Deep Neural Frameworks Improve the Accuracy of General Practitioners in the Classification of Pigmented Skin Lesions
Open accessThe DOI matches its Crossref record; 20 of the 20 references it cites were found in a registry or an index.
Abstract
This study evaluated whether deep learning frameworks trained in large datasets can help non-dermatologist physicians improve their accuracy in categorizing the seven most common pigmented skin lesions. Open-source skin images were downloaded from the International Skin Imaging Collaboration (ISIC) archive. Different deep neural networks (DNNs) (n = 8) were trained based on a random dataset constituted of 8015 images. A test set of 2003 images was used to assess the classifiers’ performance at low (300 × 224 RGB) and high (600 × 450 RGB) image resolution and aggregated data (age, sex and lesion localization). We also organized two different contests to compare the DNN performance to that of general practitioners by means of unassisted image observation. Both at low and high image resolution, the DNN framework differentiated dermatological images with appreciable performance. In all cases, the accuracy was improved when adding clinical data to the framework. Finally, the least accurate DNN outperformed general practitioners. The physician’s accuracy was statistically improved when allowed to use the output of this algorithmic framework as guidance. DNNs are proven to be high performers as skin lesion classifiers and can improve general practitioner diagnosis accuracy in a routine clinical scenario.
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.
Integrity signals
Registry
Cites
Statistics
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.
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
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
The validity and practicality of sun-reactive skin types I through VI
Fitzpatrick, T. B. (1988) · Archives of Dermatology · doi:10.1001/archderm.1988.01670060015008 · Cited by 4,339
Tschandl, P., Rosendahl, C., & Kittler, H. (2018) · Scientific Data · doi:10.1038/sdata.2018.161 · Cited by 3,628
Classification assessment methods
Tharwat, A. (2021) · Applied Computing and Informatics · doi:10.1016/j.aci.2018.08.003 · Cited by 2,417
Diagnostic accuracy of dermoscopy
Kittler, H., Pehamberger, H., Wolff, K., & Binder, M. (2002) · The Lancet Oncology · doi:10.1016/s1470-2045(02)00679-4 · Cited by 1,349
PH2 - A dermoscopic image database for research and benchmarking
Mendonca, T., Ferreira, P. M., Marques, J. S., Marcal, A. R. S., & Rozeira, J. (2013) · 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc.2013.6610779 · Cited by 1,009 · cited 3×
“Other open skin lesions datasets containing clinical and dermoscopic images from non-Caucasian human skin types are available, however, these are not as large as the HAM10000 or are not publicly available [51].”
BackgroundMethodBrinker, T. J., Hekler, A., Enk, A. H., Klode, J., Hauschild, A., Berking, C., Schilling, B., Haferkamp, S., Schadendorf, D., Holland-Letz, T., et al. (2019) · European Journal of Cancer · doi:10.1016/j.ejca.2019.04.001 · Cited by 605 · cited 1×
“Additionally, a set of other recent studies also reached dermatologist-level skin cancer classification by using DNNs [31-34].”
BackgroundTschandl, P., Codella, N., Akay, B. N., Argenziano, G., Braun, R. P., Cabo, H., Gutman, D., Halpern, A., Helba, B., Hofmann-Wellenhof, R., et al. (2019) · The Lancet Oncology · doi:10.1016/s1470-2045(19)30333-x · Cited by 555 · cited 2×
“[35].”
Deep neural networks are superior to dermatologists in melanoma image classification
Brinker, T. J., Hekler, A., Enk, A. H., Berking, C., Haferkamp, S., Hauschild, A., Weichenthal, M., Klode, J., Schadendorf, D., Holland-Letz, T., et al. (2019) · European Journal of Cancer · doi:10.1016/j.ejca.2019.05.023 · Cited by 415 · cited 1×
“Additionally, a set of other recent studies also reached dermatologist-level skin cancer classification by using DNNs [31-34].”
BackgroundDermoscopy of pigmented skin lesions – a valuable tool for early
Argenziano, G., & Soyer, H. P. (2001) · The Lancet Oncology · doi:10.1016/s1470-2045(00)00422-8 · Cited by 391
Winkler, J. K., Fink, C., Toberer, F., Enk, A., Deinlein, T., Hofmann-Wellenhof, R., Thomas, L., Lallas, A., Blum, A., Stolz, W., et al. (2019) · JAMA Dermatology · doi:10.1001/jamadermatol.2019.1735 · Cited by 365 · cited 1×
“Another important issue is related to the artifacts observed in clinical or dermoscopic images, such as surgical skin markings, dark corners, gel bubbles, superimposed color charts, overlayed rulers, and occluding hair that can affect image classification by automated algorithms [55].”
BackgroundMaron, R. C., Weichenthal, M., Utikal, J. S., Hekler, A., Berking, C., Hauschild, A., Enk, A. H., Haferkamp, S., Klode, J., Schadendorf, D., et al. (2019) · European Journal of Cancer · doi:10.1016/j.ejca.2019.06.013 · Cited by 219 · cited 1×
“In terms of the limitations within our study, firstly, the algorithms were bound to only seven different disease classes which does not reflect clinical reality as many more options should be taken into account when diagnosing [53].”
BackgroundA Benchmark for Automatic Visual Classification of Clinical Skin Disease Images
Sun, X., Yang, J., Sun, M., & Wang, K. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46466-4_13 · Cited by 197
Brinker, T. J., Hekler, A., Hauschild, A., Berking, C., Schilling, B., Enk, A. H., Haferkamp, S., Karoglan, A., von Kalle, C., Weichenthal, M., et al. (2019) · European Journal of Cancer · doi:10.1016/j.ejca.2018.12.016 · Cited by 157 · cited 1×
“Additionally, a set of other recent studies also reached dermatologist-level skin cancer classification by using DNNs [31-34].”
BackgroundSmartphone applications for triaging adults with skin lesions that are suspicious for melanoma
Chuchu, N., Takwoingi, Y., Dinnes, J., Matin, R. N., Bassett, O., Moreau, J. F., Bayliss, S. E., Davenport, C., Godfrey, K., O'Connell, S., et al. (2018) · Cochrane Database of Systematic Reviews · doi:10.1002/14651858.cd013192 · Cited by 144 · cited 2×
“modelderm, MoleMapper, MoleAnalyzer Pro) [43].”
Federman, D. G. (1999) · Archives of Family Medicine · doi:10.1001/archfami.8.2.170 · Cited by 136 · cited 2×
“These results were similar to previously published for nondermatologists [9, 10].”
BackgroundResultKamulegeya, L. H., Okello, M., Bwanika, J. M., Musinguzi, D., Lubega, W., Rusoke, D., Nassiwa, F., & Börve, A. (2019) · doi:10.1101/826057 · Cited by 111 · cited 1×
“Indeed, it was recently reported that a DNN trained on an image dataset composed mainly of Caucasian population skin images (Fitzpatrick 1 and 2 skin types) could not be extrapolated to African-black skin color patients [52].”
BackgroundKamulegeya, L., Bwanika, J., Okello, M., Rusoke, D., Nassiwa, F., Lubega, W., Musinguzi, D., & Börve, A. (2023) · African Health Sciences · doi:10.4314/ahs.v23i2.86 · Cited by 111
Tran, H., Chen, K., Lim, A. C., Jabbour, J., & Shumack, S. (2005) · Australasian Journal of Dermatology · doi:10.1111/j.1440-0960.2005.00189.x · Cited by 92 · cited 1×
“The situation is even worse if we consider that there is a shortage of dermatologists whilst diagnostic accuracy of non-expert clinicians is sensibly below than what is observed with dermatologists, reaching estimate rates between 20 and 40% [9-13].”
BackgroundMoreno, G., Tran, H., Chia, A. L., Lim, A., & Shumack, S. (2007) · Australasian Journal of Dermatology · doi:10.1111/j.1440-0960.2007.00340.x · Cited by 52 · cited 1×
“The situation is even worse if we consider that there is a shortage of dermatologists whilst diagnostic accuracy of non-expert clinicians is sensibly below than what is observed with dermatologists, reaching estimate rates between 20 and 40% [9-13].”
Background
Cited by
Daneshjou, R., Smith, M. P., Sun, M. D., Rotemberg, V., & Zou, J. (2021) · JAMA Dermatology · doi:10.1001/jamadermatol.2021.3129 · Cited by 403
Jones, O. T., Matin, R. N., van der Schaar, M., Prathivadi Bhayankaram, K., Ranmuthu, C. K. I., Islam, M. S., Behiyat, D., Boscott, R., Calanzani, N., Emery, J., et al. (2022) · The Lancet Digital Health · doi:10.1016/s2589-7500(22)00023-1 · Cited by 272
Human-AI interaction in skin cancer diagnosis: a systematic review and meta-analysis
Krakowski, I., Kim, J., Cai, Z. R., Daneshjou, R., Lapins, J., Eriksson, H., Lykou, A., & Linos, E. (2024) · npj Digital Medicine · doi:10.1038/s41746-024-01031-w · Cited by 87
Escalé-Besa, A., Vidal-Alaball, J., Miró Catalina, Q., Gracia, V. H. G., Marin-Gomez, F. X., & Fuster-Casanovas, A. (2024) · Healthcare · doi:10.3390/healthcare12121192 · Cited by 29
“…to the top one or top three diagnoses (Dulmage et al. [41], Liu, Y et al. [22], Muñoz-L ó pez, C et al. [45]), according to type of skin disease with a binary classification (Thomsen, K et al. [25]), and according to the quality of image and/or clinical data acquisition (Lucius, M et al. [44]).”
BackgroundInfluential
Metadata
- Authors
- Lucius, M., De All, J., De All, J. A., Belvisi, M., Radizza, L., Lanfranconi, M., Lorenzatti, V., & Galmarini, C. M.
- Venue
- Diagnostics
- Publisher
- MDPI AG
- Published
- 2020-11-18
- Volume
- 10
- Issue
- 11
- Pages
- 969
- ISSN
- 2075-4418 (ISSN-L)
- OpenAlex
- W3099211739
- Cited by
- 40
- References
- 20
- FWCI
- 1.60
- Type
- journal article
- Language
- en
- Open access
- gold
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
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.