Artificial Intelligence Applied to a First Screening of Naevoid Melanoma: A New Use of Fast Random Forest Algorithm in Dermatopathology
Open accessAbstract
Malignant melanoma (MM) is the "great mime" of dermatopathology, and it can present such rare variants that even the most experienced pathologist might miss or misdiagnose them. Naevoid melanoma (NM), which accounts for about 1% of all MM cases, is a constant challenge, and when it is not diagnosed in a timely manner, it can even lead to death. In recent years, artificial intelligence has revolutionised much of what has been achieved in the biomedical field, and what once seemed distant is now almost incorporated into the diagnostic therapeutic flow chart. In this paper, we present the results of a machine learning approach that applies a fast random forest (FRF) algorithm to a cohort of naevoid melanomas in an attempt to understand if and how this approach could be incorporated into the business process modelling and notation (BPMN) approach. The FRF algorithm provides an innovative approach to formulating a clinical protocol oriented toward reducing the risk of NM misdiagnosis. The work provides the methodology to integrate FRF into a mapped clinical process.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The supervised Fast Random Forest (FRF) algorithm is integrated into clinical protocols and can estimate an alerting threshold to discriminate benign from malignant lesions in the analysis of histological images.
“the supervised Fast Random Forest (FRF) algorithm is integrated into clinical protocols and can estimate an alerting threshold [39,40] to discriminate benign from malignant lesions in the analysis of histological images.”
From Artificial Intelligence in the Histopathological Assessment of Non-Neoplastic Skin Disorders: A Narrative Review with Future Perspectives (2025) · cites this paper as [39]
The Algorithm section identifies FRF as supervised, and the Results report that benign Unna naevi produced values below the proposed 12% alerting threshold for naevoid melanoma. But the Conclusions call the procedure a preliminary clinical protocol requiring larger studies; the paper does not establish that FRF is already integrated into clinical protocols or that its threshold discriminates benign from malignant lesions generally. (LLM, conf=0.86)
Experimental — accuracy not yet measuredCazzato et al.'s fast random forest algorithm flagged potential regions within biopsies using image-based cluster analysis, suggesting FRF-based screening could enhance diagnostic accuracy in challenging nevoid melanoma cases.
“Using image-based cluster analysis, the algorithm flagged potential regions within biopsies, suggesting that FRF-based screening could enhance diagnostic accuracy in challenging NM cases.”
From Artificial Intelligence in Dermatopathology: An Update and Review of the Current Literature (2026) · cites this paper as [16]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
The citation lists haven’t been fetched yet
References 22 · Cited by 18
Research works printed but not linked: 2
- [18]Rosa A. Massaro A. McDermott O. Process Mining Applied to Lean Management Model Improving Decision Making in Healthcare Organizations Proceedings of the 18th International Forum on Knowledge Asset Dynamics Matera, Italy 7–9 June 2023 222 231 978-88-96687-16-1
- [20]Patterson J.W. Weedon’s Skin Pathology 5th ed. Elsevier Amsterdam, The Netherlands 2019
- Authors
- Cazzato, G., Massaro, A., Colagrande, A., Trilli, I., Ingravallo, G., Casatta, N., Lupo, C., Ronchi, A., Franco, R., Maiorano, E., & Vacca, A.
- Affiliations
- University of Bari Aldo Moro (IT)
- LUM Jean Monnet University (IT)
- University of Chieti-Pescara (IT)
- Innovation Department, Diapath S.p.A., Via Savoldini n.71, 24057 Martinengo, Italy
- University of Campania "Luigi Vanvitelli" (IT)
- Corresponding author
- Cazzato, Gerardo
- ORCID
- Venue
- Current Oncology
- Publisher
- MDPI AG
- Published
- 2023-06-23
- Volume
- 30
- Issue
- 7
- Pages
- 6066-6078
- ISSN
- 1198-0052 (ISSN-L)
- 1718-7729
- PMC
- PMC10378276
- OpenAlex
- W4382023412
- Cited by
- 18
- References
- 22
- Citations by year
- 2026: 5 · 2025: 6 · 2024: 6 · 2023: 1
- FWCI
- 2.20
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Current Oncology (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- CINECA IRIS Institutional Research Information System (University of Bari Aldo Moro) (submittedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Archivio Istituzionale della Ricerca - Università degli Studi della Campania "Luigi Vanvitelli" (submittedVersion)
- Topics
- Health Sciences › Medicine › Oncology › Cutaneous Melanoma Detection and Management
- Physical Sciences › Computer Science › Artificial Intelligence › AI in cancer detection
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Keywords
- melanoma misdiagnosis, clinical decision support, clinical protocol development
- MeSH terms
- Random Forest, Melanoma, Cutaneous Malignant, Cutaneous Malignant Melanoma, Algorithms, Artificial Intelligence, Humans, Melanoma
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
