Structure and Illumination Constrained GAN for Medical Image Enhancement

Unified Framework for Enhancement of Low-Quality Fundus Images

Compared to desktop fundus cameras, handheld ones offer portability and affordability, although they often produce lower-quality images. This paper primarily addresses the issue of reduced image quality commonly associated with images captured by handheld fundus cameras. We first collected 538 fundus images obtained from handheld devices to form a dataset called Mule. A unified framework that consists of three main modules is then proposed to enhance the quality of fundus images. The Light Balance Module is employed first to suppress overexposure and underexposure. This is followed by the Super Resolution Module to enhance vascular details. Finally, the Vessel Enhancement Module is applied to improve image contrast. And a special preservation strategy is additionally applied to retain mocular features in the final fundus image. Objective evaluations demonstrate that the proposed framework yields the most promising results. Further experiments also suggest that it improves accuracy in downstream tasks, such as vessel segmentation, optic disc/optic cup detection, macula detection, and fundus image quality assessment. Our code is available at: https://github.com/Alen880/UFELQ.

General retinal image enhancement via reconstruction: Bridging distribution shifts using latent diffusion adaptors

Deep learning-based fundus image enhancement has attracted extensive research attention recently, which has shown remarkable effectiveness in improving the visibility of low-quality images. However, these methods are often constrained to specific datasets and degradations, leading to poor generalization capabilities and having challenges in the fine-tuning process. Therefore, a general method for fundus image enhancement is proposed for improved generalizability and flexibility, which decomposes the enhancement task into reconstruction and adaptation phases. In the reconstruction phase, self-supervised training with unpaired data is employed, allowing the utilization of extensive public datasets to improve the generalizability of the model. During the adaptation phase, the model is fine-tuned according to the target datasets and their degradations, utilizing the pre-trained weights from the reconstruction. The proposed method improves the feasibility of latent diffusion models for retinal image enhancement. Adaptation loss and enhancement adaptor are proposed in autoencoders and diffusion networks for fewer paired training data, fewer trainable parameters, and faster convergence compared with training from scratch. The proposed method can be easily fine-tuned and experiments demonstrate the adaptability for different datasets and degradations. Additionally, the reconstruction-adaptation framework can be utilized in different backbones and other modalities, which shows its generality.

Hybrid CNN-Mamba model for multi-scale fundus image enhancement

This study proposes a multi-scale fundus image enhancement approach that combines CNN with Mamba, demonstrating clear superiority across multiple benchmarks. The model consistently achieves top performance on public datasets, with the lowest FID and KID scores, and the highest PSNR and SSIM values, particularly excelling at larger image resolutions. Notably, its performance improves as the image size increases, with several metrics reaching optimal values at 1024 × 1024 resolution. Scale generalizability further highlights the model's exceptional structural preservation capability. Additionally, its high VSD and IOU scores in segmentation tasks further validate its practical effectiveness, making it a valuable tool for enhancing fundus images and improving diagnostic accuracy.

Adaptive enhancement of shoulder x-ray images using tissue attenuation and type-II fuzzy sets

Shoulder X-ray images typically have low contrast and high noise levels, making it challenging to distinguish and identify subtle anatomical structures. While existing image enhancement techniques are effective in improving contrast, they often overlook the enhancement of sharpness, especially when amplifying blurring and noise. These techniques may improve detail contrast but fail to maintain overall image clarity and the distinction between the target and the background. To address these issues, we propose a novel image enhancement method aimed at simultaneously improving both the contrast and sharpness of shoulder X-ray images. The method integrates automatic tissue attenuation techniques, which enhance the image contrast by removing non-essential tissue components while preserving important tissues and bones. Additionally, we apply an improved Type-II fuzzy set algorithm to further optimize image sharpness. By simultaneously enhancing contrast and sharpness, the method significantly improves image quality and detail distinguishability. When tested on certain images from the MURA dataset, the proposed method achieved the best or second-best results, outperforming five no-reference image quality assessment metrics. In comparative studies, the method demonstrated significant performance advantages over 10 contemporary X-ray image enhancement algorithms and was validated through ablation experiments to confirm the effectiveness of each module.

HiFi-Syn: Hierarchical granularity discrimination for high-fidelity synthesis of MR images with structure preservation

Synthesizing medical images while preserving their structural information is crucial in medical research. In such scenarios, the preservation of anatomical content becomes especially important. Although recent advances have been made by incorporating instance-level information to guide translation, these methods overlook the spatial coherence of structural-level representation and the anatomical invariance of content during translation. To address these issues, we introduce hierarchical granularity discrimination, which exploits various levels of semantic information present in medical images. Our strategy utilizes three levels of discrimination granularity: pixel-level discrimination using a Brain Memory Bank, structure-level discrimination on each brain structure with a re-weighting strategy to focus on hard samples, and global-level discrimination to ensure anatomical consistency during translation. The image translation performance of our strategy has been evaluated on three independent datasets (UK Biobank, IXI, and BraTS 2018), and it has outperformed state-of-the-art algorithms. Particularly, our model excels not only in synthesizing normal structures but also in handling abnormal (pathological) structures, such as brain tumors, despite the variations in contrast observed across different imaging modalities due to their pathological characteristics. The diagnostic value of synthesized MR images containing brain tumors has been evaluated by radiologists. This indicates that our model may offer an alternative solution in scenarios where specific MR modalities of patients are unavailable. Extensive experiments further demonstrate the versatility of our method, providing unique insights into medical image translation.

MVE-Net: A label-free microscopic image visual enhancement network via mRetinex and nonreference loss guidance

Label-free microscopic cell image analysis (segmentation, detection, counting, e.g.) is elementary for unravelling the biological functions of cells and their organelles. However, low contrast, darker brightness, background inhomogeneous, and weak edges of cells cause challenges in subsequent cell image analysis processes. To address these challenges, a Microscopic Visual Enhancement Network (MVE-Net) is proposed to improve microscopic visual effects through pre-enhancement and enhancement processes. In the pre-enhancement stage, to overcome the difficulty of acquiring paired or unpaired images for training, the mRetinex block is proposed to guide the pre-enhancement network to image contrast, cell details, and structural features. Furthermore, a multi-scale extraction module is employed to extract and fuse cell texture and structural features from the pre-enhanced images at various scales, guiding the generator training. In the enhancement stage, a nonreference loss block is designed, incorporating spatial consistency, uneven illumination smoothness, and exposure adjustment loss terms, to further enhance the contrast between cells and the background, smooth the inhomogeneous background, and adjust overall image brightness, thereby guiding the generator's enhancement process and improving the visual effect of microscopic images. Experiments on the LIVECell and PNT1A datasets demonstrate that MVE-Net outperforms state-of-the-art image enhancement methods, significantly improving image contrast, brightness, cell detail, and structural features without the need for paired or unpaired reference standard images for training.

HQG-Net: Unpaired Medical Image Enhancement With High-Quality Guidance

Unpaired medical image enhancement (UMIE) aims to transform a low-quality (LQ) medical image into a high-quality (HQ) one without relying on paired images for training. While most existing approaches are based on Pix2Pix/CycleGAN and are effective to some extent, they fail to explicitly use HQ information to guide the enhancement process, which can lead to undesired artifacts and structural distortions. In this article, we propose a novel UMIE approach that avoids the above limitation of existing methods by directly encoding HQ cues into the LQ enhancement process in a variational fashion and thus model the UMIE task under the joint distribution between the LQ and HQ domains. Specifically, we extract features from an HQ image and explicitly insert the features, which are expected to encode HQ cues, into the enhancement network to guide the LQ enhancement with the variational normalization module. We train the enhancement network adversarially with a discriminator to ensure the generated HQ image falls into the HQ domain. We further propose a content-aware loss to guide the enhancement process with wavelet-based pixel-level and multiencoder-based feature-level constraints. Additionally, as a key motivation for performing image enhancement is to make the enhanced images serve better for downstream tasks, we propose a bi-level learning scheme to optimize the UMIE task and downstream tasks cooperatively, helping generate HQ images both visually appealing and favorable for downstream tasks. Experiments on three medical datasets verify that our method outperforms existing techniques in terms of both enhancement quality and downstream task performance. The code and the newly collected datasets are publicly available at https://github.com/ChunmingHe/HQG-Net.

Fundus Image Enhancement Through Direct Diffusion Bridges

We propose FD3, a fundus image enhancement method based on direct diffusion bridges, which can cope with a wide range of complex degradations, including haze, blur, noise, and shadow. We first propose a synthetic forward model through a human feedback loop with board-certified ophthalmologists for maximal quality improvement of low-quality in-vivo images. Using the proposed forward model, we train a robust and flexible diffusion-based image enhancement network that is highly effective as a stand-alone method, unlike previous diffusion model-based approaches which act only as a refiner on top of pre-trained models. Through extensive experiments, we show that FD3 establishes superior quality not only on synthetic degradations but also on in vivo studies with low-quality fundus photos taken from patients with cataracts or small pupils.

Redefining retinal vessel segmentation: empowering advanced fundus image analysis with the potential of GANs

Retinal vessel segmentation is a critical task in fundus image analysis, providing essential insights for diagnosing various retinal diseases. In recent years, deep learning (DL) techniques, particularly Generative Adversarial Networks (GANs), have garnered significant attention for their potential to enhance medical image analysis. This paper presents a novel approach for retinal vessel segmentation by harnessing the capabilities of GANs. Our method, termed GANVesselNet, employs a specialized GAN architecture tailored to the intricacies of retinal vessel structures. In GANVesselNet, a dual-path network architecture is employed, featuring an Auto Encoder-Decoder (AED) pathway and a UNet-inspired pathway. This unique combination enables the network to efficiently capture multi-scale contextual information, improving the accuracy of vessel segmentation. Through extensive experimentation on publicly available retinal datasets, including STARE and DRIVE, GANVesselNet demonstrates remarkable performance compared to traditional methods and state-of-the-art deep learning approaches. The proposed GANVesselNet exhibits superior sensitivity (0.8174), specificity (0.9862), and accuracy (0.9827) in segmenting retinal vessels on the STARE dataset, and achieves commendable results on the DRIVE dataset with sensitivity (0.7834), specificity (0.9846), and accuracy (0.9709). Notably, GANVesselNet achieves remarkable performance on previously unseen data, underscoring its potential for real-world clinical applications. Furthermore, we present qualitative visualizations of the generated vessel segmentations, illustrating the network's proficiency in accurately delineating retinal vessels. In summary, this paper introduces GANVesselNet, a novel and powerful approach for retinal vessel segmentation. By capitalizing on the advanced capabilities of GANs and incorporating a tailored network architecture, GANVesselNet offers a quantum leap in retinal vessel segmentation accuracy, opening new avenues for enhanced fundus image analysis and improved clinical decision-making.

Explainable machine learning for assessing upper respiratory tract of racehorses from endoscopy videos

Laryngeal hemiplegia (LH) is a major upper respiratory tract (URT) complication in racehorses. Endoscopy imaging of horse throat is a gold standard for URT assessment. However, current manual assessment faces several challenges, stemming from the poor quality of endoscopy videos and subjectivity of manual grading. To overcome such limitations, we propose an explainable machine learning (ML)-based solution for efficient URT assessment. Specifically, a cascaded YOLOv8 architecture is utilized to segment the key semantic regions and landmarks per frame. Several spatiotemporal features are then extracted from key landmarks points and fed to a decision tree (DT) model to classify LH as Grade 1,2,3 or 4 denoting absence of LH, mild, moderate, and severe LH, respectively. The proposed method, validated through 5-fold cross-validation on 107 videos, showed promising performance in classifying different LH grades with 100%, 91.18%, 94.74% and 100% sensitivity values for Grade 1 to 4, respectively. Further validation on an external dataset of 72 cases confirmed its generalization capability with 90%, 80.95%, 100%, and 100% sensitivity values for Grade 1 to 4, respectively. We introduced several explainability related assessment functions, including: (i) visualization of YOLOv8 output to detect landmark estimation errors which can affect the final classification, (ii) time-series visualization to assess video quality, and (iii) backtracking of the DT output to identify borderline cases. We incorporated domain knowledge (e.g., veterinarian diagnostic procedures) into the proposed ML framework. This provides an assistive tool with clinical-relevance and explainability that can ease and speed up the URT assessment by veterinarians.

Unsupervised adversarial neural network for enhancing vasculature in photoacoustic tomography images using optical coherence tomography angiography

Photoacoustic tomography (PAT) is a powerful imaging modality for visualizing tissue physiology and exogenous contrast agents. However, PAT faces challenges in visualizing deep-seated vascular structures due to light scattering, absorption, and reduced signal intensity with depth. Optical coherence tomography angiography (OCTA) offers high-contrast visualization of vasculature networks, yet its imaging depth is limited to a millimeter scale. Herein, we propose OCPA-Net, a novel unsupervised deep learning method that utilizes the rich vascular feature of OCTA to enhance PAT images. Trained on unpaired OCTA and PAT images, OCPA-Net incorporates a vessel-aware attention module to enhance deep-seated vessel details captured from OCTA. It leverages a domain-adversarial loss function to enforce structural consistency and a novel identity invariant loss to mitigate excessive image content generation. We validate the structural fidelity of OCPA-Net on simulation experiments, and then demonstrate its vascular enhancement performance on in vivo imaging experiments of tumor-bearing mice and contrast-enhanced pregnant mice. The results show the promise of our method for comprehensive vessel-related image analysis in preclinical research applications.

Multi-degradation-adaptation network for fundus image enhancement with degradation representation learning

Fundus image quality serves a crucial asset for medical diagnosis and applications. However, such images often suffer degradation during image acquisition where multiple types of degradation can occur in each image. Although recent deep learning based methods have shown promising results in image enhancement, they tend to focus on restoring one aspect of degradation and lack generalisability to multiple modes of degradation. We propose an adaptive image enhancement network that can simultaneously handle a mixture of different degradations. The main contribution of this work is to introduce our Multi-Degradation-Adaptive module which dynamically generates filters for different types of degradation. Moreover, we explore degradation representation learning and propose the degradation representation network and Multi-Degradation-Adaptive discriminator for our accompanying image enhancement network. Experimental results demonstrate that our method outperforms several existing state-of-the-art methods in fundus image enhancement. Code will be available at https://github.com/RuoyuGuo/MDA-Net.

Digital ray: enhancing cataractous fundus images using style transfer generative adversarial networks to improve retinopathy detection

BACKGROUND/AIMS

The aim of this study was to develop and evaluate digital ray, based on preoperative and postoperative image pairs using style transfer generative adversarial networks (GANs), to enhance cataractous fundus images for improved retinopathy detection.

METHODS

For eligible cataract patients, preoperative and postoperative colour fundus photographs (CFP) and ultra-wide field (UWF) images were captured. Then, both the original CycleGAN and a modified CycleGAN (C2ycleGAN) framework were adopted for image generation and quantitatively compared using Frechet Inception Distance (FID) and Kernel Inception Distance (KID). Additionally, CFP and UWF images from another cataract cohort were used to test model performances. Different panels of ophthalmologists evaluated the quality, authenticity and diagnostic efficacy of the generated images.

RESULTS

A total of 959 CFP and 1009 UWF image pairs were included in model development. FID and KID indicated that images generated by C2ycleGAN presented significantly improved quality. Based on ophthalmologists' average ratings, the percentages of inadequate-quality images decreased from 32% to 18.8% for CFP, and from 18.7% to 14.7% for UWF. Only 24.8% and 13.8% of generated CFP and UWF images could be recognised as synthetic. The accuracy of retinopathy detection significantly increased from 78% to 91% for CFP and from 91% to 93% for UWF. For retinopathy subtype diagnosis, the accuracies also increased from 87%-94% to 91%-100% for CFP and from 87%-95% to 93%-97% for UWF.

CONCLUSION

Digital ray could generate realistic postoperative CFP and UWF images with enhanced quality and accuracy for overall detection and subtype diagnosis of retinopathies, especially for CFP.\ TRIAL REGISTRATION NUMBER: This study was registered with ClinicalTrials.gov (NCT05491798).

Synthetic lumbar MRI can aid in diagnosis and treatment strategies based on self-pix networks

CT and MR tools are commonly used to diagnose lumbar fractures (LF). However, numerous limitations have been found in practice. The aims of this study were to innovate and develop a spinal disease-specific neural network and to evaluate whether synthetic MRI of the LF affected clinical diagnosis and treatment strategies. A total of 675 LF patients who met the inclusion and exclusion criteria were included in the study. For each participant, two mid-sagittal CT and T2-weighted MR images were selected; 1350 pairs of LF images were also included. A new Self-pix based on Pix2pix and Self-Attention was constructed. A total of 1350 pairs of CT and MR images, which were randomly divided into a training group (1147 pairs) and a test group (203 pairs), were fed into Pix2pix and Self-pix. The quantitative evaluation included PSNR and SSIM (PSNR1 and SSIM1: real MR images and Pix2pix-generated MR images; PSNR2 and SSIM2: real MR images and Self-pix-generated MR images). The qualitative evaluation, including accurate diagnosis of acute fractures and accurate selection of treatment strategies based on Self-pix-generated MRI, was performed by three spine surgeons. In the LF group, PSNR1 and PSNR2 were 10.884 and 11.021 (p < 0.001), and SSIM1 and SSIM2 were 0.766 and 0.771 (p < 0.001), respectively. In the ROI group, PSNR1 and PSNR2 were 12.350 and 12.670 (p = 0.004), and SSIM1 and SSIM2 were 0.816 and 0.832 (p = 0.005), respectively. According to the qualitative evaluation, Self-pix-generated MRI showed no significant difference from real MRI in identifying acute fractures (p = 0.689), with a good sensitivity of 84.36% and specificity of 96.65%. No difference in treatment strategy was found between the Self-pix-generated MRI group and the real MRI group (p = 0.135). In this study, a disease-specific GAN named Self-pix was developed, which demonstrated better image generation performance compared to traditional GAN. The spine surgeon could accurately diagnose LF and select treatment strategies based on Self-pix-generated T2 MR images.

Enhanced tissue slide imaging in the complex domain via cross-explainable GAN for Fourier ptychographic microscopy

Achieving microscopy with large space-bandwidth products plays a key role in diagnostic imaging and is widely significant in the overall field of clinical practice. Among quantitative microscopy techniques, Fourier Ptychography (FP) provides a wide field of view and high-resolution images, suitable to the histopathological field, but onerous in computational terms. Artificial intelligence can be a solution in this sense. In particular, this research delves into the application of Generative Adversarial Networks (GAN) for the dual-channel complex FP image enhancement of human kidney samples. The study underscores the GANs' efficacy in promoting biological architectures in FP domain, thereby still guaranteeing high resolution and visibility of detailed microscopic structures. We demonstrate successful GAN-based enhanced reconstruction through two strategies: cross-explainability and expert survey. The cross-explainability is evaluated through the comparison of explanation maps for both real and imaginary components underlining its robustness. This comparison further shows that their interplay is pivotal for accurate reconstruction without hallucinations. Secondly, the enhanced reconstruction accuracy and effectiveness in a clinical workflow are confirmed through a two-step survey conducted with nephrologists.

Deep Learning Based Cystoscopy Image Enhancement

Background: Endoscopy image enhancement technology provides doctors with clearer and more detailed images for observation and diagnosis, allowing doctors to assess lesions more accurately. Unlike most other endoscopy images, cystoscopy images face more complex and diverse image degradation because of their underwater imaging characteristics. Among the various causes of image degradation, the blood haze resulting from bladder mucosal bleeding make the background blurry and unclear, severely affecting diagnostic efficiency, even leading to misjudgment. Materials and Methods: We propose a deep learning-based approach to mitigate the impact of blood haze on cystoscopy images. The approach consists of two parts as follows: a blood haze removal network and a contrast enhancement algorithm. First, we adopt Feature Fusion Attention Network (FFA-Net) and transfer learning in the field of deep learning to remove blood haze from cystoscopy images and introduce perceptual loss to constrain the network for better visual results. Second, we enhance the image contrast by remapping the gray scale of the blood haze-free image and performing weighted fusion of the processed image and the original image. Results: In the blood haze removal stage, the algorithm proposed in this article achieves an average peak signal-to-noise ratio of 29.44 decibels, which is 15% higher than state-of-the-art traditional methods. The average structural similarity and perceptual image patch similarity reach 0.9269 and 0.1146, respectively, both superior to state-of-the-art traditional methods. Besides, our method is the best in keeping color balance after removing the blood haze. In the image enhancement stage, our algorithm enhances the contrast of vessels and tissues while preserving the original colors, expanding the dynamic range of the image. Conclusion: The deep learning-based cystoscopy image enhancement method is significantly better than other traditional methods in both qualitative and quantitative evaluation. The application of artificial intelligence will provide clearer, higher contrast cystoscopy images for medical diagnosis.

AI-Assisted Detection of Biomarkers by Sensors and Biosensors for Early Diagnosis and Monitoring

The steady progress in consumer electronics, together with improvement in microflow techniques, nanotechnology, and data processing, has led to implementation of cost-effective, user-friendly portable devices, which play the role of not only gadgets but also diagnostic tools. Moreover, numerous smart devices monitor patients' health, and some of them are applied in point-of-care (PoC) tests as a reliable source of evaluation of a patient's condition. Current diagnostic practices are still based on laboratory tests, preceded by the collection of biological samples, which are then tested in clinical conditions by trained personnel with specialistic equipment. In practice, collecting passive/active physiological and behavioral data from patients in real time and feeding them to artificial intelligence (AI) models can significantly improve the decision process regarding diagnosis and treatment procedures via the omission of conventional sampling and diagnostic procedures while also excluding the role of pathologists. A combination of conventional and novel methods of digital and traditional biomarker detection with portable, autonomous, and miniaturized devices can revolutionize medical diagnostics in the coming years. This article focuses on a comparison of traditional clinical practices with modern diagnostic techniques based on AI and machine learning (ML). The presented technologies will bypass laboratories and start being commercialized, which should lead to improvement or substitution of current diagnostic tools. Their application in PoC settings or as a consumer technology accessible to every patient appears to be a real possibility. Research in this field is expected to intensify in the coming years. Technological advancements in sensors and biosensors are anticipated to enable the continuous real-time analysis of various omics fields, fostering early disease detection and intervention strategies. The integration of AI with digital health platforms would enable predictive analysis and personalized healthcare, emphasizing the importance of interdisciplinary collaboration in related scientific fields.

Computational single fundus image restoration techniques: a review

Fundus cameras are widely used by ophthalmologists for monitoring and diagnosing retinal pathologies. Unfortunately, no optical system is perfect, and the visibility of retinal images can be greatly degraded due to the presence of problematic illumination, intraocular scattering, or blurriness caused by sudden movements. To improve image quality, different retinal image restoration/enhancement techniques have been developed, which play an important role in improving the performance of various clinical and computer-assisted applications. This paper gives a comprehensive review of these restoration/enhancement techniques, discusses their underlying mathematical models, and shows how they may be effectively applied in real-life practice to increase the visual quality of retinal images for potential clinical applications including diagnosis and retinal structure recognition. All three main topics of retinal image restoration/enhancement techniques, i.e., illumination correction, dehazing, and deblurring, are addressed. Finally, some considerations about challenges and the future scope of retinal image restoration/enhancement techniques will be discussed.

Enhancing and Adapting in the Clinic: Source-Free Unsupervised Domain Adaptation for Medical Image Enhancement

Medical imaging provides many valuable clues involving anatomical structure and pathological characteristics. However, image degradation is a common issue in clinical practice, which can adversely impact the observation and diagnosis by physicians and algorithms. Although extensive enhancement models have been developed, these models require a well pre-training before deployment, while failing to take advantage of the potential value of inference data after deployment. In this paper, we raise an algorithm for source-free unsupervised domain adaptive medical image enhancement (SAME), which adapts and optimizes enhancement models using test data in the inference phase. A structure-preserving enhancement network is first constructed to learn a robust source model from synthesized training data. Then a teacher-student model is initialized with the source model and conducts source-free unsupervised domain adaptation (SFUDA) by knowledge distillation with the test data. Additionally, a pseudo-label picker is developed to boost the knowledge distillation of enhancement tasks. Experiments were implemented on ten datasets from three medical image modalities to validate the advantage of the proposed algorithm, and setting analysis and ablation studies were also carried out to interpret the effectiveness of SAME. The remarkable enhancement performance and benefits for downstream tasks demonstrate the potential and generalizability of SAME. The code is available at https://github.com/liamheng/Annotation-free-Medical-Image-Enhancement.

Image Recovery Matters: A Recovery-Extraction Framework for Robust Fetal Brain Extraction From MR Images

The extraction of the fetal brain from magnetic resonance (MR) images is a challenging task. In particular, fetal MR images suffer from different kinds of artifacts introduced during the image acquisition. Among those artifacts, intensity inhomogeneity is a common one affecting brain extraction. In this work, we propose a deep learning-based recovery-extraction framework for fetal brain extraction, which is particularly effective in handling fetal MR images with intensity inhomogeneity. Our framework involves two stages. First, the artifact-corrupted images are recovered with the proposed generative adversarial learning-based image recovery network with a novel region-of-darkness discriminator that enforces the network focusing on artifacts of the images. Second, we propose a brain extraction network for more effective fetal brain segmentation by strengthening the association between lower- and higher-level features as well as suppressing task-irrelevant features. Thanks to the proposed recovery-extraction strategy, our framework is able to accurately segment fetal brains from artifact-corrupted MR images. The experiments show that our framework achieves promising performance in both quantitative and qualitative evaluations, and outperforms state-of-the-art methods in both image recovery and fetal brain extraction.

LRENet: a location-related enhancement network for liver lesions in CT images

Objective. Liver cancer is a major global health problem expected to increase by more than 55% by 2040. Accurate segmentation of liver tumors from computed tomography (CT) images is essential for diagnosis and treatment planning. However, this task is challenging due to the variations in liver size, the low contrast between tumor and normal tissue, and the noise in the images.

APPROACH

In this study, we propose a novel method called location-related enhancement network (LRENet) which can enhance the contrast of liver lesions in CT images and facilitate their segmentation. LRENet consists of two steps: (1) locating the lesions and the surrounding tissues using a morphological approach and (2) enhancing the lesions and smoothing the other regions using a new loss function.

MAIN RESULTS

We evaluated LRENet on two public datasets (LiTS and 3Dircadb01) and one dataset collected from a collaborative hospital (Liver cancer dateset), and compared it with state-of-the-art methods regarding several metrics. The results of the experiments showed that our proposed method outperformed the compared methods on three datasets in several metrics. We also trained the Swin-Transformer network on the enhanced datasets and showed that our method could improve the segmentation performance of both liver and lesions.

SIGNIFICANCE

Our method has potential applications in clinical diagnosis and treatment planning, as it can provide more reliable and informative CT images of liver tumors.

GAN-based medical image small region forgery detection via a two-stage cascade framework

Using generative adversarial network (GAN) Goodfellow et al. (2014) for data enhancement of medical images is significantly helpful for many computer-aided diagnosis (CAD) tasks. A new GAN-based automated tampering attack, like CT-GAN Mirsky et al. (2019), has emerged. It can inject or remove lung cancer lesions to CT scans. Because the tampering region may even account for less than 1% of the original image, even state-of-the-art methods are challenging to detect the traces of such tampering. This paper proposes a two-stage cascade framework to detect GAN-based medical image small region forgery like CT-GAN. In the local detection stage, we train the detector network with small sub-images so that interference information in authentic regions will not affect the detector. We use depthwise separable convolution and residual networks to prevent the detector from over-fitting and enhance the ability to find forged regions through the attention mechanism. The detection results of all sub-images in the same image will be combined into a heatmap. In the global classification stage, using gray-level co-occurrence matrix (GLCM) can better extract features of the heatmap. Because the shape and size of the tampered region are uncertain, we use hyperplanes in an infinite-dimensional space for classification. Our method can classify whether a CT image has been tampered and locate the tampered position. Sufficient experiments show that our method can achieve excellent performance than the state-of-the-art detection methods.

A generic fundus image enhancement network boosted by frequency self-supervised representation learning

Fundus photography is prone to suffer from image quality degradation that impacts clinical examination performed by ophthalmologists or intelligent systems. Though enhancement algorithms have been developed to promote fundus observation on degraded images, high data demands and limited applicability hinder their clinical deployment. To circumvent this bottleneck, a generic fundus image enhancement network (GFE-Net) is developed in this study to robustly correct unknown fundus images without supervised or extra data. Levering image frequency information, self-supervised representation learning is conducted to learn robust structure-aware representations from degraded images. Then with a seamless architecture that couples representation learning and image enhancement, GFE-Net can accurately correct fundus images and meanwhile preserve retinal structures. Comprehensive experiments are implemented to demonstrate the effectiveness and advantages of GFE-Net. Compared with state-of-the-art algorithms, GFE-Net achieves superior performance in data dependency, enhancement performance, deployment efficiency, and scale generalizability. Follow-up fundus image analysis is also facilitated by GFE-Net, whose modules are respectively verified to be effective for image enhancement.

Blind Image Quality Assessment for Pathological Microscopic Image Under Screen and Immersion Scenarios

The high-quality pathological microscopic images are essential for physicians or pathologists to make a correct diagnosis. Image quality assessment (IQA) can quantify the visual distortion degree of images and guide the imaging system to improve image quality, thus raising the quality of pathological microscopic images. Current IQA methods are not ideal for pathological microscopy images due to their specificity. In this paper, we present deep learning-based blind image quality assessment model with saliency block and patch block for pathological microscopic images. The saliency block and patch block can handle the local and global distortions, respectively. To better capture the area of interest of pathologists when viewing pathological images, the saliency block is fine-tuned by eye movement data of pathologists. The patch block can capture lots of global information strongly related to image quality via the interaction between different image patches from different positions. The performance of the developed model is validated by the home-made Pathological Microscopic Image Quality Database under Screen and Immersion Scenarios (PMIQD-SIS) and cross-validated by the five public datasets. The results of ablation experiments demonstrate the contribution of the added blocks. The dataset and the corresponding code are publicly available at: https://github.com/mikugyf/PMIQD-SIS.

Design and Validation of a Deep Learning Model for Renal Stone Detection and Segmentation on Kidney-Ureter-Bladder Images

Kidney-ureter-bladder (KUB) imaging is used as a frontline investigation for patients with suspected renal stones. In this study, we designed a computer-aided diagnostic system for KUB imaging to assist clinicians in accurately diagnosing urinary tract stones. The image dataset used for training and testing the model comprised 485 images provided by Kaohsiung Chang Gung Memorial Hospital. The proposed system was divided into two subsystems, 1 and 2. Subsystem 1 used Inception-ResNetV2 to train a deep learning model on preprocessed KUB images to verify the improvement in diagnostic accuracy with image preprocessing. Subsystem 2 trained an image segmentation model using the ResNet hybrid, U-net, to accurately identify the contours of renal stones. The performance was evaluated using a confusion matrix for the classification model. We conclude that the model can assist clinicians in accurately diagnosing renal stones via KUB imaging. Therefore, the proposed system can assist doctors in diagnosis, reduce patients' waiting time for CT scans, and minimize the radiation dose absorbed by the body.

TransOrGAN: An Artificial Intelligence Mapping of Rat Transcriptomic Profiles between Organs, Ages, and Sexes

Animal studies are required for the evaluation of candidate drugs to ensure patient and volunteer safety. Toxicogenomics is often applied in these studies to gain understanding of the underlying mechanisms of toxicity, which is usually focused on critical organs such as the liver or kidney in young male rats. There is a strong ethical reason to reduce, refine and replace animal use (the 3Rs), where the mapping of data between organs, sexes and ages could reduce the cost and time of drug development. Herein, we proposed a generative adversarial network (GAN)-based framework entitled TransOrGAN that allowed the molecular mapping of gene expression profiles in different rodent organ systems and across sex and age groups. We carried out a proof-of-concept study based on rat RNA-seq data from 288 samples in 9 different organs of both sexes and 4 developmental stages. First, we demonstrated that TransOrGAN could infer transcriptomic profiles between any 2 of the 9 organs studied, yielding an average cosine similarity of 0.984 between synthetic transcriptomic profiles and their corresponding real profiles. Second, we found that TransOrGAN could infer transcriptomic profiles observed in females from males, with an average cosine similarity of 0.984. Third, we found that TransOrGAN could infer transcriptomic profiles in juvenile, adult, and aged animals from adolescent animals with an average cosine similarity of 0.981, 0.983, and 0.989, respectively. Altogether, TransOrGAN is an innovative approach to infer transcriptomic profiles between ages, sexes, and organ systems, offering the opportunity to reduce animal usage and to provide an integrated assessment of toxicity in the whole organism irrespective of sex or age.

Development and validation of a deep learning-based laparoscopic system for improving video quality

PURPOSE

A clear surgical field of view is a prerequisite for successful laparoscopic surgery. Surgical smoke, image blur, and lens fogging can affect the clarity of laparoscopic imaging. We aimed to develop a real-time assistance system (namely LVQIS) for removing these interfering factors during laparoscopic surgery, thereby improving laparoscopic video quality.

METHODS

LVQIS was developed with generative adversarial networks (GAN) and transfer learning, which included two classification models (ResNet-50), a motion blur removal model (MPRNet), and a smoke/fog removal model (GAN). 136 laparoscopic surgery videos were retrospectively collected in a tripartite dataset for training and validation. A synthetic dataset was simulated using the image enhancement library Albumentations and the 3D rendering software Blender. The objective evaluation results were through PSNR, SSIM and FID, and the subjective evaluation includes the operation pause time and the degree of anxiety of surgeons.

RESULTS

The synthesized dataset contained 19,245 clear images, 19,245 motion blur images, and 19,245 smoke/fog images. The ResNet-50 CNN model identified whether a single laparoscopic image had motion blur and smoke/fog with an accuracy of over 0.99. The PSNR, SSIM and FID of the de-smoke model were 29.67, 0.9551 and 74.72, respectively, and the PSNR, SSIM and FID of the de-blurring model were 26.78, 0.9020 and 80.10, respectively, which were better than other advanced de-blurring and de-smoke/fog models. In a comparative study of 100 laparoscopic surgeries, the use of LVQIS significantly reduced the operation pause time (P < 0.001) and the anxiety of surgeons (P = 0.004).

CONCLUSIONS

In this study, LVQIS is an efficient and robust system that can improve the quality of laparoscopic video, reduce surgical pause time and the anxiety of surgeons, and has the potential for real-time application in real clinical settings.

3D-MedTranCSGAN: 3D Medical Image Transformation using CSGAN

Computer vision techniques are a rapidly growing area of transforming medical images for various specific medical applications. In an end-to-end application, this paper proposes a 3D Medical Image Transformation Using a CSGAN model named a 3D-MedTranCSGAN. The 3D-MedTranCSGAN model is an integration of non-adversarial loss components and the Cyclic Synthesized Generative Adversarial Networks. The proposed model utilizes PatchGAN's discriminator network, to penalize the difference between the synthesized image and the original image. The model also computes the non-adversary loss functions such as content, perception, and style transfer losses. 3DCascadeNet is a new generator architecture introduced in the paper, which is used to enhance the perceptiveness of the transformed medical image by encoding-decoding pairs. We use the 3D-MedTranCSGAN model to do various tasks without modifying specific applications: PET to CT image transformation; reconstruction of CT to PET; modification of movement artefacts in MR images; and removing noise in PET images. We found that 3D-MedTranCSGAN outperformed other transformation methods in our experiments. For the first task, the proposed model yields SSIM is 0.914, PSNR is 26.12, MSE is 255.5, VIF is 0.4862, UQI is 0.9067 and LPIPs is 0.2284. For the second task, the model yields 0.9197, 25.7, 257.56, 0.4962, 0.9027, 0.2262. For the third task, the model yields 0.8862, 24.94, 0.4071, 0.6410, 0.2196. For the final task, the model yields 0.9521, 33.67, 33.57, 0.6091, 0.9255, 0.0244. Based on the result analysis, the proposed model outperforms the other techniques.

Luminosity rectified blind Richardson-Lucy deconvolution for single retinal image restoration

BACKGROUND AND OBJECTIVE

Due to imperfect imaging conditions, retinal images can be degraded by uneven/insufficient illumination, blurriness caused by optical aberrations and unintentional motions. Degraded images reduce the effectiveness of diagnosis by an ophthalmologist. To restore the image quality, in this research we propose the luminosity rectified Richardson-Lucy (LRRL) blind deconvolution framework for single retinal image restoration.

METHODS

We established an image formation model based on the double-pass fundus reflection feature and developed a differentiable non-convex cost function that jointly achieves illumination correction and blind deconvolution. To solve this non-convex optimization problem, we derived the closed-form expression of the gradients and used gradient descent with Nesterov-accelerated adaptive momentum estimation to accelerate the optimization, which is more efficient than the traditional half quadratic splitting method.

RESULTS

The LRRL was tested on 1719 images from three public databases. Four image quality matrixes including image definition, image sharpness, image entropy, and image multiscale contrast were used for objective assessments. The LRRL was compared against the state-of-the-art retinal image blind deconvolution methods.

CONCLUSIONS

Our LRRL corrects the problematic illumination and improves the clarity of the retinal image simultaneously, showing its superiority in terms of restoration quality and implementation efficiency. The MATLAB code is available on Github.

A two-stage framework for optical coherence tomography angiography image quality improvement

Introduction

Optical Coherence Tomography Angiography (OCTA) is a new non-invasive imaging modality that gains increasing popularity for the observation of the microvasculatures in the retina and the conjunctiva, assisting clinical diagnosis and treatment planning. However, poor imaging quality, such as stripe artifacts and low contrast, is common in the acquired OCTA and in particular Anterior Segment OCTA (AS-OCTA) due to eye microtremor and poor illumination conditions. These issues lead to incomplete vasculature maps that in turn makes it hard to make accurate interpretation and subsequent diagnosis.

Methods

In this work, we propose a two-stage framework that comprises a de-striping stage and a re-enhancing stage, with aims to remove stripe noise and to enhance blood vessel structure from the background. We introduce a new de-striping objective function in a Stripe Removal Net (SR-Net) to suppress the stripe noise in the original image. The vasculatures in acquired AS-OCTA images usually exhibit poor contrast, so we use a Perceptual Structure Generative Adversarial Network (PS-GAN) to enhance the de-striped AS-OCTA image in the re-enhancing stage, which combined cyclic perceptual loss with structure loss to achieve further image quality improvement.

Results and discussion

To evaluate the effectiveness of the proposed method, we apply the proposed framework to two synthetic OCTA datasets and a real AS-OCTA dataset. Our results show that the proposed framework yields a promising enhancement performance, which enables both conventional and deep learning-based vessel segmentation methods to produce improved results after enhancement of both retina and AS-OCTA modalities.

A Retinex-based variational model for noise suppression and nonuniform illumination correction in corneal confocal microscopy images

Objective.Corneal confocal microscopy (CCM) image analysis is a non-invasivein vivoclinical technique that can quantify corneal nerve fiber damage. However, the acquired CCM images are often accompanied by speckle noise and nonuniform illumination, which seriously affects the analysis and diagnosis of the diseases.Approach.In this paper, first we propose a variational Retinex model for the inhomogeneity correction and noise removal of CCM images. In this model, the Beppo Levi space is introduced to constrain the smoothness of the illumination layer for the first time, and the fractional order differential is adopted as the regularization term to constrain reflectance layer. Then, a denoising regularization term is also constructed with Block Matching 3D (BM3D) to suppress noise. Finally, by adjusting the uneven illumination layer, we obtain the final results. Second, an image quality evaluation metric is proposed to evaluate the illumination uniformity of images objectively.Main results.To demonstrate the effectiveness of our method, the proposed method is tested on 628 low-quality CCM images from the CORN-2 dataset. Extensive experiments show the proposed method outperforms the other four related methods in terms of noise removal and uneven illumination suppression.SignificanceThis demonstrates that the proposed method may be helpful for the diagnostics and analysis of eye diseases.

Learning for retinal image quality assessment with label regularization

BACKGROUND AND OBJECTIVE

The assessment of the image quality is crucial before the computer-aided diagnosis of fundus images. This task is very challenging. Firstly, the subjective judgments of graders on image quality lead to ambiguous labels. Secondly, despite being treated as classification in existing works, grading has regression properties that cannot be ignored. Solving the ambiguity problem and regression problem in the label space, and extracting discriminative features, have become the keys to quality assessment.

METHODS

In this paper, we proposed a framework that can assess the quality of fundus images accurately and reasonably based on deep convolutional neural networks. Drawing on the experience of human graders, a dual-path convolutional neural network with attention blocks is designed to better extract discriminative features and present the bases of decision. Label smoothing and cost-sensitive regularization are designed to solve the label ambiguity problem and the potential regression problem respectively. Besides, a large number of images are annotated by us to further improve the results.

RESULTS

We conducted our experiments on the largest retinal image quality assessment dataset with 28,792 retinal images. Our approach achieves 0.8868 precision, 0.8786 recall, 0.8820 F1, and 0.9138 Kappa score. Results show that our approach outperforms state-of-the-art methods.

CONCLUSIONS

The promising performances reveal that our methods are beneficial to retinal image quality assessment and have potential in other grading tasks.

Deep Learning Model for Computer-Aided Diagnosis of Urolithiasis Detection from Kidney-Ureter-Bladder Images

Kidney-ureter-bladder (KUB) imaging is a radiological examination with a low cost, low radiation, and convenience. Although emergency room clinicians can arrange KUB images easily as a first-line examination for patients with suspicious urolithiasis, interpreting the KUB images correctly is difficult for inexperienced clinicians. Obtaining a formal radiology report immediately after a KUB imaging examination can also be challenging. Recently, artificial-intelligence-based computer-aided diagnosis (CAD) systems have been developed to help clinicians who are not experts make correct diagnoses for further treatment more effectively. Therefore, in this study, we proposed a CAD system for KUB imaging based on a deep learning model designed to help first-line emergency room clinicians diagnose urolithiasis accurately. A total of 355 KUB images were retrospectively collected from 104 patients who were diagnosed with urolithiasis at Kaohsiung Chang Gung Memorial Hospital. Then, we trained a deep learning model with a ResNet architecture to classify KUB images in terms of the presence or absence of kidney stones with this dataset of pre-processed images. Finally, we tuned the parameters and tested the model experimentally. The results show that the accuracy, sensitivity, specificity, and F1-measure of the model were 0.977, 0.953, 1, and 0.976 on the validation set and 0.982, 0.964, 1, and 0.982 on the testing set, respectively. Moreover, the results demonstrate that the proposed model performed well compared to the existing CNN-based methods and was able to detect urolithiasis in KUB images successfully. We expect the proposed approach to help emergency room clinicians make accurate diagnoses and reduce unnecessary radiation exposure from computed tomography (CT) scans, along with the associated medical costs.

An efficient modular framework for automatic LIONC classification of MedIMG using unified medical language

Handwritten prescriptions and radiological reports: doctors use handwritten prescriptions and radiological reports to give drugs to patients who have illnesses, injuries, or other problems. Clinical text data, like physician prescription visuals and radiology reports, should be labelled with specific information such as disease type, features, and anatomical location for more effective use. The semantic annotation of vast collections of biological and biomedical texts, like scientific papers, medical reports, and general practitioner observations, has lately been examined by doctors and scientists. By identifying and disambiguating references to biomedical concepts in texts, medical semantics annotators could generate such annotations automatically. For Medical Images (MedIMG), we provide a methodology for learning an effective holistic representation (handwritten word pictures as well as radiology reports). Deep Learning (DL) methods have recently gained much interest for their capacity to achieve expert-level accuracy in automated MedIMG analysis. We discovered that tasks requiring significant responsive fields are ideal for downscaled input images that are qualitatively verified by examining functional, responsive areas and class activating maps for training models. This article focuses on the following contributions: (a) Information Extraction from Narrative MedImages, (b) Automatic categorisation on image resolution with an impact on MedIMG, and (c) Hybrid Model to Predictions of Named Entity Recognition utilising RNN + LSTM + GRM that perform admirably in every trainee for every input purpose. At the same time, supplying understandable scale weight implies that such multi-scale structures are also crucial for extracting information from high-resolution MedIMG. A portion of the reports (30%) are manually evaluated by trained physicians, while the rest were automatically categorised using deep supervised training models based on attention mechanisms and supplied with test reports. MetaMapLite proved recall and precision, but also an F1-score equivalent for primary biomedicine text search techniques and medical text examination on many databases of MedIMG. In addition to implementing as well as getting the requirements for MedIMG, the article explores the quality of medical data by using DL techniques for reaching large-scale labelled clinical data and also the significance of their real-time efforts in the biomedical study that have played an instrumental role in its extramural diffusion and global appeal.

GAN Training Acceleration Using Fréchet Descriptor-Based Coreset

Generative Adversarial Networks (GANs) are a class of deep learning models being applied to image processing. GANs have demonstrated state-of-the-art performance in applications such as image generation and image-to-image translation, just to name a few. However, with this success comes the realization that the training of GANs takes a long time and is often limited by available computing resources. In this research, we propose to construct a Coreset using Fréchet Descriptor Distances (FDD-Coreset) to accelerate the training of GAN for blob identification. We first propose a Fréchet Descriptor Distance (FDD) to measure the difference between each pair of blob images based on the statistics derived from blob distribution. The Coreset is then employed using our proposed FDD metric to select samples from the entire dataset for GAN training. A 3D-simulated dataset of blobs and a 3D MRI dataset of human kidneys are studied. Using computation time and eight performance metrics, the GAN trained on the FDD-Coreset is compared against the model trained on the entire dataset and an Inception and Euclidean Distance-based Coreset (IED-Coreset). We conclude that the FDD-Coreset not only significantly reduces the training time, but also achieves higher denoising performance and maintains approximate performance of blob identification compared with training on the entire dataset.

RFormer: Transformer-Based Generative Adversarial Network for Real Fundus Image Restoration on a New Clinical Benchmark

Ophthalmologists have used fundus images to screen and diagnose eye diseases. However, different equipments and ophthalmologists pose large variations to the quality of fundus images. Low-quality (LQ) degraded fundus images easily lead to uncertainty in clinical screening and generally increase the risk of misdiagnosis. Thus, real fundus image restoration is worth studying. Unfortunately, real clinical benchmark has not been explored for this task so far. In this paper, we investigate the real clinical fundus image restoration problem. Firstly, We establish a clinical dataset, Real Fundus (RF), including 120 low- and high-quality (HQ) image pairs. Then we propose a novel Transformer-based Generative Adversarial Network (RFormer) to restore the real degradation of clinical fundus images. The key component in our network is the Window-based Self-Attention Block (WSAB) which captures non-local self-similarity and long-range dependencies. To produce more visually pleasant results, a Transformer-based discriminator is introduced. Extensive experiments on our clinical benchmark show that the proposed RFormer significantly outperforms the state-of-the-art (SOTA) methods. In addition, experiments of downstream tasks such as vessel segmentation and optic disc/cup detection demonstrate that our proposed RFormer benefits clinical fundus image analysis and applications. The dataset, code, and models will be made publicly available at https://github.com/dengzhuo-AI/Real-Fundus.

A Novel Un-Supervised GAN for Fundus Image Enhancement with Classification Prior Loss

Fundus images captured for clinical diagnosis usually suffer from degradation factors due to variation in equipment, operators, or environment. These degraded fundus images need to be enhanced to achieve better diagnosis and improve the results of downstream tasks. As there is no paired low- and high-quality fundus image, existing methods mainly focus on supervised or semi-supervised learning methods for color fundus image enhancement (CFIE) tasks by utilizing synthetic image pairs. Consequently, domain gaps between real images and synthetic images arise. With respect to existing unsupervised methods, the most important low scale pathological features and structural information in degraded fundus images are prone to be erased after enhancement. To solve these problems, an unsupervised GAN is proposed for CFIE tasks utilizing adversarial training to enhance low quality fundus images. Synthetic image pairs are no longer required during the training. A specially designed U-Net with skip connection in our enhancement network can effectively remove degradation factors while preserving pathological features and structural information. Global and local discriminators adopted in the GAN lead to better illumination uniformity in the enhanced fundus image. To better improve the visual quality of enhanced fundus images, a novel non-reference loss function based on a pretrained fundus image quality classification network was designed to guide the enhancement network to produce high quality images. Experiments demonstrated that our method could effectively remove degradation factors in low-quality fundus images and produce a competitive result compared with previous methods in both quantitative and qualitative metrics.

Edge-enhanced dual discriminator generative adversarial network for fast MRI with parallel imaging using multi-view information

In clinical medicine, magnetic resonance imaging (MRI) is one of the most important tools for diagnosis, triage, prognosis, and treatment planning. However, MRI suffers from an inherent slow data acquisition process because data is collected sequentially in k-space. In recent years, most MRI reconstruction methods proposed in the literature focus on holistic image reconstruction rather than enhancing the edge information. This work steps aside this general trend by elaborating on the enhancement of edge information. Specifically, we introduce a novel parallel imaging coupled dual discriminator generative adversarial network (PIDD-GAN) for fast multi-channel MRI reconstruction by incorporating multi-view information. The dual discriminator design aims to improve the edge information in MRI reconstruction. One discriminator is used for holistic image reconstruction, whereas the other one is responsible for enhancing edge information. An improved U-Net with local and global residual learning is proposed for the generator. Frequency channel attention blocks (FCA Blocks) are embedded in the generator for incorporating attention mechanisms. Content loss is introduced to train the generator for better reconstruction quality. We performed comprehensive experiments on Calgary-Campinas public brain MR dataset and compared our method with state-of-the-art MRI reconstruction methods. Ablation studies of residual learning were conducted on the MICCAI13 dataset to validate the proposed modules. Results show that our PIDD-GAN provides high-quality reconstructed MR images, with well-preserved edge information. The time of single-image reconstruction is below 5ms, which meets the demand of faster processing.

Retinal Image Enhancement Using Cycle-Constraint Adversarial Network

Retinal images are the most intuitive medical images for the diagnosis of fundus diseases. Low-quality retinal images cause difficulties in computer-aided diagnosis systems and the clinical diagnosis of ophthalmologists. The high quality of retinal images is an important basis of precision medicine in ophthalmology. In this study, we propose a retinal image enhancement method based on deep learning to enhance multiple low-quality retinal images. A generative adversarial network is employed to build a symmetrical network, and a convolutional block attention module is introduced to improve the feature extraction capability. The retinal images in our dataset are sorted into two sets according to their quality: low and high quality. Generators and discriminators alternately learn the features of low/high-quality retinal images without the need for paired images. We analyze the proposed method both qualitatively and quantitatively on public datasets and a private dataset. The study results demonstrate that the proposed method is superior to other advanced algorithms, especially in enhancing color-distorted retinal images. It also performs well in the task of retinal vessel segmentation. The proposed network effectively enhances low-quality retinal images, aiding ophthalmologists and enabling computer-aided diagnosis in pathological analysis. Our method enhances multiple types of low-quality retinal images using a deep learning network.