The DOI matches its Crossref record; 82 of the 82 references it cites were found in a registry or an index; 69 of the 120 sources its claims cite were checked for support.
Artificial Intelligence in Dermatopathology: An Update and Review of the Current Literature
Open accessAbstract
Artificial intelligence (AI) has pervaded many fields of medicine in the last few years on the wave of similar changes in other disciplines. Adoption of AI-driven technologies will progress in pathology in the years to come and will also transform our subspecialty of dermatopathology. From the adoption of AI in teaching to its use in clinical practice and in advancing our field through improved research capabilities, we expect a great deal of changes that will hopefully improve our assessment of tissue sections for several cutaneous pathologies. In this review, we offer an overview of where the use of these tools currently stands in dermatopathology and the potential directions that will transform the way we practice and do research. We cover AI’s role in diagnosing various skin conditions, such as melanocytic lesions and other cutaneous skin cancers, and inflammatory dermatoses. The review further covers AI’s contributions to workflow automation, like mitotic figure detection and counting, predictive analytics (e.g., melanoma prognosis), and educational tools (e.g., AI-driven simulators). It also addresses critical technical aspects, including data curation, algorithm development, and model validation. We aim to provide a comprehensive overview of how AI is transforming dermatopathology, from diagnosis and prognosis to education and clinical integration.
Overview
CoChat Summary
## TL;DR
This narrative review surveys how AI is being used in dermatopathology, from diagnosis and workflow automation to prognosis, education, and technical validation. It concludes that AI is promising but still limited by dataset scarcity, generalizability issues, interpretability problems, and the need for stronger validation.
## Key findings
- The review found AI applications across several dermatopathology tasks, including diagnosis of melanocytic lesions, non-melanoma skin cancers, inflammatory dermatoses, workflow automation, prognostication, and educational tools.
- Reported model performances included a basal cell carcinoma framework with 91.4% balanced accuracy on about 1400 image patches, and a cutaneous adnexal tumor CNN with 89.92% overall classification accuracy across 248 samples.
- In nevi-versus-melanoma work, the paper reports examples where an ensemble CNN achieved AUCs as high as 0.97, and another CNN study reported 94% sensitivity, 90% specificity, and 92% accuracy, compared with dermatopathologists’ 88.88% sensitivity, 91.77% specificity, and 90.33% accuracy.
- For workflow automation, an AI mitosis detector on 99 digitized melanocytic lesion slides improved concordance in nevoid melanoma from 68% to 75%, but overall concordance stayed around 90% and false positives remained a problem.
## Methods
- This is a narrative review of the literature, not a systematic review.
- The authors performed a non-systematic PubMed search for studies published between 1985 and 2026, using keywords related to dermatopathology, histopathology, artificial intelligence, machine learning, deep learning, and multimodal large language models, plus manual screening of references.
- They included studies based on relevance to AI applications in dermatopathology, with emphasis on histopathology-based investigations, and did not apply formal inclusion/exclusion criteria or a systematic quality appraisal.
## Limitations
- The paper explicitly states that, because it is a narrative review, no formal inclusion/exclusion criteria or systematic quality appraisal framework was applied.
- The paper states that many models remain investigational and require further validation with large multi-center external cohorts, prospective studies, calibration analyses, and real-world utility assessments.
- Evident from the text: the review synthesizes a broad and heterogeneous literature, so its conclusions depend on the underlying studies’ variable datasets, metrics, and validation approaches rather than on a single standardized analysis.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures

Verification
Claims the paper makes
60 Valid · 2 Not valid · 46 Not checked · 56 Own results and interpretations
Every source the paper cites for this claim was checked and supports it.
In the paper
“Hybrid models are utilized in certain studies to couple CNN-based feature extraction with traditional machine learning classifiers, thereby improving diagnostic performance [21].”
§ Artificial Intelligence in Dermatopathology: An Update and Review of the Current Literature > 3. AI Applications in Dermatopathology: Current Landscape > 3.1. Diagnostic Support for Specific Skin Conditions
What the sources say
- Supportedcites [21] Nogales, A., Garrido, M.C., Guitian, A., Rodriguez-Peralto, J.-L., Villanueva, C.P., Díaz-Prieto, D., and García-Tejedor, Á.J. (2025). A Hybrid Artificial Intelligence Framework for Melanoma Diagnosis Using Histopathological Images. Technologies, 13. source
The Methods describe convolutional layers extracting image features, which are then passed to classical classifiers including KNN and random forest. The Results identify those as the best-performing classifiers, and the Discussion says the hybrid model outperformed other CNN models for classifying melanoma and nevi. The supplied passages do not give page numbers. (LLM, conf=0.93)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“Across three dermatopathology centers, the AI model achieved sensitivities ranging from ~97.7% to 98.6% and specificities around 96.8–98.7%.”
§ Artificial Intelligence in Dermatopathology: An Update and Review of the Current Literature > 3. AI Applications in Dermatopathology: Current Landscape > 3.1. Diagnostic Support for Specific Skin Conditions
What the sources say
- Supportedcites [19] Duschner, 2023, Applying an artificial intelligence deep learning approach to routine dermatopathological diagnosis of basal cell carcinoma, J. Dtsch. Dermatol. Ges. source
The Abstract identifies the model as U-Net-based and reports BCC-detection sensitivities of 98.23%, 97.67%, and 98.57%, with specificities of 98.51%, 96.77%, and 98.73%, across three centers. The “Subtype classification and automated measurement of tumor thickness” section also confirms that the algorithm was trained to distinguish BCC subtypes. The quoted sensitivity and specificity ranges refer to BCC detection, not subtype-classification accuracy. (LLM, conf=0.94)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“a 2024 study introduced an artificial intelligence platform, ArcticAI, which is designed to enhance intraoperative margin assessment during Mohs micrographic surgery for BCC [24].”
§ Artificial Intelligence in Dermatopathology: An Update and Review of the Current Literature > 3. AI Applications in Dermatopathology: Current Landscape > 3.1. Diagnostic Support for Specific Skin Conditions
What the sources say
- Supportedcites [24] Levy, 2024, Intraoperative margin assessment for basal cell carcinoma with deep learning and histologic tumor mapping to surgical site, npj Precis. Oncol. source
The Introduction says the authors developed ArcticAI to improve tissue preparation, histological inspection, and tumor mapping, using Mohs micrographic surgery for basal cell carcinoma as the model system. The Methods (Technology overview) identifies it as an AI-based platform for rapid tumor-margin assessment. (LLM, conf=0.94)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“The AI flagged numerous false positives (e.g., pigmented or keratinocyte nuclei), underscoring the need for cautious interpretation.”
§ Artificial Intelligence in Dermatopathology: An Update and Review of the Current Literature > 3. AI Applications in Dermatopathology: Current Landscape > 3.2. Workflow Optimization and Automation
What the sources say
- Partly supportedcites [32] Sturm, B., Creytens, D., Smits, J., Ooms, A.H.A.G., Eijken, E., Kurpershoek, E., Küsters-Vandevelde, H.V.N., Wauters, C., Blokx, W.A.M., and van der Laak, J.A.W.M. (2022). Computer-Aided Assessment of Melanocytic Lesions by Means of a Mitosis Algorithm. Diagnostics, 12. sourceExperimental — accuracy not yet measured
The Materials and Methods (§2.2) and Discussion (§4) report numerous false-positive detections, including melanin pigment and several kinds of nuclei. But the Results (§3) describe detections in keratinocytes as non-lesional mitoses, not keratinocyte nuclei falsely identified as mitoses. The paper also cautions against misinterpreting such detections. (LLM, conf=0.87)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“in nevoid melanoma cases, the algorithm improved concordance from 68% to 75%.”
§ Artificial Intelligence in Dermatopathology: An Update and Review of the Current Literature > 3. AI Applications in Dermatopathology: Current Landscape > 3.2. Workflow Optimization and Automation
What the sources say
- Supportedcites [32] Sturm, B., Creytens, D., Smits, J., Ooms, A.H.A.G., Eijken, E., Kurpershoek, E., Küsters-Vandevelde, H.V.N., Wauters, C., Blokx, W.A.M., and van der Laak, J.A.W.M. (2022). Computer-Aided Assessment of Melanocytic Lesions by Means of a Mitosis Algorithm. Diagnostics, 12. source
The Abstract reports that concordance with the consensus diagnosis in the 10 nevoid melanoma cases increased with AI assistance, from 68% without it to 75% with it. This is concordance among pathologists using the mitosis algorithm, not a standalone algorithm diagnosis. (LLM, conf=0.95)
plainstates a causen = 8- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on an assumption not shown here.
- Supported
63 more claims · 18 with their cited sources checked
Create a free account to see all 164 claims and every source verdict.
- It may state two separate claims as one. 69
- Probably not a substantial claim. 51
- The quote only partly states this claim. 1
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
“The researchers trained a model on 595 hematoxylin and eosin-stained images of melanocytic lesions (350 nevi and 345 melanomas). The model’s performance was directly compared with that of 11 practicing pathologists using a separate test set of 100 image sections (1:1 ratio of melanomas to nevi).”
§ Artificial Intelligence in Dermatopathology: An Update and Review of the Current Literature > 3. AI Applications in Dermatopathology: Current Landscape > 3.1. Diagnostic Support for Specific Skin Conditions
Reported 595 · recomputed 695
Citations
Cites
QuPath: Open source software for digital pathology image analysis
Bankhead, P., Loughrey, M. B., Fernández, J. A., Dombrowski, Y., McArt, D. G., Dunne, P. D., McQuaid, S., Gray, R. T., Murray, L. J., Coleman, H. G., et al. (2017) · Scientific Reports · doi:10.1038/s41598-017-17204-5 · Cited by 8,908 · cited 2×
“Emerging solutions include standardized open-source tools (e.g., QuPath) [44], collaborative multi-institutional annotation efforts, and AI-assisted annotation workflows that reduce manual burden [45].”
Decision curve analysis: a novel method for evaluating prediction models
Vickers, A. J., & Elkin, E. B. (2006) · Medical Decision Making · doi:10.1177/0272989x06295361 · Cited by 6,341 · cited 1×
“Decision curve analysis may further help determine whether AI-assisted predictions provide meaningful net clinical benefit compared with conventional diagnostic strategies [75].”
Calibration: the Achilles heel of predictive analytics
On behalf of Topic Group ‘Evaluating diagnostic tests and prediction models’ of the STRATOS initiative, Van Calster, B., McLernon, D. J., van Smeden, M., Wynants, L., & Steyerberg, E. W. (2019) · BMC Medicine · doi:10.1186/s12916-019-1466-7 · Cited by 2,650 · cited 2×
“Increasing emphasis has therefore been placed on model calibration, which examines whether predicated probabilities correspond to true clinical risk [73].”
Hay, R. J., Johns, N. E., Williams, H. C., Bolliger, I. W., Dellavalle, R. P., Margolis, D. J., Marks, R., Naldi, L., Weinstock, M. A., Wulf, S. K., et al. (2014) · Journal of Investigative Dermatology · doi:10.1038/jid.2013.446 · Cited by 1,706 · cited 1×
“Among the challenges in this field are the vast range of skin pathologies, with more than 2000 recognized skin diseases spanning from inflammatory dermatoses, infections, deposit disease, and neoplastic lesions [1].”
Explainable artificial intelligence (XAI) in deep learning-based medical image analysis
van der Velden, B. H. M., Kuijf, H. J., Gilhuijs, K. G. A., & Viergever, M. A. (2022) · Medical Image Analysis · doi:10.1016/j.media.2022.102470 · Cited by 1,328 · cited 3×
“Efforts such as XAI are essential for translating deep learning models into clinically usable tools [65].”
InfluentialHuman-computer collaboration for skin cancer recognition
Tschandl, P., Rinner, C., Apalla, Z., Argenziano, G., Codella, N., Halpern, A., Janda, M., Lallas, A., Longo, C., Malvehy, J., et al. (2020) · Nature Medicine · doi:10.1038/s41591-020-0942-0 · Cited by 1,024 · cited 2×
“Annotation provides critical ground truth for supervised learning but is limited by time demands and inter-observer variability [43].”
History of artificial intelligence in medicine
Kaul, V., Enslin, S., & Gross, S. A. (2020) · Gastrointestinal Endoscopy · doi:10.1016/j.gie.2020.06.040 · Cited by 1,011 · cited 2×
“This shift enabled more clinical applications across different specialties, including radiology, pathology, and gastroenterology [10].”
Artificial intelligence in medicine
Ramesh, A., Kambhampati, C., Monson, J., & Drew, P. (2004) · Annals of The Royal College of Surgeons of England · doi:10.1308/147870804290 · Cited by 991
Tellez, D., Litjens, G., Bándi, P., Bulten, W., Bokhorst, J.-M., Ciompi, F., & van der Laak, J. (2019) · Medical Image Analysis · doi:10.1016/j.media.2019.101544 · Cited by 649 · cited 1×
“Techniques such as rotation, scaling, color normalization, or stain augmentation help simulate variability encountered in real-world slides [57].”
Artificial Intelligence and Digital Pathology: Challenges and Opportunities
Tizhoosh, H. R., & Pantanowitz, L. (2018) · Journal of Pathology Informatics · doi:10.4103/jpi.jpi_53_18 · Cited by 544 · cited 1×
“Additionally, interoperability with existing workflows, such as integration with laboratory information systems (LISs), is essential for practical use but technically complex [79].”
Cited by
Chatziioannidis, D., Tsintotas, K. A., & Tzekis, P. (2026) · Electronics · doi:10.3390/electronics15153346 · Cited by 1
Metadata
- Authors
- Abu-Dayeh, A., Cazzato, G., & Giubellino, A.
- Affiliations
- Department of Laboratory Medicine and Pathology, University of Minnesota, Minneapolis, MN 55414, USA
- University of Bari Aldo Moro (IT)
- University of Minnesota (US)
- Corresponding author
- Giubellino, Alessio
- Venue
- Diagnostics
- Publisher
- MDPI AG
- Published
- 2026-06-01
- Volume
- 16
- Issue
- 11
- Pages
- 1702
- ISSN
- 2075-4418 (ISSN-L)
- PMC
- PMC13257086
- OpenAlex
- W7163067755
- Semantic Scholar
- 0441ab66d206bb8546498c9c17edda08f4597a70
- Cited by
- 1
- References
- 78
- Citations by year
- 2026: 1
- FWCI
- 0.97
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Diagnostics (publishedVersion, cc-by)
- PubMed (publishedVersion)
- PubMed Central (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › AI in cancer detection
- Health Sciences › Medicine › Oncology › Cutaneous Melanoma Detection and Management
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Keywords
- artificial intelligence, dermatopathology, skin cancer diagnosis, melanocytic lesions, melanoma prognosis, inflammatory skin diseases, mitotic figure detection, workflow automation, model validation, data curation, algorithm development, pathology education
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-09, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-09 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Artificial Intelligence in Dermatopathology: An Update and Review of the Current Literature. https://app.cochat.ai/papers/10.3390/diagnostics16111702
- The paper
- Abu-Dayeh, A., Cazzato, G., & Giubellino, A. (2026). Artificial Intelligence in Dermatopathology: An Update and Review of the Current Literature. Diagnostics. https://doi.org/10.3390/diagnostics16111702
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.