TY - CHAP
T1 - Evaluation of Big Data based CNN Models in Classification of Skin Lesions with Melanoma
AU - Naronglerdrit, Prasitthichai
AU - Mporas, Iosif
N1 - Copyright © 2021, The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. This is the accepted manuscript version of Naronglerdrit P., Mporas I. (2021) Evaluation of Big Data Based CNN Models in Classification of Skin Lesions with Melanoma. In: Kose U., Alzubi J. (eds) Deep Learning for Cancer Diagnosis. Studies in Computational Intelligence, vol 908. Springer, Singapore. https://doi.org/10.1007/978-981-15-6321-8_5
PY - 2021/1/1
Y1 - 2021/1/1
N2 - This chapter presents a methodology for diagnosis of pigmented skin lesions using convolutional neural networks. The architecture is based on convolutional neural networks and it is evaluated using new CNN models as well as retrained modification of pre-existing CNN models were used. The experimental results showed that CNN models pre-trained on big datasets for general purpose image classification when re-trained in order to identify skin lesion types offer more accurate results when compared to convolutional neural network models trained explicitly from the dermatoscopic images. The best performance was achieved by retraining a modified version of ResNet-50 convolutional neural network with accuracy equal to 93.89%. Analysis on skin lesion pathology type was also performed with classification accuracy for melanoma and basal cell carcinoma being equal to 79.13% and 82.88%, respectively.
AB - This chapter presents a methodology for diagnosis of pigmented skin lesions using convolutional neural networks. The architecture is based on convolutional neural networks and it is evaluated using new CNN models as well as retrained modification of pre-existing CNN models were used. The experimental results showed that CNN models pre-trained on big datasets for general purpose image classification when re-trained in order to identify skin lesion types offer more accurate results when compared to convolutional neural network models trained explicitly from the dermatoscopic images. The best performance was achieved by retraining a modified version of ResNet-50 convolutional neural network with accuracy equal to 93.89%. Analysis on skin lesion pathology type was also performed with classification accuracy for melanoma and basal cell carcinoma being equal to 79.13% and 82.88%, respectively.
U2 - 10.1007/978-981-15-6321-8
DO - 10.1007/978-981-15-6321-8
M3 - Chapter (peer-reviewed)
VL - 908
SP - 79
EP - 98
BT - Deep Learning for Cancer Diagnosis, Series Volume: 908, DOI: , eBook ISBN:
ER -