A Skin Disease Classification Model Based on DenseNet and ConvNeXt Fusion

Abstract

Skin disease is one of the most common diseases. Due to the intricate categories of skin diseases, their symptoms being very similar in the early stage, and the lesion samples being extremely unbalanced, their classification is challenging. At the same time, under the conditions of limited data, the generalization ability of a single reliable convolutional neural network model is weak, the feature extraction ability is insufficient, and the classification accuracy is low. Therefore, in this paper, we proposed a convolutional neural network model for skin disease classification based on model fusion. Through model fusion, deep and shallow feature fusion, and the introduction of an attention module, the feature extraction capacity of the model was strengthened. In addition, a series of works such as model pre-training, data augmentation, and parameter fine-tuning were conducted to upgrade the classification performance of the model. The experimental results showed that when working on our private dataset dominated by acne-like skin diseases, our proposed model outperformed the two baseline models of DenseNet201 and ConvNeXt_L by 4.42% and 3.66%, respectively. On the public HAM10000 dataset, the accuracy and f1-score of the proposed model were 95.29% and 89.99%, respectively, which also achieved good results compared with other state-of-the-art models.

A GAN-Based Data Augmentation Method for Imbalanced Multi-Class Skin Lesion Classification

“…It consists of 10,015 dermoscopic skin lesion images from seven skin diseases with a resolution of 600 450  pixels. For fair comparison, this study follows 8 VOLUME XX, 2017 [22], [27], [69], [70], [71] and split the HAM10000 dataset into a training and test set by the proportion of 8:2, each containing 8012 images and 2003 images, respectively (Table I). These works are followed based on two reasons.…”

“…First, similar to this research, the works in [22], [27] also employed GAN-based data augmentation method to mitigate limited data and class imbalance problem. Second, very recent works in [69], [70], [71] have demonstrated notable performance in the field of skin lesion classification.…”

“…The works in [77], [78], [79] correspond to the top three winners' team of ISIC 2018 Challenge, which use HAM10000 dataset as training set [80], representing a high level of performance in this classification task. The works in [22], [27], [69], [70], [71] correspond to recent representative methods from the literature. It should be noted that, the ISIC 2018 Challenge provided the HAM10000 dataset as training data, with additional testing data not included in the original dataset.…”

FOTCA: hybrid transformer-CNN architecture using AFNO for accurate plant leaf disease image recognition

“…Based on earlier work by scholars ( Kumar et al., 2023a ; Kumar et al., 2023b ; Wei et al., 2023 ), we selected DenseNet-169, Inception-v3, VGG19, ResNet-50, ResNet-101, ViT as our comparative experimental models to verify the feasibility and reliability of the improved model plant leaf disease image recognition method in this paper. We all iteratively update the pre-trained model parameters based on each model to accelerate the model convergence during the training process.…”

SkinLiTE: Lightweight Supervised Contrastive Learning Model for Enhanced Skin Lesion Detection and Disease Typification in Dermoscopic Images

“…Contrastingly, Tripathi, et al [33] and Bala, et al [34] developed specific models named XCNN and MonkeyNet, tailored for their respective tasks, suggesting a preference for custom-built architectures in certain scenarios. Wei, et al [35] combined DenseNet and ConvNeXt models, incorporating attention modules and pre-training, data augmentation, and fine-tuning processes to classify skin diseases, especially focusing on acne-like diseases. El Gannour, et al [36] utilized two ensemble learning techniques, integrating image and clinical data inputs for improved skin disease classification accuracy.…”