BEEx Is an Open-Source Tool That Evaluates Batch Effects in Medical Images to Enable Multicenter Studies.

Medical imaging plays an essential role in modern healthcare, providing non-invasive tools for diagnosing and monitoring various medical conditions. However, the resolution limitations of imaging hardware often result in suboptimal images, which can hinder the precision of clinical decision-making. Single image super-resolution (SISR) techniques offer a solution by reconstructing high-resolution (HR) images from low-resolution (LR) counterparts, enhancing the visual quality of medical images. In this paper, we propose an enhanced Residual Feature Learning Network (RFLN) tailored specifically for medical imaging. Our contributions include replacing the residual local feature blocks with standard residual blocks, increasing the model depth for improved feature extraction, and incorporating enhanced spatial attention (ESA) mechanisms to refine the feature selection. Extensive experiments on medical imaging datasets demonstrate that the proposed model achieves superior performance in terms of both quantitative metrics, such as PSNR and SSIM, and qualitative visual quality compared to existing state-of-the-art models. The enhanced RFLN not only effectively mitigates noise but also preserves critical anatomical details, making it a promising solution for high-precision medical imaging applications.

MG-CNFNet: A multiple grained channel normalized fusion networks for medical image deblurring

Deep learning (DL) has been applied in proofs of concept across biomedical imaging, including across modalities and medical specialties. Labeled data are critical to training and testing DL models, but human expert labelers are limited. In addition, DL traditionally requires copious training data, which is computationally expensive to process and iterate over. Consequently, it is useful to prioritize using those images that are most likely to improve a model's performance, a practice known as instance selection. The challenge is determining how best to prioritize. It is natural to prefer straightforward, robust, quantitative metrics as the basis for prioritization for instance selection. However, in current practice, such metrics are not tailored to, and almost never used for, image datasets. To address this problem, we introduce ENRICH-Eliminate Noise and Redundancy for Imaging Challenges-a customizable method that prioritizes images based on how much diversity each image adds to the training set. First, we show that medical datasets are special in that in general each image adds less diversity than in nonmedical datasets. Next, we demonstrate that ENRICH achieves nearly maximal performance on classification and segmentation tasks on several medical image datasets using only a fraction of the available images and without up-front data labeling. ENRICH outperforms random image selection, the negative control. Finally, we show that ENRICH can also be used to identify errors and outliers in imaging datasets. ENRICH is a simple, computationally efficient method for prioritizing images for expert labeling and use in DL.

The widespread applications of machine learning algorithms in high-stakes fields have demanded the development of effective explanation methods to help decision-makers interpret the black-box models. Post-hoc local feature attribution methods are among the most prevalent explainability approaches. These methods are often evaluated through human-subject studies, whose heterogeneous designs make results difficult to compare and sometimes unreliable. Although several quantitative evaluation metrics have been proposed, many suffer from different limitations, such as computational complexity and generalizability to complicated real-world datasets. Additionally, to our knowledge, there is no current work that establishes a comprehensive and reliable benchmark to evaluate all aspects of the methods in high-stakes settings. In this work, we introduce and assess a benchmark including a suite of quantitative metrics that provide a thorough evaluation of post-hoc feature attribution methods on 2 medical imaging datasets. We formalize 4 evaluation lenses, propose a new metric, and also apply a novel technique to prepare AI explanation artifacts for processing by explanation evaluation metrics. Our results show cross-dataset consistency, meaningful-random separation, sensitivity to model strength, and the value of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{4}$</tex> lenses. We provide reliable statistics that can guide the choice of explanation algorithms in medical imaging and beyond and help prioritize methods before investing in costly human-subject experiments.

Anomaly detection (AD) is a challenging problem in computer vision. Particularly in the field of medical imaging, AD poses even more challenges due to a number of reasons, including insufficient availability of ground truth (annotated) data. In recent years, AD models based on generative adversarial networks (GANs) have made significant progress. However, their effectiveness in biomedical imaging remains underexplored. In this paper, we present an overview of using GANs for AD, as well as an investigation of state-of-the-art GAN-based AD methods for biomedical imaging and the challenges encountered in detail. We have also specifically investigated the advantages and limitations of AD methods on medical image datasets, conducting experiments using 3 AD methods on 7 medical imaging datasets from different modalities and organs/tissues. Given the highly different findings achieved across these experiments, we further analyzed the results from both data-centric and model-centric points of view. The results showed that none of the methods had reliable performance for detecting abnormalities in medical images. Factors such as the number of training samples, the subtlety of the anomaly, and the dispersion of the anomaly in the images are among the phenomena that highly impact the performance of the AD models. The obtained results were highly variable (AUC: 0.475-0.991; Sensitivity: 0.17-0.98; Specificity: 0.14-0.97). In addition, we provide recommendations for deployment of AD models in medical imaging and foresee important research directions.

The federated learning (FL) paradigm is well-suited for the field of medical image analysis, as it can effectively cope with machine learning on isolated multi-center data while protecting the privacy of participating parties. However, current research on optimization algorithms in FL often focuses on limited datasets and scenarios, primarily centered around natural images, with insufficient comparative experiments in medical contexts. In this work, we conduct a comprehensive evaluation of several state-of-the-art FL algorithms in the context of medical imaging. We conduct a fair comparison of classification models trained using various FL algorithms across multiple medical imaging datasets. Additionally, we evaluate system performance metrics, such as communication cost and computational efficiency, while considering different FL architectures. Our findings show that medical imaging datasets pose substantial challenges for current FL optimization algorithms. No single algorithm consistently delivers optimal performance across all medical FL scenarios, and many optimization algorithms may under-perform when applied to these datasets. Our experiments provide a benchmark and guidance for future research and application of FL in medical imaging contexts. Furthermore, we propose an efficient and robust method that combines generative techniques using denoising diffusion probabilistic models with label smoothing to augment datasets, widely enhancing the performance of FL on classification tasks across various medical imaging datasets.

Deep learning methods, especially convolutional neural networks, have become more and more popular in medical image classifications. However, training a deep neural network from scratch can be a luxury for many medical image datasets as the process requires a large and well-balanced sample to output satisfactory results. Unlike natural image datasets, medical images are expensive to collect owing to labor and equipment costs. Besides, the class labels in medical image datasets are usually severely imbalanced subject to the availability of patients. Further, aggregating medical images from multiple sources can be challenging due to policy restrictions, privacy concerns, communication costs, and data heterogeneity caused by equipment differences and labeling discrepancies. In this paper, we propose to address these issues with the help of transfer learning and artificial samples created by generative models. Instead of requesting medical images from source data, our method only needs a parsimonious supplement of model parameters pre-trained on the source data. The proposed method preserves the data privacy in the source data and significantly reduces the communication cost. Our study shows transfer learning together with artificial samples can improve the pneumonia classification accuracy on a small but heavily imbalanced chest X-ray image dataset by \(11.53\%\) which performs even better than directly augmenting that source data into the training process.KeywordsDeep learningGenerative modelsMedical image classificationPrivacy preservationTransfer learning

Microarray batch effect (BE) has been the primary bottleneck for large-scale integration of data from multiple experiments. Current BE correction methods either need known batch identities (ComBat) or have the potential to overcorrect, by removing true but unknown biological differences (Surrogate Variable Analysis SVA). It is well known that experimental conditions such as array or reagent batches, PCR amplification or ozone levels can affect the measured expression levels; often the direction of perturbation of the measured expression is the same in different datasets. However, there are no BE correction algorithms that attempt to estimate the individual effects of technical differences and use them to correct expression data. In this manuscript, we show that a set of signatures, each of which is a vector the length of the number of probes, calculated on a reference set of microarray samples can predict much of the batch effect in other validation sets. We present a rationale of selecting a reference set of samples designed to estimate technical differences without removing biological differences. Putting both together, we introduce the Batch Effect Signature Correction (BESC) algorithm that uses the BES calculated on the reference set to efficiently predict and remove BE. Using two independent validation sets, we show that BESC is capable of removing batch effect without removing unknown but true biological differences. Much of the variations due to batch effect is shared between different microarray datasets. That shared information can be used to predict signatures (i.e. directions of perturbation) due to batch effect in new datasets. The correction can be precomputed without using the samples to be corrected (blind), done on each sample individually (single sample) and corrects only known technical effects without removing known or unknown biological differences (conservative). Those three characteristics make it ideal for high-throughput correction of samples for a microarray data repository. We also compare the performance of BESC to three other batch correction methods: SVA, Removing Unwanted Variation (RUV) and Hidden Covariates with Prior (HCP). An R Package besc implementing the algorithm is available from http://explainbio.com.

Multimodal medical image fusion plays an important role in clinical applications. However, multimodal medical image fusion methods ignore the feature dependence among modals, and the feature fusion ability with different granularity is not strong. A Long-Range Correlation-Guided Dual-Encoder Fusion Network for Medical Images is proposed in this paper. The main innovations of this paper are as follows: Firstly, A Cross-dimension Multi-scale Feature Extraction Module (CMFEM) is designed in the encoder, by extracting multi-scale features and aggregating coarse-to-fine features, the model realizes fine-grained feature enhancement in different modalities. Secondly, a Long-range Correlation Fusion Module (LCFM) is designed, by calculating the long-range correlation coefficient between local features and global features, the same granularity features are fused by the long-range correlation fusion module. long-range dependencies between modalities are captured by the model, and different granularity features are aggregated. Finally, this paper is validated on clinical multimodal lung medical image dataset and brain medical data dataset. On the lung medical image dataset, IE, AG, {textbf {Q}}^{{textbf {AB/F}}}, and EI metrics are improved by 4.53%, 4.10%, 6.19%, and 6.62% respectively. On the brain medical image dataset, SF, VIF, and {textbf {Q}}^{{textbf {AB/F}}} metrics are improved by 3.88%, 15.71%, and 7.99% respectively. This model realizes better fusion performance, which plays an important role in the fusion of multimodal medical images.

Multimodal imaging is gaining in importance in the field of personalized medicine and can be described as a current trend in medical imaging diagnostics. The segmentation, classification and analysis of tissue structures is essential. The goal of this study is the evaluation of established segmentation methods on different medical image data sets acquired with different diagnostic procedures. Established segmentation methods were implemented using the latest state of the art and applied to medical image data sets. In order to benchmark the segmentation performance quantitatively, medical image data sets were superimposed with artificial Gaussian noise, and the mis-segmentation as a function of the image SNR or CNR was compared to a gold standard. The evaluation of the image segmentation showed that the best results of pixel-based segmentation (&lt; 3%) can be achieved with methods of machine learning, multithreshold and advanced level-set method - even at high artificial noise (SNR&lt; 18). Finally, the complexity of the object geometry and the contrast of the ROI to the surrounding tissue must also be considered to select the best segmentation algorithm.

Graph neural networks (GNNs) have achieved extraordinary enhancements in various areas including the fields medical imaging and network neuroscience where they displayed a high accuracy in diagnosing challenging neurological disorders such as autism. In the face of medical data scarcity and high-privacy, training such data-hungry models remains challenging. Federated learning brings an efficient solution to this issue by allowing to train models on multiple datasets, collected independently by different hospitals, in fully data-preserving manner. Although both state-of-the-art GNNs and federated learning techniques focus on boosting classification accuracy, they overlook a critical unsolved problem: investigating the reproducibility of the most discriminative biomarkers (i.e., features) selected by the GNN models within a federated learning paradigm. Quantifying the reproducibility of a predictive medical model against perturbations of training and testing data distributions presents one of the biggest hurdles to overcome in developing translational clinical applications. To the best of our knowledge, this presents the first work investigating the reproducibility of federated GNN models with application to classifying medical imaging and brain connectivity datasets. We evaluated our framework using various GNN models trained on medical imaging and connectomic datasets. More importantly, we showed that federated learning boosts both the accuracy and reproducibility of GNN models in such medical learning tasks. Our source code is available at https://github.com/basiralab/reproducibleFedGNN.KeywordsGraph neural networksFederated LearningReproducibilityBrain connectivity graphsPredictive medicine

Abstract: A brain tumour is described by the presence of abnormal cells in the brain's tissues. Brain tumours can be benign (not cancerous) or malignant (cancerous). The malignant brain tumour is one of the leading and common cancers in the world. There are two types of tumours, primary tumours that develop in the brain and secondary tumours that start in another region of the body and then spread to the brain. The precise identification of the size and location of a brain tumour is crucial in the diagnosis of a brain tumour and is often diagnosed with magnetic resonance imaging (MRI). This book chapter discusses the major types of brain tumours and the advancements in computeraided approaches for detecting brain tumours. The manuscript gives an overview of various recent machine learning and medical image processing approaches developed recently for the identification and classification of brain tumours. Several medical image dataset available for the research works in this domain is also briefed in this article. The major research challenges which can be addressed by the researchers in the domain of brain tumour detection are also discussed in this article. Method: This manuscript mainly discusses various medical image processing methods to detect brain tumors. Results: The manuscript discusses the efficiency of the existing schemes and the key areas where further improvement is required Conclusion: This manuscript gives an overview of various categories of brain tumors, existing methods to detect brain tumors, the critical challenges in this domain and the medical image dataset available for the study.

Unlocking the Potential of Medical Imaging with ChatGPT's Intelligent Diagnostics

U-Net convolutional neural networks have become a cornerstone in medical image processing, particularly for complex segmentation tasks. However, with the proliferation of various U-Net variants, it is imperative to evaluate their performance across diverse medical datasets to determine the most suitable architecture for specific applications. This study presents a comprehensive empirical analysis of six U-Net variants namely; U-Net, U-Net++, ResU-Net, TransU-Net, V-Net, and U-Net 3+, focusing on their effectiveness in medical image segmentation. The research used two publicly available medical image datasets: the Breast Cancer Ultrasound dataset and the Brain MRI FLAIR dataset. These datasets were selected to represent a range of challenges in medical imaging, including varying levels of noise, contrast, color channels, and lesion visibility. Each U-Net variant was trained and evaluated under consistent experimental conditions, with the Dice Similarity Coefficient (DSC), Jaccard Index and accuracy serving as the primary evaluation metrics. The findings reveal that while all U-Net variants exhibit strong performance, ResU-Net consistently outperformed other architectures across both datasets. Specifically, ResU-Net achieved the highest Jaccard Index and DSC values, recording 0.5158 and 0.5930 for the Breast Cancer Ultrasound dataset and 0.7923 and 0.8730 for the Brain MRI dataset, respectively. Additionally, ResU-Net demonstrated the highest accuracy of 96.20% in the Breast Ultrasound dataset and emerged as the first runner-up in the Brain MRI dataset with an accuracy of 99.70%, trailing slightly behind U-Net++, which achieved 97.72%. Conversely, U-Net 3+ registered the lowest performance in both datasets. ResU-Net's superior performance suggests that its residual connections enhance feature propagation and mitigate the vanishing gradient problem, leading to better segmentation outcomes. Conversely, U-Net 3+ showed lower performance, indicating that complexity doesn’t always equate to better results. This study offers insights into the strengths and weaknesses of different U-Net variants, guiding practitioners in choosing the most suitable model for their specific applications especially in medical image processing domain.

Recently, due to the increasing requirements of medical imaging applications and the professional requirements of annotating medical images, few-shot learning has gained increasing attention in the medical image semantic segmentation field. To perform segmentation with limited number of labeled medical images, most existing studies use Prototypical Networks (PN) and have obtained compelling success. However, these approaches overlook the query imagefeatures extracted from the proposed representation network, failing to preserving the spatial connection between query and support images. In this paper, we propose a novel self-supervised few-shot medical image segmentation network and introduce a novel Cycle-Resemblance Attention (CRA) module to fully leverage the pixel-wise relation between query and support medical images. Notably, we first line up multiple attention blocks to refine more abundant relation information. Then, we present CRAPNet by integrating the CRA module with a classic prototype network, where pixel-wise relations between query and support features are well recaptured for segmentation. Extensive experiments on two different medical image datasets, e.g., abdomen MRI and abdomen CT, demonstrate the superiority of our model over existing state-of-the-art methods.