Meta-Transfer Derm-Diagnosis: Exploring Few-Shot Learning and Transfer Learning for Skin Disease Classification in Long-Tail Distribution
2404.16814
0
0
🔄
Abstract
Addressing the challenges of rare diseases is difficult, especially with the limited number of reference images and a small patient population. This is more evident in rare skin diseases, where we encounter long-tailed data distributions that make it difficult to develop unbiased and broadly effective models. The diverse ways in which image datasets are gathered and their distinct purposes also add to these challenges. Our study conducts a detailed examination of the benefits and drawbacks of episodic and conventional training methodologies, adopting a few-shot learning approach alongside transfer learning. We evaluated our models using the ISIC2018, Derm7pt, and SD-198 datasets. With minimal labeled examples, our models showed substantial information gains and better performance compared to previously trained models. Our research emphasizes the improved ability to represent features in DenseNet121 and MobileNetV2 models, achieved by using pre-trained models on ImageNet to increase similarities within classes. Moreover, our experiments, ranging from 2-way to 5-way classifications with up to 10 examples, showed a growing success rate for traditional transfer learning methods as the number of examples increased. The addition of data augmentation techniques significantly improved our transfer learning based model performance, leading to higher performances than existing methods, especially in the SD-198 and ISIC2018 datasets. All source code related to this work will be made publicly available soon at the provided URL.
Create account to get full access
Overview
- Addressing rare skin diseases is challenging due to limited data and a small patient population
- The diverse ways in which image datasets are gathered and their distinct purposes add to these challenges
- The study examines the benefits and drawbacks of episodic and conventional training methodologies, adopting a few-shot learning approach alongside transfer learning
- Evaluations were performed using the ISIC2018, Derm7pt, and SD-198 datasets
Plain English Explanation
Rare skin diseases are difficult to study because there are not many examples of them, and the patients are spread out. Additionally, the data collected on these diseases comes from different sources and is used for different purposes, adding to the challenge.
This study looked at the pros and cons of two training methods - episodic and conventional - that use a small number of examples to learn about rare skin diseases. They used pre-trained models, which means the models were first trained on a large, general dataset (ImageNet) before being trained on the skin disease datasets. This helped the models better identify features within the skin disease classes.
The researchers evaluated their models using three different skin disease datasets: ISIC2018, Derm7pt, and SD-198. Even with just a few examples, their models showed significant improvements over previously trained models. Data augmentation, which creates new training examples by transforming existing ones, also helped improve the transfer learning-based model performance.
Technical Explanation
The researchers conducted a detailed examination of episodic and conventional training methodologies, adopting a few-shot learning approach alongside transfer learning. They evaluated their models using the ISIC2018, Derm7pt, and SD-198 datasets.
Their experiments, ranging from 2-way to 5-way classifications with up to 10 examples, showed a growing success rate for traditional transfer learning methods as the number of examples increased. The DenseNet121 and MobileNetV2 models demonstrated improved feature representation capabilities when using pre-trained models on ImageNet, which increased similarities within classes.
The addition of data augmentation techniques significantly improved the transfer learning-based model performance, leading to higher performances than existing methods, especially in the SD-198 and ISIC2018 datasets.
Critical Analysis
The researchers acknowledged the limited number of reference images and small patient population as a significant challenge in addressing rare skin diseases. They also noted the diverse ways in which image datasets are gathered and their distinct purposes as an additional obstacle.
While the study's findings suggest that transfer learning and data augmentation can help overcome the challenges of rare skin disease classification, the researchers did not discuss potential limitations or biases in the datasets used. It would be helpful to understand how representative the ISIC2018, Derm7pt, and SD-198 datasets are of the broader population of rare skin diseases and whether there are any demographic or geographic biases.
Additionally, the researchers did not explore the generalizability of their approach to other rare disease domains beyond dermatology. It would be valuable to understand how well the techniques employed in this study could be applied to address challenges in other rare disease areas.
Conclusion
This study highlights the potential of few-shot learning and transfer learning to address the challenges of rare skin disease classification, where limited data and small patient populations make it difficult to develop unbiased and broadly effective models.
By leveraging pre-trained models and data augmentation techniques, the researchers were able to achieve substantial information gains and better performance compared to previously trained models, even with minimal labeled examples.
The insights from this research could have important implications for the development of more accurate and accessible diagnostic tools for rare skin diseases, ultimately improving patient outcomes. Further research is needed to explore the generalizability of these techniques to other rare disease domains and address potential biases in the underlying datasets.
This summary was produced with help from an AI and may contain inaccuracies - check out the links to read the original source documents!
Related Papers
Skin Cancer Images Classification using Transfer Learning Techniques
Md Sirajul Islam, Sanjeev Panta
0
0
Skin cancer is one of the most common and deadliest types of cancer. Early diagnosis of skin cancer at a benign stage is critical to reducing cancer mortality. To detect skin cancer at an earlier stage an automated system is compulsory that can save the life of many patients. Many previous studies have addressed the problem of skin cancer diagnosis using various deep learning and transfer learning models. However, existing literature has limitations in its accuracy and time-consuming procedure. In this work, we applied five different pre-trained transfer learning approaches for binary classification of skin cancer detection at benign and malignant stages. To increase the accuracy of these models we fine-tune different layers and activation functions. We used a publicly available ISIC dataset to evaluate transfer learning approaches. For model stability, data augmentation techniques are applied to improve the randomness of the input dataset. These approaches are evaluated using different hyperparameters such as batch sizes, epochs, and optimizers. The experimental results show that the ResNet-50 model provides an accuracy of 0.935, F1-score of 0.86, and precision of 0.94.
6/21/2024
Diffusion Model Driven Test-Time Image Adaptation for Robust Skin Lesion Classification
Ming Hu, Siyuan Yan, Peng Xia, Feilong Tang, Wenxue Li, Peibo Duan, Lin Zhang, Zongyuan Ge
0
0
Deep learning-based diagnostic systems have demonstrated potential in skin disease diagnosis. However, their performance can easily degrade on test domains due to distribution shifts caused by input-level corruptions, such as imaging equipment variability, brightness changes, and image blur. This will reduce the reliability of model deployment in real-world scenarios. Most existing solutions focus on adapting the source model through retraining on different target domains. Although effective, this retraining process is sensitive to the amount of data and the hyperparameter configuration for optimization. In this paper, we propose a test-time image adaptation method to enhance the accuracy of the model on test data by simultaneously updating and predicting test images. We modify the target test images by projecting them back to the source domain using a diffusion model. Specifically, we design a structure guidance module that adds refinement operations through low-pass filtering during reverse sampling, regularizing the diffusion to preserve structural information. Additionally, we introduce a self-ensembling scheme automatically adjusts the reliance on adapted and unadapted inputs, enhancing adaptation robustness by rejecting inappropriate generative modeling results. To facilitate this study, we constructed the ISIC2019-C and Dermnet-C corruption robustness evaluation benchmarks. Extensive experiments on the proposed benchmarks demonstrate that our method makes the classifier more robust across various corruptions, architectures, and data regimes. Our datasets and code will be available at url{https://github.com/minghu0830/Skin-TTA_Diffusion}.
5/21/2024
🤿
An interpretable imbalanced semi-supervised deep learning framework for improving differential diagnosis of skin diseases
Futian Weng, Yuanting Ma, Jinghan Sun, Shijun Shan, Qiyuan Li, Jianping Zhu, Yang Wang, Yan Xu
0
0
Dermatological diseases are among the most common disorders worldwide. This paper presents the first study of the interpretability and imbalanced semi-supervised learning of the multiclass intelligent skin diagnosis framework (ISDL) using 58,457 skin images with 10,857 unlabeled samples. Pseudo-labelled samples from minority classes have a higher probability at each iteration of class-rebalancing self-training, thereby promoting the utilization of unlabeled samples to solve the class imbalance problem. Our ISDL achieved a promising performance with an accuracy of 0.979, sensitivity of 0.975, specificity of 0.973, macro-F1 score of 0.974 and area under the receiver operating characteristic curve (AUC) of 0.999 for multi-label skin disease classification. The Shapley Additive explanation (SHAP) method is combined with our ISDL to explain how the deep learning model makes predictions. This finding is consistent with the clinical diagnosis. We also proposed a sampling distribution optimisation strategy to select pseudo-labelled samples in a more effective manner using ISDLplus. Furthermore, it has the potential to relieve the pressure placed on professional doctors, as well as help with practical issues associated with a shortage of such doctors in rural areas.
6/11/2024
Data Alignment for Zero-Shot Concept Generation in Dermatology AI
Soham Gadgil, Mahtab Bigverdi
0
0
AI in dermatology is evolving at a rapid pace but the major limitation to training trustworthy classifiers is the scarcity of data with ground-truth concept level labels, which are meta-labels semantically meaningful to humans. Foundation models like CLIP providing zero-shot capabilities can help alleviate this challenge by leveraging vast amounts of image-caption pairs available on the internet. CLIP can be fine-tuned using domain specific image-caption pairs to improve classification performance. However, CLIP's pre-training data is not well-aligned with the medical jargon that clinicians use to perform diagnoses. The development of large language models (LLMs) in recent years has led to the possibility of leveraging the expressive nature of these models to generate rich text. Our goal is to use these models to generate caption text that aligns well with both the clinical lexicon and with the natural human language used in CLIP's pre-training data. Starting with captions used for images in PubMed articles, we extend them by passing the raw captions through an LLM fine-tuned on the field's several textbooks. We find that using captions generated by an expressive fine-tuned LLM like GPT-3.5 improves downstream zero-shot concept classification performance.
4/22/2024