Dermo-DOCTOR: A framework for concurrent skin lesion detection and recognition using a deep convolutional neural network with end-to-end dual encoders

Automated skin lesion analysis for simultaneous detection and recognition is still challenging inter-class homogeneity intra-class heterogeneity, leading to low generic capability of a single convolutional neural network (CNN) with limited datasets. This article proposes an end-to-end deep CNN-based framework the lesions, named Dermo-DOCTOR, consisting two encoders. The feature maps from encoders are fused channel-wise, called Fused Feature Map (FFM). FFM utilized decoding in sub-network, concatenating each stage encoders’ outputs corresponding decoder layers retrieve lost spatial information due pooling For three fully connected layers, utilizing FFM, aggregated obtain final class. We train evaluate proposed Dermo-Doctor publicly available benchmark datasets, such as ISIC-2016 ISIC-2017. achieved segmentation results exhibit mean intersection over unions 85.0% 80.0% respectively ISIC-2017 test Dermo-DOCTOR also demonstrates praiseworthy success recognition, providing areas under receiver operating characteristic curves 0.98 0.91 those experimental show that outperforms alternative methods mentioned literature, designed recognition. As provides better on different even training data, it can be auspicious computer-aided assistive tool dermatologists.