Abstract:Background and objective: Medicine image classification are important methods of traditional medical image analysis, but the trainable data in medical image classification is highly imbalanced and the accuracy of medical image classification models is low. In view of the above two common problems in medical image classification. This study aims to: (i) effectively solve the problem of poor training effect caused by the imbalance of class imbalanced data sets. (ii) propose a network framework suitable for improving medical image classification results, which needs to be superior to existing methods. Methods: In this paper, we put in the diffusion model multi-scale feature fusion network (DMSFF), which mainly uses the diffusion generation model to overcome imbalanced classes (DMOIC) on highly imbalanced medical image datasets. At the same time, it is processed according to the cropped image augmentation strategy through cropping (IASTC). Based on this, we use the new dataset to design a multi-scale feature fusion network (MSFF) that can fully utilize multiple hierarchical features. The DMSFF network can effectively solve the problems of small and imbalanced samples and low accuracy in medical image classification. Results: We evaluated the performance of the DMSFF network on highly imbalanced medical image classification datasets APTOS2019 and ISIC2018. Compared with other classification models, our proposed DMSFF network achieved significant improvements in classification accuracy and F1 score on two datasets, reaching 0.872, 0.731, and 0.906, 0.836, respectively. Conclusions: Our newly proposed DMSFF architecture outperforms existing methods on two datasets, and verifies the effectiveness of generative model inverse balance for imbalance class datasets and feature enhancement by multi-scale feature fusion. Further, the method can be applied to other class imbalanced data sets where the results will be improved.

DDFA: a displacement and diffusion-based feature augmentation method for imbalanced image recognition

Boosting Unsupervised Contrastive Learning Using Diffusion-Based Data Augmentation from Scratch

Latent-based Diffusion Model for Long-tailed Recognition

CUDA: Curriculum of Data Augmentation for Long-Tailed Recognition

Feature Space Augmentation for Long-Tailed Data

Anisotropic Diffusion Probabilistic Model for Imbalanced Image Classification

Data-Centric Long-Tailed Image Recognition

SAFA: Sample-Adaptive Feature Augmentation for Long-Tailed Image Classification

Towards Effective Data-Free Knowledge Distillation via Diverse Diffusion Augmentation

Instance-Specific Semantic Augmentation for Long-Tailed Image Classification

Detail Reinforcement Diffusion Model: Augmentation Fine-Grained Visual Categorization in Few-Shot Conditions

Feature Re-Balancing for Long-Tailed Visual Recognition.

Decoupled Data Augmentation for Improving Image Classification

FDAA: A feature distribution-aware transferable adversarial attack method

ECS-SC: Long-tailed classification via data augmentation based on easily confused sample selection and combination

Text-guided Fourier Augmentation for long-tailed recognition

A diffusion model multi-scale feature fusion network for imbalanced medical image classification research

DALDA: Data Augmentation Leveraging Diffusion Model and LLM with Adaptive Guidance Scaling

A Simple Data Augmentation for Feature Distribution Skewed Federated Learning

Boosting Dense Long-Tailed Object Detection from Data-Centric View

Experimenting with an Evaluation Framework for Imbalanced Data Learning (EFIDL)