CoChat

AI outperformed every dermatologist in dermoscopic melanoma diagnosis, using an optimized deep-CNN architecture with custom mini-batch logic and loss function

Open access

The DOI matches its Crossref record; 28 of the 28 references it cites were found in a registry or an index.

Abstract

Melanoma, one of the most dangerous types of skin cancer, results in a very high mortality rate. Early detection and resection are two key points for a successful cure. Recent researches have used artificial intelligence to classify melanoma and nevus and to compare the assessment of these algorithms to that of dermatologists. However, training neural networks on an imbalanced dataset leads to imbalanced performance, the specificity is very high but the sensitivity is very low. This study proposes a method for improving melanoma prediction on an imbalanced dataset by reconstructed appropriate CNN architecture and optimized algorithms. The contributions involve three key features as custom loss function, custom mini-batch logic, and reformed fully connected layers. In the experiment, the training dataset is kept up to date including 17,302 images of melanoma and nevus which is the largest dataset by far. The model performance is compared to that of 157 dermatologists from 12 university hospitals in Germany based on the same dataset. The experimental results prove that our proposed approach outperforms all 157 dermatologists and achieves higher performance than the state-of-the-art approach with area under the curve of 94.4%, sensitivity of 85.0%, and specificity of 95.0%. Moreover, using the best threshold shows the most balanced measure compare to other researches, and is promisingly application to medical diagnosis, with sensitivity of 90.0% and specificity of 93.8%. To foster further research and allow for replicability, we made the source code and data splits of all our experiments publicly available.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41598-021-96707-8

Cites

28 of 28 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Proposed melanoma prediction system (this figure was created with Draw.io).
Figure 1 Proposed melanoma prediction system (this figure was created with Draw.io).CC BY · PubMed Central
Figure 2 Dense connections in DenseNet (this figure was created with Draw.io).
Figure 2 Dense connections in DenseNet (this figure was created with Draw.io).CC BY · PubMed Central
Figure 3 Triangular cycle decreases the cycle amplitude by half after each period, while keeping the base lr constant (this figure was created with Draw.io).
Figure 3 Triangular cycle decreases the cycle amplitude by half after each period, while keeping the base lr constant (this figure was created with Draw.io).CC BY · PubMed Central
Figure 4 Loss of ORI, BON, and BLF models during training and on the validation dataset.
Figure 4 Loss of ORI, BON, and BLF models during training and on the validation dataset.CC BY · PubMed Central
Figure 5 Sensitivity and specificity of ORI, BON, and BLF models during training and on the validation dataset.
Figure 5 Sensitivity and specificity of ORI, BON, and BLF models during training and on the validation dataset.CC BY · PubMed Central
Figure 6 Performance evaluated of three best models of ORI, BON, and BLF over Test-10 dataset using a prediction threshold of 0.5.
Figure 6 Performance evaluated of three best models of ORI, BON, and BLF over Test-10 dataset using a prediction threshold of 0.5.CC BY · PubMed Central
Figure 7 The receiver operating characteristic curves of ORI, BON, and BLF over the Test-10 dataset.
Figure 7 The receiver operating characteristic curves of ORI, BON, and BLF over the Test-10 dataset.CC BY · PubMed Central
Figure 8 Performance of ORI, BON, and BLF over the MClass-D dataset using a prediction threshold of 0.5.
Figure 8 Performance of ORI, BON, and BLF over the MClass-D dataset using a prediction threshold of 0.5.CC BY · PubMed Central
Figure 9 The receiver operating characteristic curves of ORI, BON, and BLF over the MClass-D dataset.
Figure 9 The receiver operating characteristic curves of ORI, BON, and BLF over the MClass-D dataset.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 28 references

Research works printed but not linked: 1

  1. Codella, N. C. et al. International symposium on biomedical imaging (ISBI), hosted by the international skin imaging collaboration (ISIC). in 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018). 168–172 (IEEE, 2017).

Cited by

See all 123

Metadata

Authors
Pham, T.-C., Luong, C.-M., Hoang, V.-D., & Doucet, A.
Affiliations
  • Thuyloi University (VN)
  • Trường Đại học Khoa học và Công nghệ Hà Nội (VN)
  • Ho Chi Minh City University of Technology and Engineering (VN)
  • Laboratoire Informatique, Image et Interaction (L3i) (FR)
Corresponding author
Doucet, Antoine
Venue
Scientific Reports
Published
2021-09-01
Volume
11
Issue
1
Pages
17485-17485
ISSN
2045-2322 (ISSN-L)
OpenAlex
W3198806597
Cited by
101
References
28
Citations by year
2026: 13 · 2025: 23 · 2024: 27 · 2023: 24 · 2022: 13 · 2021: 1
FWCI
5.91
Type
journal article
Language
en
Open access
Licence
Locations
  • Scientific Reports (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Greater South Information System
Topics
  • Health Sciences › Medicine › Oncology › Cutaneous Melanoma Detection and Management
  • Physical Sciences › Computer Science › Artificial Intelligence › AI in cancer detection
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
Keywords
melanoma diagnosis, dermoscopy, convolutional neural networks, class imbalance, custom loss function, dermatologist comparison, early detection of skin cancer, sensitivity and specificity
MeSH terms
Dermatologists, Deep Learning, Algorithms, Artificial Intelligence, Humans, Melanoma, ROC Curve, Skin Neoplasms, Neural Networks, Computer, Dermoscopy

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

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: AI outperformed every dermatologist in dermoscopic melanoma diagnosis, using an optimized deep-CNN architecture with custom mini-batch logic and loss function. https://app.cochat.ai/papers/10.1038/s41598-021-96707-8
The paper
Pham, T.-C., Luong, C.-M., Hoang, V.-D., & Doucet, A. (2021). AI outperformed every dermatologist in dermoscopic melanoma diagnosis, using an optimized deep-CNN architecture with custom mini-batch logic and loss function. Scientific Reports. https://doi.org/10.1038/s41598-021-96707-8

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.