Medical Image Classification Research Articles

Research on Brain Tumor Medical Image Classification Based on Diffusion Self-Supervised Representation Learning

Research on Brain Tumor Medical Image Classification Based on Diffusion Self-Supervised Representation Learning

Score-Based Counterfactual Generation for Interpretable Medical Image Classification and Lesion Localization.

Deep neural networks (DNNs) have immense potential for precise clinical decision-making in the field of biomedical imaging. However, accessing high-quality data is crucial for ensuring the high-performance of DNNs. Obtaining medical imaging data is often challenging in terms of both quantity and quality. To address these issues, we propose a score-based counterfactual generation (SCG) framework to create counterfactual images from latent space, to compensate for scarcity and imbalance of data. In addition, some uncertainties in external physical factors may introduce unnatural features and further affect the estimation of the true data distribution. Therefore, we integrated a learnable FuzzyBlock into the classifier of the proposed framework to manage these uncertainties. The proposed SCG framework can be applied to both classification and lesion localization tasks. The experimental results revealed a remarkable performance boost in classification tasks, achieving an average performance enhancement of 3-5% compared to previous state-of-the-art (SOTA) methods in interpretable lesion localization.

An Evolutionary Deep Learning Method Based on Improved Heap-Based Optimization for Medical Image Classification and Diagnosis

An Evolutionary Deep Learning Method Based on Improved Heap-Based Optimization for Medical Image Classification and Diagnosis

Multi-Label Generalized Zero Shot Chest Xray Classification By Combining Image-Text Information With Feature Disentanglement.

In fully supervised learning-based medical image classification, the robustness of a trained model is influenced by its exposure to the range of candidate disease classes. Generalized Zero Shot Learning (GZSL) aims to correctly predict seen and novel unseen classes. Current GZSL approaches have focused mostly on the single-label case. However, it is common for chest X-rays to be labelled with multiple disease classes. We propose a novel multi-modal multi-label GZSL approach that leverages feature disentanglement andmulti-modal information to synthesize features of unseen classes. Disease labels are processed through a pre-trained BioBert model to obtain text embeddings that are used to create a dictionary encoding similarity among different labels. We then use disentangled features and graph aggregation to learn a second dictionary of inter-label similarities. A subsequent clustering step helps to identify representative vectors for each class. The multi-modal multi-label dictionaries and the class representative vectors are used to guide the feature synthesis step, which is the most important component of our pipeline, for generating realistic multi-label disease samples of seen and unseen classes. Our method is benchmarked against multiple competing methods and we outperform all of them based on experiments conducted on the publicly available NIH and CheXpert chest X-ray datasets.

MedOptNet: Meta-Learning Framework for Few-Shot Medical Image Classification.

In the medical research domain, limited data and high annotation costs have made efficient classification under few-shot conditions a popular research area. This paper proposes a meta-learning framework, termed MedOptNet, for few-shot medical image classification. The framework enables the use of various high-performance convex optimization models as classifiers, such as multi-class kernel support vector machines, ridge regression, and other models. End-to-end training is then implemented using dual problems and differentiation in the paper. Additionally, various regularization techniques are employed to enhance the model's generalization capabilities. Experiments on the BreakHis, ISIC2018, and Pap smear medical few-shot datasets demonstrate that the MedOptNet framework outperforms benchmark models. Moreover, the model training time is also compared to prove its effectiveness in the paper, and an ablation study is conducted to validate the effectiveness of each module.

Pyramid Pixel Context Adaption Network for Medical Image Classification With Supervised Contrastive Learning.

Spatial attention (SA) mechanism has been widely incorporated into deep neural networks (DNNs), significantly lifting the performance in computer vision tasks via long-range dependency modeling. However, it may perform poorly in medical image analysis. Unfortunately, the existing efforts are often unaware that long-range dependency modeling has limitations in highlighting subtle lesion regions. To overcome this limitation, we propose a practical yet lightweight architectural unit, pyramid pixel context adaption (PPCA) module, which exploits multiscale pixel context information to recalibrate pixel position in a pixel-independent manner dynamically. PPCA first applies a well-designed cross-channel pyramid pooling (CCPP) to aggregate multiscale pixel context information, then eliminates the inconsistency among them by the well-designed pixel normalization (PN), and finally estimates per pixel attention weight via a pixel context integration. By embedding PPCA into a DNN with negligible overhead, the PPCA network (PPCANet) is developed for medical image classification. In addition, we introduce supervised contrastive learning to enhance feature representation by exploiting the potential of label information via supervised contrastive loss (CL). The extensive experiments on six medical image datasets show that the PPCANet outperforms state-of-the-art (SOTA) attention-based networks and recent DNNs. We also provide visual analysis and ablation study to explain the behavior of PPCANet in the decision-making process.

Research on Multi-Scale Feature Fusion Network Algorithm Based on Brain Tumor Medical Image Classification

Research on Multi-Scale Feature Fusion Network Algorithm Based on Brain Tumor Medical Image Classification

RSSL: A Unified Reliability-Aware Semi-Supervised Learning Framework for Medical Image Classification

RSSL: A Unified Reliability-Aware Semi-Supervised Learning Framework for Medical Image Classification

Improving Tumor Classification by Reusing Self-predicted Segmentation of Medical Images as Guiding Knowledge.

Differential diagnosis of tumors is important for computer-aided diagnosis. In computer-aided diagnosis systems, expert knowledge of lesion segmentation masks is limited as it is only used during preprocessing or as supervision to guide feature extraction. To improve the utilization of lesion segmentation masks, this study proposes a simple and effective multitask learning network that improves medical image classification using self-predicted segmentation as guiding knowledge; we call this network RS 2-net. In RS 2-net, the predicted segmentation probability map obtained from the initial segmentation inference is added to the original image to form a new input, which is then reinput to the network for the final classification inference. We validated the proposed RS 2-net using three datasets: the pNENs-Grade dataset, which tested the prediction of pancreatic neuroendocrine neoplasm grading, and the HCC-MVI dataset, which tested the prediction of microvascular invasion of hepatocellular carcinoma, and ISIC 2017 public skin lesion dataset. The experimental results indicate that the proposed strategy of reusing self-predicted segmentation is effective, and RS 2-net outperforms other popular networks and existing state-of-the-art studies. Interpretive analytics based on feature visualization demonstrates that the improved classification performance of our reuse strategy is due to the semantic information that can be acquired in advance in a shallow network.

Artificial intelligence for human gunshot wound classification

Certain features are helpful in the identification of gunshot entrance and exit wounds, such as the presence of muzzle imprints, peripheral tears, stippling, bone beveling, and wound border irregularity. Some cases are less straightforward and wounds can thus pose challenges to an emergency room doctor or forensic pathologist. In recent years, deep learning has shown promise in various automated medical image classification tasks.This study explores the feasibility of using a deep learning model to classify entry and exit gunshot wounds in digital color images. A collection of 2418 images of entrance and exit gunshot wounds were procured. Of these, 2028 entrance and 1314 exit wounds were cropped, focusing on the area around each gunshot wound. A ConvNext Tiny deep learning model was trained using the Fastai deep learning library, with a train/validation split ratio of 70/30, until a maximum validation accuracy of 92.6% was achieved. An additional 415 entrance and 293 exit wound images were collected for the test (holdout) set. The model achieved an accuracy of 87.99%, precision of 83.99%, recall of 87.71%, and F1-score 85.81% on the holdout set. Correctly classified were 88.19% of entrance wounds and 87.71% of exit wounds. The results are comparable to what a forensic pathologist can achieve without other morphologic cues. This study represents one of the first applications of artificial intelligence to the field of forensic pathology. This work demonstrates that deep learning models can discern entrance and exit gunshot wounds in digital images with high accuracy.

Token labeling-guided multi-scale medical image classification

Vision transformer has been widely used in medical image analysis. However, in most of current methods, only the class token is concerned during training, while the output patch tokens’ information is not well utilized. To track this problem, we propose a two-stage token labeling guided multi-scale model for medical image classification. In the first stage, we pre-train a classification model to extract critical areas as token labeling. In the second stage, we adopt coarse and fine branches to encode visual features, which adapts to the various lesions in medical images. Then, the class token output by each branch is fused for classification. The token labeling is used to supervise the representation learning of patch tokens, which can integrate the local information into the learning. The experimental results on Laryngoscope8, ISIC 2018, and REFUGE data sets show that after adding token labeling, this dual-branch classification model achieves significantly better performance than the model using only class token loss, which demonstrates the effectiveness of our method for medical image classification tasks.

Multiple instance learning for medical image classification based on instance importance

Multiple Instance Learning (MIL) is designed to classify instances where class labels are associated with sets of instances, a common occurrence in biomedical data, especially when multiple images are derived from a single object measurement. Among the various MIL techniques, instance-based classification emerges as the main approach. Individual instances are assigned class labels, and the final bag-level classification is determined through an aggregation process, usually based on average or maximum values. While instance-based models may exhibit slightly lower performance compared to more general methods, they offer the unique advantage of interpretability, allowing the decision-making process to be traced to each individual instance. This article presents an innovative instance-based approach that not only assigns class labels to instances, but also gives them meaning. The final aggregation uses a weighted average, allowing the model to identify instances carrying valuable information while distinguishing those that are redundant in the classification process. This enhancement gives the model the ability to recognize the meaning of individual instances, thereby increasing both its overall interpretability and classification performance.

Growth threshold for pseudo labeling and pseudo label dropout for semi-supervised medical image classification

Semi-supervised learning (SSL) provides methods to improve model performance through unlabeled samples. In medical image analysis, the challenges of multi-category classification and imbalance learning must be addressed effectively. Pseudo labeling is not specifically designed for multi-category and category imbalance problems. In this paper, we propose the Growth Threshold for Pseudo Labeling (GTPL) and Pseudo Label Dropout (PLD), which can be used separately or in combination. GTPL changes the threshold value of each category by combining the confidence of labeled and unlabeled samples. PLD alleviates the category imbalance by randomly discarding some of the pseudo labels. We apply GTPL and PLD to FixMatch and CoMatch and effectively improve their semi-supervised classification performance. We validate the effectiveness of our approach in skin lesion diagnosis on two long-tailed distributions of public medical images on the ISIC 2018 and ISIC 2019 challenge datasets, obtaining AUCs of 89.19%, 92.71%, 94.71%, and 94.76%, respectively, on four scales of labeled data from ISIC 2018.

Distributed Federated Learning-Based Deep Learning Model for Privacy MRI Brain Tumor Detection

Distributed training can facilitate the processing of large medical image datasets, and improve the accuracy and efficiency of disease diagnosis while protecting patient privacy, which is crucial for achieving efficient medical image analysis and accelerating medical research progress. This paper presents an innovative approach to medical image classification, leveraging Federated Learning (FL) to address the dual challenges of data privacy and efficient disease diagnosis. Traditional Centralized Machine Learning models, despite their widespread use in medical imaging for tasks such as disease diagnosis, raise significant privacy concerns due to the sensitive nature of patient data. As an alternative, FL emerges as a promising solution by allowing the training of a collective global model across local clients without centralizing the data, thus preserving privacy. Focusing on the application of FL in Magnetic Resonance Imaging (MRI) brain tumor detection, this study demonstrates the effectiveness of the Federated Learning framework coupled with EfficientNet-B0 and the FedAvg algorithm in enhancing both privacy and diagnostic accuracy. Through a meticulous selection of preprocessing methods, algorithms, and hyperparameters, and a comparative analysis of various Convolutional Neural Network (CNN) architectures, the research uncovers optimal strategies for image classification. The experimental results reveal that EfficientNet-B0 outperforms other models like ResNet in handling data heterogeneity and achieving higher accuracy and lower loss, highlighting the potential of FL in overcoming the limitations of traditional models. The study underscores the significance of addressing data heterogeneity and proposes further research directions for broadening the applicability of FL in medical image analysis.

Enhancing Medical Image Classification: A Deep Learning Perspective with Multi Wavelet Transform

Classification of medical images is a very important area of research for both the medical industry and academia. In recent years, automated classification algorithms have become very important in most medical applications, saving time and effort, such as disease detection and diagnostic radiology. Deep learning offers a plethora of advantages when applied to medical image classification, revolutionizing medical diagnosis and patient care. In this study, deep convolutional neural networks (DCNNs) is used to classify medical im-ages and multi-wavelet transform will be applied to extract features. The proposed method aims to improve medical image classification accuracy, thereby assisting healthcare professionals in making more accurate and efficient diagnoses. DCNNs based on the VGG16 model were trained and used in this study. Combining VGG16, a powerful convolutional neural network (CNN), with multiwavelet transform offers several advantages for image processing and analysis tasks, particularly in areas like image classification and feature extraction. To evaluate the performance of the proposed method six publicly available brain tumour MRI datasets are analysed with DCNNs. A fully connected layer is used to categorize the extracted features. According to the results, the deep CNN model combined with the multi-wavelet trans-form achieves an impressive accuracy of 96.43 %. It is evident from this high level of accuracy that the proposed approach is effective in accurately classifying medical images.

Optimizing brain tumor classification with hybrid CNN architecture: Balancing accuracy and efficiency through oneAPI optimization

A brain tumour is a malignant condition that spreads extremely quickly and requires rapid detection. In recent years, it has become apparent that deep learning is a promising technique for classifying brain tumours. To identify these tumours, this work proposes a hybrid CNN architecture, implementing InceptionV3, ResNet-50, VGG16, and DenseNet. To evaluate this approach, a total of 3929 images was obtained from Kaggle, including 2556 non-tumorigenic and 1373 tumorigenic specimens. Initially, the method extracts features from MRI pictures, then segments the tumour using the mask images. Subsequently, the segmented tumour image is fused with the original image, and finally the fused image is classified with the assistance of four distinct CNN models, specifically InceptionV3, ResNet, DenseNet, and VGG16. To enhance the performance of the hybrid architecture, these models were further optimized with oneAPI, which allows a comparative analysis of each individual model for classification. To evaluate the effectiveness of medical image classification models, a variety of evaluation criteria have been utilized extensively. We highlight the use of mean Intersection over Union (mIoU) as a more acceptable and intuitive statistic for evaluating the balance between true positives, false positives, and false negatives. This is in response to recent breakthroughs that have been made.

Review of AlexNet for Medical Image Classification

In recent years, the rapid development of deep learning has led to a wide range of applications in the field of medical image classification. The variants of neural network models with ever-increasing performance share some commonalities: to try to mitigate overfitting, improve generalization, avoid gradient vanishing and exploding, etc. AlexNet first utilizes the dropout technique to mitigate overfitting and the ReLU activation function to avoid gradient vanishing. Therefore, we focus our discussion on AlexNet, which has contributed greatly to the development of CNNs in 2012. After reviewing over 40 papers, including journal papers and conference papers, we give a narrative on the technical details, advantages, and application areas of AlexNet.

Improving vision transformer for medical image classification via token-wise perturbation

Transformer has achieved impressive successes for various computer vision tasks. However, most of existing studies require to pretrain the Transformer backbone on a large-scale labeled dataset (e.g., ImageNet) for achieving satisfactory performance, which is usually unavailable for medical images. Additionally, due to the gap between medical and natural images, the improvement generated by the ImageNet pretrained weights significantly degrades while transferring the weights to medical image processing tasks. In this paper, we propose Bootstrap Own Latent of Transformer (BOLT), a self-supervised learning (SSL) approach specifically for medical image classification with the Transformer backbone. Our BOLT consists of two networks, namely online and target branches, for self-supervised representation learning. Concretely, the online network is trained to predict the target network representation of the same patch embedding tokens with a different perturbation. To maximally excavate the impact of Transformer from limited medical data, we propose an auxiliary difficulty ranking task. The Transformer is enforced to identify which branch (i.e., online/target) is processing the more difficult perturbed tokens. Overall, the Transformer endeavours itself to distill the transformation-invariant features from the perturbed tokens to simultaneously achieve difficulty measurement and maintain the consistency of self-supervised representations. The proposed BOLT is evaluated on three medical image processing tasks, i.e., skin lesion classification, knee fatigue fracture grading and diabetic retinopathy grading. The experimental results validate the superiority of our BOLT for medical image classification, compared to ImageNet pretrained weights and state-of-the-art SSL approaches.

Convolutional Neural Networks for Medical Image Segmentation and Classification: A Review

Convolutional Neural Networks for Medical Image Segmentation and Classification: A Review

MEHW‐SVM multi‐kernel approach for improved brain tumour classification

AbstractThe human brain, the primary constituent of the nervous system, exhibits distinctive complexities that present considerable difficulties for healthcare practitioners, specifically in categorizing brain tumours. Magnetic resonance imaging is a widely favoured imaging modality for detecting brain tumours due to its extensive range of image characteristics and utilization of non‐ionizing radiation. The primary objective of the current investigation is to differentiate between three distinct classifications of brain tumours by introducing a novel methodology. The utilization of a combined feature extraction technique that integrates novel global grey level co‐occurrence matrix and local binary patterns is employed, thereby offering a comprehensive representation of the structural and textural information contained within the images. Principal component analysis is used to improve the model's efficiency for effective feature selection and dimensionality reduction. This study presents a novel framework incorporating four separate kernel functions, Minkowski–Gaussian, exponential support vector machine (SVM), histogram intersection SVM, and wavelet kernel, into a SVM classifier. The ensemble kernel employed in this study is specifically designed to classify glioma, meningioma, and pituitary tumours. Its implementation enhances the model's robustness and adaptability, surpassing the performance of conventional single‐kernel SVM approaches. This study substantially contributes to medical image classification by utilizing innovative kernel functions and advanced machine‐learning techniques. The findings demonstrate the potential for enhanced diagnostic accuracy in brain tumour cases. The presented approach shows promise in effectively addressing the intricate challenges associated with classifying brain tumours.