Intelligent skin cancer diagnosis using improved particle swarm optimization and deep learning models. / Tan, Teck Yan; Zhang, Li; Lim, Chee Peng.

In: Applied Soft Computing, Vol. 84, 105725, 11.2019.

Research output: Contribution to journalArticlepeer-review

Published

Standard

Intelligent skin cancer diagnosis using improved particle swarm optimization and deep learning models. / Tan, Teck Yan; Zhang, Li; Lim, Chee Peng.

In: Applied Soft Computing, Vol. 84, 105725, 11.2019.

Research output: Contribution to journalArticlepeer-review

Harvard

APA

Vancouver

Author

Tan, Teck Yan ; Zhang, Li ; Lim, Chee Peng. / Intelligent skin cancer diagnosis using improved particle swarm optimization and deep learning models. In: Applied Soft Computing. 2019 ; Vol. 84.

BibTeX

@article{00cbf1d985c1454ab1c70f68719f912e,
title = "Intelligent skin cancer diagnosis using improved particle swarm optimization and deep learning models",
abstract = "In this research, we propose an intelligent decision support system for skin cancer detection. Since generating an effective lesion representation is a vital step to ensure the success of lesion classification, the discriminative power of different types of features is exploited. Specifically, we combine clinically important asymmetry, border irregularity, colour and dermoscopic structure features with texture features extracted using Grey Level Run Length Matrix, Local Binary Patterns, and Histogram of Oriented Gradients operators for lesion representation. Then, we propose two enhanced Particle Swarm Optimization (PSO) models for feature optimization. The first model employs adaptive acceleration coefficients, multiple remote leaders, in-depth sub-dimension feature search and re-initialization mechanisms to overcome stagnation. The second model uses random acceleration coefficients, instead of adaptive ones, based on non-linear circle, sine and helix functions, respectively, to increase diversification and intensification. Ensemble classifiers are also constructed with each base model trained using each optimized feature subset. A deep convolutional neural network is devised whose hyper-parameters are fine-tuned using the proposed PSO models. Extensive experimental studies using dermoscopic skin lesion data, medical data from the UCI machine learning repository, and ALL-IDB2 image data are conducted to evaluate the model efficiency systematically. The results from empirical evaluations and statistical tests indicate the superiority of the proposed models over other advanced PSO variants and classical search methods pertaining to discriminative feature selection and optimal hyper-parameter identification for deep learning networks in lesion classification as well as other disease diagnosis.",
author = "Tan, {Teck Yan} and Li Zhang and Lim, {Chee Peng}",
year = "2019",
month = nov,
doi = "10.1016/j.asoc.2019.105725",
language = "English",
volume = "84",
journal = "Applied Soft Computing",
issn = "1568-4946",
publisher = "Elsevier BV",

}

RIS

TY - JOUR

T1 - Intelligent skin cancer diagnosis using improved particle swarm optimization and deep learning models

AU - Tan, Teck Yan

AU - Zhang, Li

AU - Lim, Chee Peng

PY - 2019/11

Y1 - 2019/11

N2 - In this research, we propose an intelligent decision support system for skin cancer detection. Since generating an effective lesion representation is a vital step to ensure the success of lesion classification, the discriminative power of different types of features is exploited. Specifically, we combine clinically important asymmetry, border irregularity, colour and dermoscopic structure features with texture features extracted using Grey Level Run Length Matrix, Local Binary Patterns, and Histogram of Oriented Gradients operators for lesion representation. Then, we propose two enhanced Particle Swarm Optimization (PSO) models for feature optimization. The first model employs adaptive acceleration coefficients, multiple remote leaders, in-depth sub-dimension feature search and re-initialization mechanisms to overcome stagnation. The second model uses random acceleration coefficients, instead of adaptive ones, based on non-linear circle, sine and helix functions, respectively, to increase diversification and intensification. Ensemble classifiers are also constructed with each base model trained using each optimized feature subset. A deep convolutional neural network is devised whose hyper-parameters are fine-tuned using the proposed PSO models. Extensive experimental studies using dermoscopic skin lesion data, medical data from the UCI machine learning repository, and ALL-IDB2 image data are conducted to evaluate the model efficiency systematically. The results from empirical evaluations and statistical tests indicate the superiority of the proposed models over other advanced PSO variants and classical search methods pertaining to discriminative feature selection and optimal hyper-parameter identification for deep learning networks in lesion classification as well as other disease diagnosis.

AB - In this research, we propose an intelligent decision support system for skin cancer detection. Since generating an effective lesion representation is a vital step to ensure the success of lesion classification, the discriminative power of different types of features is exploited. Specifically, we combine clinically important asymmetry, border irregularity, colour and dermoscopic structure features with texture features extracted using Grey Level Run Length Matrix, Local Binary Patterns, and Histogram of Oriented Gradients operators for lesion representation. Then, we propose two enhanced Particle Swarm Optimization (PSO) models for feature optimization. The first model employs adaptive acceleration coefficients, multiple remote leaders, in-depth sub-dimension feature search and re-initialization mechanisms to overcome stagnation. The second model uses random acceleration coefficients, instead of adaptive ones, based on non-linear circle, sine and helix functions, respectively, to increase diversification and intensification. Ensemble classifiers are also constructed with each base model trained using each optimized feature subset. A deep convolutional neural network is devised whose hyper-parameters are fine-tuned using the proposed PSO models. Extensive experimental studies using dermoscopic skin lesion data, medical data from the UCI machine learning repository, and ALL-IDB2 image data are conducted to evaluate the model efficiency systematically. The results from empirical evaluations and statistical tests indicate the superiority of the proposed models over other advanced PSO variants and classical search methods pertaining to discriminative feature selection and optimal hyper-parameter identification for deep learning networks in lesion classification as well as other disease diagnosis.

U2 - 10.1016/j.asoc.2019.105725

DO - 10.1016/j.asoc.2019.105725

M3 - Article

VL - 84

JO - Applied Soft Computing

JF - Applied Soft Computing

SN - 1568-4946

M1 - 105725

ER -