Multi-Learner Based Deep Meta-Learning for Few-Shot Medical Image Classification.

Existing Few-Shot Learning (FSL) methods predominantly focus on developing different types of sophisticated models to extract the transferable prior knowledge for recognizing novel classes, while they almost pay less attention to the feature learning part in FSL which often simply leverage some well-known CNN as the feature learner. However, feature is the core medium for encoding such transferable knowledge. Feature learning is easy to be trapped in the over-fitting particularly in the scarcity of the training data, and thereby degenerates the performances of FSL. The handcrafted features, such as Histogram of Oriented Gradient (HOG) and Local Binary Pattern (LBP), have no requirement on the amount of training data, and used to perform quite well in many small-scale data scenarios, since their extractions involve no learning process, and are mainly based on the empirically observed and summarized prior feature engineering knowledge. In this paper, we intend to develop a general and simple approach for generally boosting FSL via exploiting such prior knowledge in the feature learning phase. To this end, we introduce two novel handcrafted feature regression modules, namely HOG and LBP regression, to the feature learning parts of deep learning-based FSL models. These two modules are separately plugged into the different convolutional layers of backbone based on the characteristics of the corresponding handcrafted features to guide the backbone optimization from different feature granularity, and also ensure that the learned feature can encode the handcrafted feature knowledge which improves the generalization ability of feature and alleviate the over-fitting of the models. Three recent state-of-the-art FSL approaches are leveraged for examining the effectiveness of our method. Extensive experiments on miniImageNet, CIFAR-FS and FC100 datasets show that the performances of all these FSL approaches are well boosted via applying our method on all three datasets. Our codes and models have been released.

Non-negative subspace feature representation for few-shot learning in medical imaging

Few-shot learning aims to learn generalized knowledge from limited data, which is a challenging study. Metric-based few-shot learning is the most widely studied direction, and most metric-based methods focus on how to learn a more robust feature embedding space or distance metric for classification. Few-shot learning is trained with a large number of tasks. Most existing methods do not realize that the role of the same sample in different learning tasks may be different, which will affect the effectiveness of classification. This paper proposes a task-adaptive feature adjustment method to obtain new features with task context information and more discrimination to solve this problem. We also introduce the Sphere manifold into few-shot learning, use the manifold metric instead of the Euclidean metric, and deduce the gradient optimization on the manifold. We propose a novel few-shot learning method with an excellent performance by combining these two innovations. We conduct several experiments on the few-shot learning benchmark datasets to demonstrate the advancement of our methods. The experimental results have significantly improved over the baseline and are highly competitive with the mainstream and effective methods.

Few-shot learning has emerged as a pivotal approach to address the challenges of limited labelled data in classification tasks. Its significance is emphasized in scenarios, such as network intrusion detection, where the prevalence of zero-day attacks and the scarcity of representative datasets exacerbate the difficulty of accurate model training and evaluation. This study explores the application of prototypical networks for few-shot meta-learning in the 3-way, 10-shot mode, and incorporates multiple n-way, k-shot configurations (3- way I-shot, 3- way 2-shot, and 3- way 3-shot) during the fine-tuning stage. This study uses a benchmark network intrusion detection dataset, NSL-KDD, to evaluate the proposed method. Unlike traditional supervised learning approaches and few-shot learning approaches, the proposed approach is trained, validated, and evaluated on 27 entirely disjoint classes, ensuring robustness and adaptability with validation and testing on classes excluded from the training stage. Fine-tuning experiments demonstrate rapid adaptation to unseen classes within 5 minutes by using minimal epochs and classification tasks. The experiments observe accuracy in the range of 76%-91% by using a fine-tuned model and significantly improve classification accuracy from 56% to 91% before and after fine-tuning the model. The study also captures a comparative analysis of the proposed approach with traditional supervised learning methods and various few-shot learning approaches used on the NSL-KDD dataset, highlighting the advantages of the proposed approach in terms of performance, speed, and adaptation to unseen classes with limited samples.

Artificial intelligence (AI) has advanced rapidly, but it has limited impact on biomedical text understanding due to a lack of annotated datasets (a.k.a. few-shot learning). Multi-task learning, which uses data from multiple datasets and tasks with related syntax and semantics, has potential to address this issue. However, the effectiveness of this approach heavily relies on the quality of the available data and its transferability between tasks. In this paper, we propose a framework, built upon a state-of-the-art multi-task method (i.e. MT-DNN), that leverages different publicly available biomedical datasets to enhance relation extraction performance. Our model employs a transformer-based architecture with shared encoding layers across multiple tasks, and task-specific classification layers to generate task-specific representations. To further improve performance, we utilize a knowledge distillation technique. In our experiments, we assess the impact of incorporating biomedical datasets in a multi-task learning setting and demonstrate that it consistently outperforms state-of-the-art few-shot learning methods in cases of limited data. This results in significant improvement across most datasets and few-shot scenarios, particularly in terms of recall scores.

Pneumonia remains a leading cause of mortality among children under five years of age, accounting for approximately 1.6 million deaths annually. Early detection is critical for reducing child mortality; however, traditional diagnostic methods rely on trained personnel and medical resources, often scarce in low-resource settings. Machine learning has emerged as a promising approach to pneumonia detection, but its effectiveness is hindered by the limited availability of labeled data required for training robust models. This study reviews existing machine learning techniques for pneumonia classification, including convolutional neural networks (CNNs), transfer learning, and few-shot learning approaches. Furthermore, we propose an optimized few-shot learning model that integrates Siamese networks with transfer learning to improve pneumonia detection using minimal labeled data. The model leverages MobileNetV3 as a pre-trained feature extractor, producing high-quality embeddings that enhance similarity learning within the Siamese framework. Additionally, triplet loss is incorporated to ensure a more discriminative embedding space, facilitating robust classification. Experimental results indicate that our model outperforms conventional machine learning classifiers, achieving an accuracy of 92.04%, precision of 91.20%, recall of 90.32%, and F1-score of 90.09%. The integration of Siamese networks with triplet loss enhances generalization while mitigating overfitting, making it particularly suitable for resource-constrained environments. These findings highlight the effectiveness of few-shot learning in pneumonia detection, offering a computationally efficient and scalable solution that bridges critical gaps in medical imaging.

Medical imaging is an indispensable tool in clinical diagnosis and therapeutic decision-making, encompassing a wide range of modalities such as radiography, ultrasound, CT, and MRI. With the rapid advancement of deep learning technologies, significant progress has been made in medical image analysis. However, existing deep learning methods are often limited by dataset size, which can lead to overfitting, while traditional approaches relying on hand-crafted features lack specificity and fail to fully capture complex pathological information. To address these challenges, we propose RadioGuide-DCN, an innovative radiomics-guided decorrelated classification network. Our method integrates radiomics features as prior information into deep neural networks and employs a feature decorrelation loss mechanism combined with an anti-attention feature fusion module to effectively reduce feature redundancy and enhance the model's capacity to capture both local details and global patterns. Specifically, we utilize a Kolmogorov-Arnold Network (KAN) classifier with learnable activation functions to further boost performance across various medical imaging datasets. Experimental results demonstrate that RadioGuide-DCN achieves an accuracy of 93.63% in BUSI image classification and consistently outperforms conventional radiomics and deep learning methods in multiple medical imaging classification tasks, significantly improving classification accuracy and AUC scores. Our study offers a novel paradigm for integrating deep learning with traditional imaging approaches and holds broad clinical application potential, particularly in tumor detection, image classification, and disease diagnosis.

As a high mortality disease, cancer seriously affects people's life and well-being. Reliance on pathologists to assess disease progression from pathological images is inaccurate and burdensome. Computer aided diagnosis (CAD) system can effectively assist diagnosis and make more credible decisions. However, a large number of labeled medical images that contribute to improve the accuracy of machine learning algorithm, especially for deep learning in CAD, are difficult to collect. Therefore, in this work, an improved few-shot learning method is proposed for medical image recognition. In addition, to make full use of the limited feature information in one or more samples, a feature fusion strategy is involved in our model. On the dataset of BreakHis and skin lesions, the experimental results show that our model achieved the classification accuracy of 91.22% and 71.20% respectively when only 10 labeled samples are given, which is superior to other state-of-the-art methods.

Introduction/Objective Recent advancements in artificial intelligence (AI), particularly with Large Language Models like ChatGPT-4Vision (GPT-4V), have expanded the potential for automated interpretation of medical images. This study evaluates the diagnostic accuracy of GPT-4V in histopathological analysis of neurodegenerative diseases and compares its performance with traditional Convolutional Neural Networks (CNNs). Methods/Case Report We utilized 1515 histopathological images, including hematoxylin and eosin (H&E) staining and tau immunohistochemistry, from patients with various neurodegenerative diseases, such as Alzheimer’s disease, progressive supranuclear palsy, and corticobasal degeneration. GPT-4V’s performance was assessed through multi- step prompts to assess how textual context influences the image interpretation. In the quantitative analysis, the diagnostic accuracy of GPT-4V and the CNN-based model YOLOv8 was assessed for classifying three tau lesions—astrocytic plaques, neuritic plaques, and tufted astrocytes. The analysis used both zero-shot and few-shot learning methods, where GPT-4V and YOLOv8 were evaluated on their ability to distinguish these lesions using ten hold-out test images and 490 training images per lesion. Results (if a Case Study enter NA) GPT-4V accurately recognized Lewy bodies in all cases, but it showed limited accuracy in identifying neurofibrillary tangles (20%) and Pick bodies (0%) on H&E slides. Notably, GPT-4V often suggested Alzheimer’s disease as a potential diagnosis, relying more on contextual clues, such as the prevalence of certain diseases, rather than direct image analysis. In the tau immunohistochemical analysis, GPT-4V’s diagnoses were strongly influenced by additional information about staining and brain region. However, few-shot learning markedly improved GPT-4V’s diagnostic capabilities from 40% accuracy with zero-shot learning to 90% with few-shot learning, reaching best accuracy with 20-shot learning. This performance compared favorably against YOLOv8, which also achieved 90% accuracy but required 100-shot learning. Conclusion While GPT-4V faces challenges in independently interpreting histopathological images, few-shot learning significantly improves its accuracy. This approach is particularly promising for neuropathology, where acquiring extensive labeled datasets is challenging. The findings highlight the need for ongoing refinement of AI applications in pathology, emphasizing balanced integration of textual and visual data to optimize diagnostic efficacy and reliability.

Few-shot learning in image classification aims at learning a model to identify novel categories from a few training samples. Many few-shot learning methods have been proposed to improve the model performance. However, they still suffer from data dependencies, weak feature representation ability, and poor performance on complex tasks. To address these three issues, we propose a global and local attention-based multiscale prototypical network (called GLAP), which includes a feature extraction network and a multiscale classification head. The feature extraction network is stacked by four global attention modules (GAMs) and four local attention modules (LAMs). On the one hand, each GAM learns the global context information in the whole image by calculating the global attention map of all pixels, thereby extracting general features to alleviate data dependencies. On the other hand, each LAM learns the local correspondences between features extracted by the corresponding GAM and features extracted by pre-trained ResNet50 at different scales, which combines high-level semantic features with low-level visual features to enhance feature representation ability. Afterward, the multiscale classification head is designed, which adopts a multiscale metric-based meta-learning (MMML) training method. MMML first calculates the category prototypes of four scales for metric classification, with the aim of generating intermediate predictions of four scales. Then, MMML generates the classification result by combining the intermediate predictions of four scales, thus improving the model performance on complex tasks. Extensive experiments on three benchmark datasets demonstrate that GLAP performs better than state-of-the-art few-shot learning methods.

Few-shot learning is a method to acquire learning ability in a small amount of sample data scenarios. This paper aims to study an online and offline interaction model applied to international Chinese education and teaching based on the few-shot learning method. It first expounds an overview of international Chinese education, including the connotation and characteristics of the hybrid mode of online teaching and offline teaching in international Chinese education, then designs an online and offline interactive model, and finally compares it with a baseline model and a traditional single teaching model. The experimental data shows that the accuracy and recall rate of the international Chinese education online and offline interaction model based on few-shot learning both reach 70%, which verifies its effectiveness.

Due to a lack of labeled samples, deep learning methods generally tend to have poor classification performance in practical applications. Few-shot learning (FSL), as an emerging learning paradigm, has been widely utilized in hyperspectral image (HSI) classification with limited labeled samples. However, the existing FSL methods generally ignore the domain shift problem in cross-domain scenes and rarely explore the associations between samples in the source and target domain. To tackle the above issues, a graph-based domain adaptation FSL (GDAFSL) method is proposed for HSI classification with limited training samples, which utilizes the graph method to guide the domain adaptation learning process in a uniformed framework. First, a novel deep residual hybrid attention network (DRHAN) is designed to extract discriminative embedded features efficiently for few-shot HSI classification. Then, a graph-based domain adaptation network (GDAN), which combines graph construction with domain adversarial strategy, is proposed to fully explore the domain correlation between source and target embedded features. By utilizing the fully explored domain correlations to guide the domain adaptation process, a domain invariant feature metric space is learned for few-shot HSI classification. Comprehensive experimental results conducted on three public HSI datasets demonstrate that GDAFSL is superior to the state-of-the-art with a small sample size.

Making machines learn like humans is the ultimate goal of artificial intelligence. Few-shot learning attempts to simulate the learning mechanism of humans, which is a task that can learn novel concepts from very few labeled samples. Due to the lack of sufficient labeled training data, existing methods often increase the risk of over-fitting or cause a considerable gap between clean and augmented data. To solve these problems, we present a novel compare network to perform robust few-shot learning in a meta-learned end-to-end manner. Specifically, we argue that it is desirable to learn a robust encoder that can draw inferences about other cases from one example. To this end, we improve the accuracy of few-shot learning by mining the internal mechanism of deep networks, which can leverage label information more effectively. By introducing shift-invariant blocks and a self-attention block in our architecture, all these components are seamlessly integrated into our framework, which can give feedback to each other without data augmentation. Furthermore, we provide ablative analyses of different blocks to help understand how each term contributes to performance. Extensive experiments demonstrate that our method can provide more robust results and outperform state-of-the-art few-shot learning methods.

This paper addresses the task of class-imbalanced medical image classification using complication labels. Medical image classification plays an essential role in improving patient care and reducing the burden on healthcare systems, along with other computer aided diagnosis tasks. However, there are two common obstacles with medical image image classification. One is that there is a high cost of acquiring annotations from doctors/experts. Two is that in many cases there is a differences in the number of incidents that appear per symptom which lead to class-imbalance. The above factors make medical image classification difficult to deal with. Additionally, there are cases where multiple symptoms appear in a mixed manner (complications), such as diabetic complications where both symptoms of infection and ischaemia could appear at the same time, which also could influence the classification result negatively. In this paper, a method that jointly tackles the obstacles mentioned above is presented. First, we conduct extensive experiment with various class-imbalanced learning methods introduced in previous works and also propose a method that improves the baseline class-imbalanced approaches by utilizing complication labels in pre-training. Second, we validate the proposed method with the diabetic foot ulcer dataset introduced in Diabetic Foot Ulcer Challenge 2021.

In many real-world problems, collecting a large number of labeled samples is infeasible. Few-shot learning (FSL) is the dominant approach to address this issue, where the objective is to quickly adapt to novel categories in presence of a limited number of samples. FSL tasks have been predominantly solved by leveraging the ideas from gradient-based meta-learning and metric learning approaches. However, recent works have demonstrated the significance of powerful feature representations with a simple embedding network that can outperform existing sophisticated FSL algorithms. In this work, we build on this insight and propose a novel training mechanism that simultaneously enforces equivariance and invariance to a general set of geometric transformations. Equivariance or invariance has been employed standalone in the previous works; however, to the best of our knowledge, they have not been used jointly. Simultaneous optimization for both of these contrasting objectives allows the model to jointly learn features that are not only independent of the input transformation but also the features that encode the structure of geometric transformations. These complementary sets of features help generalize well to novel classes with only a few data samples. We achieve additional improvements by incorporating a novel self-supervised distillation objective. Our extensive experimentation shows that even without knowledge distillation our proposed method can outperform current state-of-the-art FSL methods on five popular benchmark datasets.