Alleviating the Incompatibility Between Cross Entropy Loss and Episode Training for Few-Shot Skin Disease Classification

Zhu, Wei; Liao, Haofu; Li, Wenbin; Li, Weijian; Luo, Jiebo

doi:10.1007/978-3-030-59725-2_32

Cited by 15 publications

(10 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several improved skin lesion-classification methods based on FSL have also been introduced, containing multiperspective improvements on vanilla methods. Liu et al introduced an embedding-based comparison method improved from RelationNet, 18 Zhu et al proposed a novel loss-function calculation named queryrelative loss over meta-learning, 19 Li et al and Prabhu et al raised a similar issue of focusing on clustering methods using category judgment in feature maps, 20,21 and Mahajan et al proposed an enhanced convolution strategy, ie, groupequivariant convolutions (G-convolutions) to alleviate classification error brought about by data augmentation, such as rotations and reflections. 22 However, these studies were limited to nonuniversal approaches or few disease classes.…”

Section: State Of the Artmentioning

confidence: 99%

“…CDD-Net focuses on feature-level improvement, with the application of the developed extraction and fusion strategy, which makes it possible to utilize almost all the current FSL methods, rather than improvement only in specific tools, such as metric learning-based methods 16,20,21 or meta-learning-based methods. 18,19 To test our method, we built a large skin disease dataset, Derm104, with 104 classes of skin lesions, which provides a large benchmark for skin disease FSL classification. It merges 1781 self-collected images from a public skin disease atlas website (www.med126.com/pf) and 2702 images from the public dataset SD-198, 23 for a total of 4483 images.…”

Section: State Of the Artmentioning

confidence: 99%

See 1 more Smart Citation

Few-Shot Classification with Multiscale Feature Fusion for Clinical Skin Disease Diagnosis

Chen,

Liu,

Yang

2024

CCID

View full text Add to dashboard Cite

Introduction: Skin disease is one of the most common diseases and can affect people of all ages and races. However, the diagnosis of skin diseases via observation is a highly challenging task for both doctors and patients, and would benefit from the use of an intelligent system. Building a large benchmark with professional dermatologists is resource-intensive, and we believe that few-shot learning (FSL) methods would be helpful in solving the problem of annotated data scarcity. In this paper, we propose CDD-Net (Context Feature Fusion and Dual Attention Dermatology Net), a plug-in module for FSL clinical skin disease classification. Methods: Current FSL methods used in skin disease classification are limited to nonuniversal approaches and few disease classes. Our CDD-Net has a flexible structure, including a context feature-fusion module and dual-attention module to extract discriminating texture feature and emphasize contributive regions and channels. The context feature-fusion module localizes discriminatory texture details of skin lesions by integrating features from different layers, while the dual-attention module highlights discriminative regions via channel-wise and pixel-wise depictions based on weight vectors and restrains the contributions of irrelevant areas. We also present Derm104, a new clinical skin disease data benchmark that has significant coverage of rare diseases and reliable annotation between primary species and subspecies for better validation of our approach. Results: Our experiments validated the versatility of CDD-Net for different FSL methods and achieved an improvement in accuracy of up to 9.14 percentage points compared with the vanilla network, which can be considered state of the art. The ablation study also showed that the dual-attention module and context feature-fusion module worked efficiently in CDD-Net.

show abstract

Section: State Of the Artmentioning

confidence: 99%

Section: State Of the Artmentioning

confidence: 99%

Few-Shot Classification with Multiscale Feature Fusion for Clinical Skin Disease Diagnosis

Chen,

Liu,

Yang

2024

CCID

View full text Add to dashboard Cite

show abstract

“…One common solution is few-shot learning (FSL), which aims to learn transferable knowledge on common classes (base classes) and apply it to the rare ones (novel classes) with only a handful of labeled data. Current FSL methods for few-shot skin disease classification can be broadly divided into two groups: meta-learning based (Prabhu et al, 2019;Mahajan et al, 2020;Zhang et al, 2020;Zhu et al, 2020Zhu et al, , 2021Singh et al, 2021) and transfer-learning based (Guo et al, 2020;Chen et al, 2021;Medina et al, 2020;Phoo and Hariharan, 2020). In the category of meta-learning-based methods, existing solutions are mainly based on MAML (Finn et al, 2017) and Prototypical Networks (Snell et al, 2017), with the focus on different problems in rare skin disease classification.…”

Section: Introductionmentioning

confidence: 99%

Sub-cluster-aware Network for Few-shot Skin Disease Classification

LI¹,

Li²,

Xu³

et al. 2022

Preprint

View full text Add to dashboard Cite

This paper studies the few-shot skin disease classification problem. Based on a crucial observation that skin disease images often exist multiple sub-clusters within a class (i.e., the appearances of images within one class of disease vary and form multiple distinct sub-groups), we design a novel Sub-Cluster-Aware Network, namely SCAN, for rare skin disease diagnosis with enhanced accuracy. As the performance of few-shot learning highly depends on the quality of the learned feature encoder, the main principle guiding the design of SCAN is the intrinsic sub-clustered representation learning for each class so as to better describe feature distributions. Specifically, SCAN follows a dual-branch framework, where the first branch is to learn class-wise features to distinguish different skin diseases, and the second one aims to learn features which can effectively partition each class into several groups so as to preserve the sub-clustered structure within each class. To achieve the objective of the second branch, we present a cluster loss to learn image similarities via unsupervised clustering. To ensure that the samples in each sub-cluster are from the same class, we further design a purity loss to refine the unsupervised clustering results. We evaluate the proposed approach on two public datasets for few-shot skin disease classification. The experimental results validate that our framework outperforms the other state-of-the-art methods by around 2% to 4% on the SD-198 and Derm7pt datasets.

show abstract

“…How to implement a similar self-distilling strategy on an unsupervised base dataset, though, is not obvious. On the other hand, we are aware of only few methods [34,12,18,21] for FSL of medical image classification, and to the best of our knowledge, all of them relied on heavy labeling of the base dataset, causing a great burden for practical applications. Lastly, the meta-learning process and the target task are often isolated in most existing FSL approaches, and the meta-learner has little knowledge about its end task.…”

Section: Introductionmentioning

confidence: 99%

Unsupervised Representation Learning Meets Pseudo-Label Supervised Self-Distillation: A New Approach to Rare Disease Classification

Sun,

Wei,

et al. 2021

Preprint

View full text Add to dashboard Cite

Rare diseases are characterized by low prevalence and are often chronically debilitating or life-threatening. Imaging-based classification of rare diseases is challenging due to the severe shortage in training examples. Few-shot learning (FSL) methods tackle this challenge by extracting generalizable prior knowledge from a large base dataset of common diseases and normal controls, and transferring the knowledge to rare diseases. Yet, most existing methods require the base dataset to be labeled and do not make full use of the precious examples of the rare diseases. To this end, we propose in this work a novel hybrid approach to rare disease classification, featuring two key novelties targeted at the above drawbacks. First, we adopt the unsupervised representation learning (URL) based on self-supervising contrastive loss, whereby to eliminate the overhead in labeling the base dataset. Second, we integrate the URL with pseudo-label supervised classification for effective self-distillation of the knowledge about the rare diseases, composing a hybrid approach taking advantages of both unsupervised and (pseudo-) supervised learning on the base dataset. Experimental results on classification of rare skin lesions show that our hybrid approach substantially outperforms existing FSL methods (including those using fully supervised base dataset) for rare disease classification via effective integration of the URL and pseudo-label driven self-distillation, thus establishing a new state of the art.

show abstract

Alleviating the Incompatibility Between Cross Entropy Loss and Episode Training for Few-Shot Skin Disease Classification

Cited by 15 publications

References 11 publications

Few-Shot Classification with Multiscale Feature Fusion for Clinical Skin Disease Diagnosis

Few-Shot Classification with Multiscale Feature Fusion for Clinical Skin Disease Diagnosis

Sub-cluster-aware Network for Few-shot Skin Disease Classification

Unsupervised Representation Learning Meets Pseudo-Label Supervised Self-Distillation: A New Approach to Rare Disease Classification

Contact Info

Product

Resources

About