An adaptive weighted ensemble learning network for diabetic retinopathy classification.

Diabetic retinopathy (DR) is one of the leading causes of blindness. However, because the data distribution of classes is not always balanced, it is challenging for automated early DR detection using deep learning techniques. In this paper, we propose an adaptive weighted ensemble learning method for DR detection based on optical coherence tomography (OCT) images. Specifically, we develop an ensemble learning model based on three advanced deep learning models for higher performance. To better utilize the cues implied in these base models, a novel decision fusion scheme is proposed based on the Bayesian theory in terms of the key evaluation indicators, to dynamically adjust the weighting distribution of base models to alleviate the negative effects potentially caused by the problem of unbalanced data size. Extensive experiments are performed on two public datasets to verify the effectiveness of the proposed method. A quadratic weighted kappa of 0.8487 and an accuracy of 0.9343 on the DRAC2022 dataset, and a quadratic weighted kappa of 0.9007 and an accuracy of 0.8956 on the APTOS2019 dataset are obtained, respectively. The results demonstrate that our method has the ability to enhance the ovearall performance of DR detection on OCT images.

Adaptive and Iterative Learning with Multi-perspective Regularizations for Metal Artifact Reduction.

Metal artifact reduction (MAR) is important for clinical diagnosis with CT images. The existing state-of-the-art deep learning methods usually suppress metal artifacts in sinogram or image domains or both. However, their performance is limited by the inherent characteristics of the two domains, i.e., the errors introduced by local manipulations in the sinogram domain would propagate throughout the whole image during backprojection and lead to serious secondary artifacts, while it is difficult to distinguish artifacts from actual image features in the image domain. To alleviate these limitations, this study analyzes the desirable properties of wavelet transform in-depth and proposes to perform MAR in the wavelet domain. First, wavelet transform yields components that possess spatial correspondence with the image, thereby preventing the spread of local errors to avoid secondary artifacts. Second, using wavelet transform could facilitate identification of artifacts from image since metal artifacts are mainly high-frequency signals. Taking these advantages of the wavelet transform, this paper decomposes an image into multiple wavelet components and introduces multi-perspective regularizations into the proposed MAR model. To improve the transparency and validity of the model, all the modules in the proposed MAR model are designed to reflect their mathematical meanings. In addition, an adaptive wavelet module is also utilized to enhance the flexibility of the model. To optimize the model, an iterative algorithm is developed. The evaluation on both synthetic and real clinical datasets consistently confirms the superior performance of the proposed method over the competing methods.

Dual-TranSpeckle: Dual-pathway transformer based encoder-decoder network for medical ultrasound image despeckling.

The majority of existing deep learning-based image denoising algorithms mainly focus on processing the overall image features, ignoring the fine differences between the semantic and pixel features. Hence, we propose Dual-TranSpeckle (DTS), a medical ultrasound image despeckling network built on a dual-path Transformer. The DTS introduces two different paths, named "semantic path" and "pixel path," to facilitate the parallel transfer of feature information within the image. The semantic path passes a global view of the input semantic features, and the image features are passed through a Semantic Block to extract global semantic information from pixel-level features. The pixel path is employed to transmit finer-grained pixel features. Within the dual-path network framework, two essential modules, namely Dual Block and Merge Block, are designed. These leverage the Transformer architecture during the encoding and decoding stages. The Dual Block module facilitates information interaction between the semantic and pixel features by considering the interdependencies across both paths. Meanwhile, the Merge Block module enables parallel transfer of feature information by merging the dual path features, thereby facilitating the self-attention calculations for the overall feature representation. Our DTS is extensively evaluated on two public datasets and one private dataset. The DTS network demonstrates significant enhancements in quantitative evaluation results in terms of peak signal-to-noise ratio (PSNR), structural similarity (SSIM), feature similarity (FSIM), and naturalness image quality evaluator (NIQE). Furthermore, our qualitative analysis confirms that the DTS has significant improvements in despeckling performance, effectively suppressing speckle noise while preserving essential image structures.

Iterative Residual Optimization Network for Limited-Angle Tomographic Reconstruction.

Limited-angle tomographic reconstruction is one of the typical ill-posed inverse problems, leading to edge divergence with degraded image quality. Recently, deep learning has been introduced into image reconstruction and achieved great results. However, existing deep reconstruction methods have not fully explored data consistency, resulting in poor performance. In addition, deep reconstruction methods are still mathematically inexplicable and unstable. In this work, we propose an iterative residual optimization network (IRON) for limited-angle tomographic reconstruction. First, a new optimization objective function is established to overcome false negative and positive artifacts induced by limited-angle measurements. We integrate neural network priors as a regularizer to explore deep features within residual data. Furthermore, the block-coordinate descent is employed to achieve a novel iterative framework. Second, a convolution assisted transformer is carefully elaborated to capture both local and long-range pixel interactions simultaneously. Regarding the visual transformer, the multi-head attention is further redesigned to reduce computational costs and protect reconstructed image features. Third, based on the relative error convergence property of the convolution assisted transformer, a mathematical convergence analysis is also provided for our IRON. Both numerically simulated and clinically collected real cardiac datasets are employed to validate the effectiveness and advantages of the proposed IRON. The results show that IRON outperforms other state-of-the-art methods.

Hybrid U-Net and Swin-transformer network for limited-angle cardiac computed tomography.

Objective.Cardiac computed tomography (CT) is widely used for diagnosis of cardiovascular disease, the leading cause of morbidity and mortality in the world. Diagnostic performance depends strongly on the temporal resolution of the CT images. To image the beating heart, one can reduce the scanning time by acquiring limited-angle projections. However, this leads to increased image noise and limited-angle-related artifacts. The goal of this paper is to reconstruct high quality cardiac CT images from limited-angle projections.Approach. The ability to reconstruct high quality images from limited-angle projections is highly desirable and remains a major challenge. With the development of deep learning networks, such as U-Net and transformer networks, progresses have been reached on image reconstruction and processing. Here we propose a hybrid model based on the U-Net and Swin-transformer (U-Swin) networks. The U-Net has the potential to restore structural information due to missing projection data and related artifacts, then the Swin-transformer can gather a detailed global feature distribution.Main results. Using synthetic XCAT and clinical cardiac COCA datasets, we demonstrate that our proposed method outperforms the state-of-the-art deep learning-based methods.Significance. It has a great potential to freeze the beating heart with a higher temporal resolution.

Spectrum learning for super-resolution tomographic reconstruction.

Objective. Computed Tomography (CT) has been widely used in industrial high-resolution non-destructive testing. However, it is difficult to obtain high-resolution images for large-scale objects due to their physical limitations. The objective is to develop an improved super-resolution technique that preserves small structures and details while efficiently capturing high-frequency information.Approach. The study proposes a new deep learning based method called spectrum learning (SPEAR) network for CT images super-resolution. This approach leverages both global information in the image domain and high-frequency information in the frequency domain. The SPEAR network is designed to reconstruct high-resolution images from low-resolution inputs by considering not only the main body of the images but also the small structures and other details. The symmetric property of the spectrum is exploited to reduce weight parameters in the frequency domain. Additionally, a spectrum loss is introduced to enforce the preservation of both high-frequency components and global information.Main results. The network is trained using pairs of low-resolution and high-resolution CT images, and it is fine-tuned using additional low-dose and normal-dose CT image pairs. The experimental results demonstrate that the proposed SPEAR network outperforms state-of-the-art networks in terms of image reconstruction quality. The approach successfully preserves high-frequency information and small structures, leading to better results compared to existing methods. The network's ability to generate high-resolution images from low-resolution inputs, even in cases of low-dose CT images, showcases its effectiveness in maintaining image quality.Significance. The proposed SPEAR network's ability to simultaneously capture global information and high-frequency details addresses the limitations of existing methods, resulting in more accurate and informative image reconstructions. This advancement can have substantial implications for various industries and medical diagnoses relying on accurate imaging.

Deep Few-view High-resolution Photon-counting Extremity CT at Halved Dose for a Clinical Trial.

The latest X-ray photon-counting computed tomography (PCCT) for extremity allows multi-energy high-resolution (HR) imaging for tissue characterization and material decomposition. However, both radiation dose and imaging speed need improvement for contrast-enhanced and other studies. Despite the success of deep learning methods for 2D few-view reconstruction, applying them to HR volumetric reconstruction of extremity scans for clinical diagnosis has been limited due to GPU memory constraints, training data scarcity, and domain gap issues. In this paper, we propose a deep learning-based approach for PCCT image reconstruction at halved dose and doubled speed in a New Zealand clinical trial. Particularly, we present a patch-based volumetric refinement network to alleviate the GPU memory limitation, train network with synthetic data, and use model-based iterative refinement to bridge the gap between synthetic and real-world data. The simulation and phantom experiments demonstrate consistently improved results under different acquisition conditions on both in- and off-domain structures using a fixed network. The image quality of 8 patients from the clinical trial are evaluated by three radiologists in comparison with the standard image reconstruction with a full-view dataset. It is shown that our proposed approach is essentially identical to or better than the clinical benchmark in terms of diagnostic image quality scores. Our approach has a great potential to improve the safety and efficiency of PCCT without compromising image quality.

Surveillance of pathogenic bacteria on a food matrix using machine-learning-enabled paper chromogenic arrays.

Global food systems can benefit significantly from continuous monitoring of microbial food safety, a task for which tedious operations, destructive sampling, and the inability to monitor multiple pathogens remain challenging. This study reports significant improvements to a paper chromogenic array sensor - machine learning (PCA-ML) methodology sensing concentrations of volatile organic compounds (VOCs) emitted on a species-specific basis by pathogens by streamlining dye selection, sensor fabrication, database construction, and machine learning and validation. This approach enables noncontact, time-dependent, simultaneous monitoring of multiple pathogens (Listeria monocytogenes, Salmonella, and E. coli O157:H7) at levels as low as 1 log CFU/g with over 90% accuracy. The report provides theoretical and practical frameworks demonstrating that chromogenic response, including limits of detection, depends on time integrals of VOC concentrations. The paper also discusses the potential for implementing PCA-ML in the food supply chain for different food matrices and pathogens, with species- and strain-specific identification.