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

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...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Tri-Cong Pham, Chi-Mai Luong, Van-Dung Hoang, Antoine Doucet
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2021
Materias:
R
Q
Acceso en línea:https://doaj.org/article/cf0870ad10214c32b722e2edc72aa7af
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:cf0870ad10214c32b722e2edc72aa7af
record_format dspace
spelling oai:doaj.org-article:cf0870ad10214c32b722e2edc72aa7af2021-12-02T16:38:48ZAI outperformed every dermatologist in dermoscopic melanoma diagnosis, using an optimized deep-CNN architecture with custom mini-batch logic and loss function10.1038/s41598-021-96707-82045-2322https://doaj.org/article/cf0870ad10214c32b722e2edc72aa7af2021-09-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-96707-8https://doaj.org/toc/2045-2322Abstract 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.Tri-Cong PhamChi-Mai LuongVan-Dung HoangAntoine DoucetNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-13 (2021)
institution DOAJ
collection DOAJ
language EN
topic Medicine
R
Science
Q
spellingShingle Medicine
R
Science
Q
Tri-Cong Pham
Chi-Mai Luong
Van-Dung Hoang
Antoine Doucet
AI outperformed every dermatologist in dermoscopic melanoma diagnosis, using an optimized deep-CNN architecture with custom mini-batch logic and loss function
description 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.
format article
author Tri-Cong Pham
Chi-Mai Luong
Van-Dung Hoang
Antoine Doucet
author_facet Tri-Cong Pham
Chi-Mai Luong
Van-Dung Hoang
Antoine Doucet
author_sort Tri-Cong Pham
title AI outperformed every dermatologist in dermoscopic melanoma diagnosis, using an optimized deep-CNN architecture with custom mini-batch logic and loss function
title_short AI outperformed every dermatologist in dermoscopic melanoma diagnosis, using an optimized deep-CNN architecture with custom mini-batch logic and loss function
title_full AI outperformed every dermatologist in dermoscopic melanoma diagnosis, using an optimized deep-CNN architecture with custom mini-batch logic and loss function
title_fullStr AI outperformed every dermatologist in dermoscopic melanoma diagnosis, using an optimized deep-CNN architecture with custom mini-batch logic and loss function
title_full_unstemmed AI outperformed every dermatologist in dermoscopic melanoma diagnosis, using an optimized deep-CNN architecture with custom mini-batch logic and loss function
title_sort ai outperformed every dermatologist in dermoscopic melanoma diagnosis, using an optimized deep-cnn architecture with custom mini-batch logic and loss function
publisher Nature Portfolio
publishDate 2021
url https://doaj.org/article/cf0870ad10214c32b722e2edc72aa7af
work_keys_str_mv AT tricongpham aioutperformedeverydermatologistindermoscopicmelanomadiagnosisusinganoptimizeddeepcnnarchitecturewithcustomminibatchlogicandlossfunction
AT chimailuong aioutperformedeverydermatologistindermoscopicmelanomadiagnosisusinganoptimizeddeepcnnarchitecturewithcustomminibatchlogicandlossfunction
AT vandunghoang aioutperformedeverydermatologistindermoscopicmelanomadiagnosisusinganoptimizeddeepcnnarchitecturewithcustomminibatchlogicandlossfunction
AT antoinedoucet aioutperformedeverydermatologistindermoscopicmelanomadiagnosisusinganoptimizeddeepcnnarchitecturewithcustomminibatchlogicandlossfunction
_version_ 1718383600973381632