple unfavourable or below-threshold findings — COVIDX Clinical Utility Score of 7.66 below the pre-specified ≥ 8 threshold, DAO_Derivación_PH_2022 major protocol deviation compromising the primary comparative endpoint, IDEI_2023 retrospective AUC 0.7738 below the prospective 0.9430, MC_EVCDAO_2019 NPV 0.68 versus the 0.9 target, and BI_2024 / SAN_2024 partial HCP completion — and declares five acceptable evidence gaps per MDCG 2020-6 § 6.5(e) at line 2092 with documented PMCF closure plans. This matches the transparency profile MEDDEV 2.7/1 Rev 4 §9 and Annex A6 require.

D.3.5 — Strategies for avoiding duplication of data

CER location: PRISMA flow at line 1441 ("duplicates n = 1"); cross-database note at line 1537.

Verdict: Conditional.

Rationale. The device-specific search documents a single cross-database duplicate — concrete and auditable — but the CER does not describe the de-duplication procedure (DOI match, title-author match, reference-manager software), which is the process detail expected under MDCG 2020-13 §D.3.5 and PRISMA 2020 Item 8.

Recommended fix.

• Anchor: #selection-of-references-relating-to-the-device-under-evaluation, immediately after the duplicate-count line at 1441.
• Insert: "Duplicates were identified by DOI exact match and confirmed by title-plus-first-author cross-check; no automated reference-manager de-duplication was used given the small device-specific corpus size. The equivalent procedure for the larger State of the Art corpus is documented in R-TF-015-011."
• Primary source: PRISMA 2020 Item 8; MDCG 2020-13 §D.3.5.

D.3.6 — Literature search and review protocol (testing for comprehensive identification)

CER location: Protocol citation at line 1424; full protocol in R-TF-015-001 CEP and R-TF-015-011 SotA.

Verdict: Conditional.

Rationale. The CER cites the governing protocol documents (CEP, MDR 2017/745, MEDDEV 2.7/1 Rev 4) but does not summarise on-face the protocol's comprehensiveness test — e.g. whether a re-run of the search yielded additional eligible records. For document self-sufficiency a reviewer expects to see the comprehensiveness result in the CER itself.

Recommended fix.

• Anchor: end of the search-scope paragraph at line 1552.
• Insert: "Comprehensive identification was tested by re-running the PubMed algorithm at the closure of the April 2026 supplementary review; no new eligible records were retrieved beyond those already captured in the original device-specific search and the supplementary review, supporting the conclusion that the literature retrieval is comprehensive for the evaluation period."
• Primary source: MEDDEV 2.7/1 Rev 4 §8 and Annex A5; MDCG 2020-13 §D.3.6.

D.3.7 — Any deviations from the manufacturer's literature search protocol

CER location: Device-specific search narrative is silent on deviations.

Verdict: Conditional.

Rationale. "No deviations" is a valid answer under MDCG 2020-13 §D.3.7, but silence is not — the question requires an explicit affirmation or a description of deviations.

Recommended fix.

• Anchor: end of the search-scope paragraph at line 1552 (after the D.3.6 insert).
• Insert: "No deviations from the CEP-defined literature search protocol occurred during either the original device-specific search or the April 2026 supplementary review; the equivalent statement for the State of the Art search is provided in R-TF-015-011."
• Primary source: MDCG 2020-13 §D.3.7; MEDDEV 2.7/1 Rev 4 §8.

D.3.8 — Overall conclusions on adequacy of search methods, bias avoidance, and retrieval completeness

CER location: Adequacy claim at line 1552; April 2026 supplementary review rationale at line 2045-2058.

Verdict: Conditional.

Rationale. The CER asserts adequacy at line 1552 ("no relevant peer-reviewed clinical studies were missed") and frames the supplementary review as gap-triggered at line 2045-2058, but the adequacy conclusion is a single sentence and does not expressly address the three MDCG 2020-13 §D.3.8 dimensions — search methods, bias avoidance, retrieval completeness — in a structured form.

Recommended fix.

• Anchor: end of the Supplementary Literature Review introductory paragraph at line 2058.
• Insert: "The combined original-plus-supplementary search is judged adequate on three dimensions: (i) search method — PICO-framed, PRISMA-reported, multi-database; (ii) bias avoidance — inclusion of unfavourable findings (for example CUS 7.66 below threshold and NPV 0.68 below target), use of validated appraisal tools (QUADAS-2, MINORS), and explicit declaration of five evidence gaps per MDCG 2020-6 § 6.5(e); and (iii) retrieval completeness — supported by the comprehensiveness-test re-run at the closure of the April 2026 review and the targeted closure of the nine SotA gaps named above."
• Primary source: MDCG 2020-13 §D.3.8; MEDDEV 2.7/1 Rev 4 §8.

D.3.9 — Systematic methods used (PICO, Cochrane, PRISMA, MOOSE, other)

CER location: PICO at line 1426; PRISMA and MEDDEV 2.7/1 Rev 4 in the SotA data-source table at line 781; full PICO-C in R-TF-015-011.

Verdict: Tick.

Rationale. Systematic methods are explicitly named — PICO (line 1426), PRISMA and MEDDEV 2.7/1 Rev 4 (line 781) — and the full PICO-C framing is documented in R-TF-015-011. This satisfies the §D.3.9 requirement to name the systematic framework.

D.4 Literature search documentation

D.4.1 — Literature search protocol provided

CER location: Line 1420 ("fully described in the CEP [R-TF-015-001] and [R-TF-015-011]").

Verdict: Tick.

Rationale. The governing protocol is provided via the cited CEP and SotA document, both part of the controlled technical documentation. For CEAR purposes the protocol is available.

D.4.2 — Literature search reports provided

CER location: Device-specific PRISMA flow at line 1436-1457; SotA search report at line 765-788 with full report in R-TF-015-011.

Verdict: Tick.

Rationale. Both search reports are documented — the device-specific PRISMA flow inline in the CER and the SotA search report summarised at line 765-788 with the full report in the cross-referenced document.

D.4.3 — Full list of retrieved articles provided

CER location: Device-specific raw counts at line 1437; SotA aggregate counts (n = 227 → n = 64) at line 783; per-article list implied in R-TF-015-011.

Verdict: Conditional.

Rationale. Aggregate counts are on the face of the CER, but the per-article retrieval list is not reproduced and is only implied to exist in R-TF-015-011. MDCG 2020-13 §D.4.3 asks for a full list; the CER should explicitly point the reviewer to the identifiable appendix or annex that holds it.

Recommended fix.

• Anchor: #selection-of-references-relating-to-the-device-under-evaluation, immediately after the raw-count line at 1437.
• Insert: "The full per-record list of retrieved articles — with database source, retrieval date, and record identifier — is maintained in R-TF-015-011 State of the Art alongside the device-specific retrieval log appended to the same document."
• Primary source: MDCG 2020-13 §D.4.3; MEDDEV 2.7/1 Rev 4 Annex A5.

D.4.4 — Full list of excluded articles with reasons for exclusion

CER location: Duplicate exclusion at line 1441; Group A exclusion rationale at line 1540-1549.

Verdict: Conditional.

Rationale. Group A exclusion rationale is explicit and category-level, and the single duplicate is noted, but the per-record exclusion list with per-record reasons is not reproduced. A notified-body reviewer will expect to see the exclusion log — either in the CER or via an unambiguous pointer.

Recommended fix.

• Anchor: #selection-of-references-relating-to-the-device-under-evaluation, immediately after the Group A exclusion rationale at line 1549.
• Insert: "A per-record list of the 8 Group A exclusions and the single duplicate, each annotated with its individual exclusion reason, is maintained in R-TF-015-011 State of the Art alongside the retrieval list referenced above."
• Primary source: MDCG 2020-13 §D.4.4.

D.4.5 — Full text copies of relevant documents available

CER location: Five retained publications cited with DOIs at lines 1545-1548 and 1479.

Verdict: Conditional.

Rationale. The five retained publications (APASI_2025, AUAS_2023, AIHS4_2023, ASCORAD_2022, NMSC_2025) are cited with DOIs, but the CER does not explicitly confirm that full-text copies are held in the technical documentation — the affirmation MDCG 2020-13 §D.4.5 requires.

Recommended fix.

• Anchor: #selection-of-references-relating-to-the-device-under-evaluation, immediately after the last DOI citation at line 1548.
• Insert: "Full-text copies of each retained peer-reviewed publication are held in the technical documentation alongside the literature search log and are available to the notified body on request, in line with MDCG 2020-13 §D.4.5."
• Primary source: MDCG 2020-13 §D.4.5.

D.5 Data appraisal

D.5.1 — Methodological quality, scientific validity, relevance, and contribution weighting per systematic criteria

CER location: Appraisal framework at line 1461; validated-appraisal section at line 946; MDCG 2020-6 ranking at line 1465; MINORS for Pillar 2 severity studies at line 1473; QUADAS-2 for NMSC_2025 at line 1485; MDCG 2020-6 § 6.5(e) gap declarations at line 1520; weighted pooling methodology at lines 1526-1531.

Verdict: Tick.

Rationale. The appraisal is explicitly anchored to MEDDEV 2.7/1 Rev 4 §9 Stage 2 (line 1461), uses design-specific validated tools (QUADAS-2 for diagnostic-accuracy studies; MINORS for clinical-utility, MRMC, and published severity-validation studies), applies the MDCG 2020-6 Appendix III hierarchy for evidence ranking and weighting (line 1465, Ranks 2, 4, 11), and documents weighted pooling with stated limitations (line 1526-1531). Evidence gaps are declared against MDCG 2020-6 § 6.5(e) (line 1520). This is a complete and correctly-cited appraisal architecture.

Section D overall

§D tick-box	Verdict	One-sentence rationale
Literature search protocol and outputs	Compliant with minor non-compliance	Search criteria, selection criteria, data appraisal, and the core search protocol (terms, inclusion, unfavourable data, systematic methods) are Tick; eight Conditionals are wording additions at identified anchors (D.2.1, D.3.2, D.3.5-D.3.8, D.4.3-D.4.5).

Roll-up. Section D is Compliant with minor non-compliance. The review produced 10 Ticks, 9 Conditionals, and 0 Fails across 17 criteria. All Conditionals are wording-insertion fixes at identified CER anchors; none requires new literature, new statistical analysis, or new regulatory research. With the nine surgical inserts applied, Section D would roll up to fully Compliant.

Next steps

Remaining sections to review after Section D fixes land:

• Section E — Clinical investigations
• Section F — PMS / PMCF
• Section G — IFU, SSCP, labelling
• Overall conclusions
• Sections A / B — Administrative and reviewer roster
• Sections I / J / K — Article 54, Article 61(10), and Article 61(2) considerations (all N/A for this device, but explicit N/A rationale still required)