Search

A point cloud is a simple and concise 3D representation, but point cloud generation is a long-term challenging task in 3D vision. However, most existing methods only focus on their effectiveness of generation and auto-encoding separately. Furthermore, both generative adversarial networks (GANs) and auto-encoders (AEs) are the most popular generative models. But there is a lack of related research that investigates the implicit connections between them in the field of point cloud generation. Thus, we propose a new bidirectional network (BI-Net) trained with collaborative learning, introducing more priors through the alternate parameter optimizations of a GAN and AE combination, which is different from the way of combining them at the network structure and loss function level. Specifically, BI-Net acts as a GAN and AE in different data processing directions, where their network structures can be reused. If optimizing only the GAN without the AE, there is no direct constraint of ground truth on the generator's parameter optimization. This unique approach enables better network optimization and leads to superior generation results. Moreover, we propose a nearest neighbor mutual exclusion (NNME) loss to further homogenize the spatial distribution of generated points during the reverse direction. Extensive experiments were conducted, and the results show that the BI-Net produces competitive and high-quality results on reasonable structure and uniform distributions compared to existing state-of-the-art methods. We believe that our network structure (BI-Net) with collaborative learning could provide a new promising method for future point cloud generation tasks. [ABSTRACT FROM AUTHOR]

Purpose: To present fully automated DL-based prostate cancer detection system for prostate MRI. Methods: MRI scans from two institutions, were used for algorithm training, validation, testing. MRI-visible lesions were contoured by an experienced radiologist. All lesions were biopsied using MRI-TRUS-guidance. Lesions masks, histopathological results were used as ground truth labels to train UNet, AH-Net architectures for prostate cancer lesion detection, segmentation. Algorithm was trained to detect any prostate cancer ≥ ISUP1. Detection sensitivity, positive predictive values, mean number of false positive lesions per patient were used as performance metrics. Results: 525 patients were included for training, validation, testing of the algorithm. Dataset was split into training (n = 368, 70%), validation (n = 79, 15%), test (n = 78, 15%) cohorts. Dice coefficients in training, validation sets were 0.403, 0.307, respectively, for AHNet model compared to 0.372, 0.287, respectively, for UNet model. In validation set, detection sensitivity was 70.9%, PPV was 35.5%, mean number of false positive lesions/patient was 1.41 (range 0–6) for UNet model compared to 74.4% detection sensitivity, 47.8% PPV, mean number of false positive lesions/patient was 0.87 (range 0–5) for AHNet model. In test set, detection sensitivity for UNet was 72.8% compared to 63.0% for AHNet, mean number of false positive lesions/patient was 1.90 (range 0–7), 1.40 (range 0–6) in UNet, AHNet models, respectively. Conclusion: We developed a DL-based AI approach which predicts prostate cancer lesions at biparametric MRI with reasonable performance metrics. While false positive lesion calls remain as a challenge of AI-assisted detection algorithms, this system can be utilized as an adjunct tool by radiologists. [ABSTRACT FROM AUTHOR]

It is significant to identify the moldy status of stored maize by fungi infection in the early stage. Hyperspectral imaging (HSI) combined with the sparse auto-encoders (SAE) and convolutional neural network (CNN) algorithms was used to classify the moldy grades of maize kernels. The HSI data were obtained in the range of 400–1000 nm, and four grades from health to heavy mildew were distinguished using the measured fungal spores of maize. The depth spectral features were represented using SAE and the image features were extracted by CNN. K nearest neighbors, support vector machine (SVM), and partial least squares discriminant analysis classifiers were combined with the spectral and image features to establish classification models to identify the different moldy grades of maize kernels. The comparison results indicated that the fusion of SAE and CNN combined with the SVM classifier to construct the SAE-CNN-SVM model had the most satisfactory identification result with high correct recognition rates of 99.47% and 98.94% for the training and testing sets, respectively, and the values of sensitivity and specificity were 0.95–1. The moldy grades were presented intuitively on the maize image based on pixels or kernel-wise. Therefore, the HSI with the SAE-CNN-SVM model had good recognition ability for the early detection of moldy maize kernels, which could potentially provide technical support for the development of online detection of moldy maize kernels during storage. [ABSTRACT FROM AUTHOR]

Active learning is a unique abstraction of machine learning techniques where the model/algorithm could guide users for annotation of a set of data points that would be beneficial to the model, unlike passive machine learning. The primary advantage being that active learning frameworks select data points that can accelerate the learning process of a model and can reduce the amount of data needed to achieve full accuracy as compared to a model trained on a randomly acquired data set. Multiple frameworks for active learning combined with deep learning have been proposed, and the majority of them are dedicated to classification tasks. Herein, we explore active learning for the task of segmentation of medical imaging data sets. We investigate our proposed framework using two datasets: 1.) MRI scans of the hippocampus, 2.) CT scans of pancreas and tumors. This work presents a query-by-committee approach for active learning where a joint optimizer is used for the committee. At the same time, we propose three new strategies for active learning: 1.) increasing frequency of uncertain data to bias the training data set; 2.) Using mutual information among the input images as a regularizer for acquisition to ensure diversity in the training dataset; 3.) adaptation of Dice log-likelihood for Stein variational gradient descent (SVGD). The results indicate an improvement in terms of data reduction by achieving full accuracy while only using 22.69% and 48.85% of the available data for each dataset, respectively. [ABSTRACT FROM AUTHOR]

Performing deep neural network (DNN) inference in real time requires excessive network resources, which poses a big challenge to the resource-limited industrial Internet of things (IIoT) networks. To address the challenge, in this paper, we introduce an end-edge-cloud orchestration architecture, in which the inference task assignment and DNN model placement are flexibly coordinated. Specifically, the DNN models, trained and pre-stored in the cloud, are properly placed at the end and edge to perform DNN inference. To achieve efficient DNN inference, a multi-dimensional resource management problem is formulated to maximize the average inference accuracy while satisfying the strict delay requirements of inference tasks. Due to the mix-integer decision variables, it is difficult to solve the formulated problem directly. Thus, we transform the formulated problem into a Markov decision process which can be solved efficiently. Furthermore, a deep reinforcement learning based resource management scheme is proposed to make real-time optimal resource allocation decisions. Simulation results are provided to demonstrate that the proposed scheme can efficiently allocate the available spectrum, caching, and computing resources, and improve average inference accuracy by 31.4 % compared with the deep deterministic policy gradient benchmark. [ABSTRACT FROM AUTHOR]

This study proposes high-frequency container port congestion measures based on Automatic Identification System (AIS) data. Vessel movement information of 3,957 container ships from March 2017 to April 2017 is included. The world top 20 container ports' berth and anchorage areas are identified through Density Based Spatial Clustering of Applications with Noise (DBSCAN) and convex hull methods, and their hourly port congestion statuses are depicted in terms of the traffic volume and turnaround time. The constructed congestion measures overcome the disadvantages of the traditionally used port or industry data, which is heterogenous, behind the time and not easy to obtain publicly. The higher frequency (hourly) of the proposed measures can effectively monitor any slight change in port performance. A Long Short-Term Memory (LSTM) neural network model is then proposed for congestion prediction using constructed congestion measures. Point prediction and sequence prediction are both performed. We innovatively introduce congestion propagation effects into the prediction model as input features. Using Shanghai, Singapore and Ningbo ports as case studies, results show that the inclusion of congestion propagation effect can improve the prediction performance especially for sequence prediction. This study provides significant implications and decision support for container shipping market participants. [ABSTRACT FROM AUTHOR]

Echocardiography is the first-line diagnostic technique for heart diseases. Although artificial intelligence techniques have made great improvements in the analysis of echocardiography, the major limitations remain to be the built neural networks are normally adapted to a few diseases and specific equipment. Here, we present an end-to-end deep learning framework named AIEchoDx that differentiates four common cardiovascular diseases (Atrial Septal Defect, Dilated Cardiomyopathy, Hypertrophic Cardiomyopathy, prior Myocardial Infarction) from normal subjects with performance comparable to that of consensus of three senior cardiologists in AUCs (99.50% vs 99.26%, 98.75% vs 92.75%, 99.57% vs 97.21%, 98.52% vs 84.20%, and 98.70% vs 89.41%), respectively. Meanwhile, AIEchoDx accurately recognizes critical lesion regions of interest along with each disease by visualizing the decision-making process. Furthermore, our analysis indicates that heterogeneous diseases, like dilated cardiomyopathy, could be classified into two phenogroups with distinct clinical characteristics. Finally, AIEchoDx performs efficiently as an anomaly detection tool when applying handheld device-produced videos. Together, AIEchoDx provides a potential diagnostic assistant tool in either cart-based echocardiography equipment or handheld echocardiography device for primary and point-of-care medical personnel with high diagnostic performance, and the application of lesion region identification and heterogeneous disease phenogrouping, which may broaden the application of artificial intelligence in echocardiography. [ABSTRACT FROM AUTHOR]

Background and Aim Methods Results Conclusions Reliable bowel preparation assessment is important in colonoscopy. However, current scoring systems are limited by laborious and time‐consuming tasks and interobserver variability. We aimed to develop an artificial intelligence (AI) model to assess bowel cleanliness and evaluate its clinical applicability.A still image‐driven AI model to assess the Boston Bowel Preparation Scale (BBPS) was developed and validated using 2361 colonoscopy images. For evaluating real‐world applicability, the model was validated using 113 10‐s colonoscopy video clips and 30 full colonoscopy videos to identify “adequate (BBPS 2–3)” or “inadequate (BBPS 0–1)” preparation. The model was tested with an external dataset of 29 colonoscopy videos. The clinical applicability of the model was evaluated using 225 consecutive colonoscopies. Inter‐rater variability was analyzed between the AI model and endoscopists.The AI model achieved an accuracy of 94.0% and an area under the receiver operating characteristic curve of 0.939 with the still images. Model testing with an external dataset showed an accuracy of 95.3%, an area under the receiver operating characteristic curve of 0.976, and a sensitivity of 100% for the detection of inadequate preparations. The clinical applicability study showed an overall agreement rate of 85.3% between endoscopists and the AI model, with Fleiss' kappa of 0.686. The agreement rate was lower for the right colon compared with the transverse and left colon, with Fleiss' kappa of 0.563, 0.575, and 0.789, respectively.The AI model demonstrated accurate bowel preparation assessment and substantial agreement with endoscopists. Further refinement of the AI model is warranted for effective monitoring of qualified colonoscopy in large‐scale screening programs. [ABSTRACT FROM AUTHOR]

Objective: Automated methods for prostate segmentation on MRI are typically developed under ideal scanning and anatomical conditions. This study evaluates three different prostate segmentation AI algorithms in a challenging population of patients with prior treatments, variable anatomic characteristics, complex clinical history, or atypical MRI acquisition parameters. Materials and methods: A single institution retrospective database was queried for the following conditions at prostate MRI: prior prostate-specific oncologic treatment, transurethral resection of the prostate (TURP), abdominal perineal resection (APR), hip prosthesis (HP), diversity of prostate volumes (large ≥ 150 cc, small ≤ 25 cc), whole gland tumor burden, magnet strength, noted poor quality, and various scanners (outside/vendors). Final inclusion criteria required availability of axial T2-weighted (T2W) sequence and corresponding prostate organ segmentation from an expert radiologist. Three previously developed algorithms were evaluated: (1) deep learning (DL)-based model, (2) commercially available shape-based model, and (3) federated DL-based model. Dice Similarity Coefficient (DSC) was calculated compared to expert. DSC by model and scan factors were evaluated with Wilcox signed-rank test and linear mixed effects (LMER) model. Results: 683 scans (651 patients) met inclusion criteria (mean prostate volume 60.1 cc [9.05–329 cc]). Overall DSC scores for models 1, 2, and 3 were 0.916 (0.707–0.971), 0.873 (0–0.997), and 0.894 (0.025–0.961), respectively, with DL-based models demonstrating significantly higher performance (p < 0.01). In sub-group analysis by factors, Model 1 outperformed Model 2 (all p < 0.05) and Model 3 (all p < 0.001). Performance of all models was negatively impacted by prostate volume and poor signal quality (p < 0.01). Shape-based factors influenced DL models (p < 0.001) while signal factors influenced all (p < 0.001). Conclusion: Factors affecting anatomical and signal conditions of the prostate gland can adversely impact both DL and non-deep learning-based segmentation models. [ABSTRACT FROM AUTHOR]

Although ship detection methods based on deep learning have achieved remarkable progress, the design of the universal ship detection (USD) system is rarely studied. The main challenge of USD lies in the notorious domain bias and shift problem across multiple domains. This article implements USD based on domain-invariant representations to alleviate this issue. Specifically, the proposed method integrates a multilevel domain classification network (MDCN) and a domain-centric cut-paste module (DCM). First, the backbone network is facilitated to learn domain-independent image features from multiple domains through MDCN, thereby reducing the disturbance of domain-specific features to universal detector. Furthermore, the proposed method combines the domain-related synthetic samples generated by DCM to provide MDCN with strong supervision information, which further motivates the network to be more attentive to the domain-invariant representations at the instance level. Finally, we conduct experiments on multiple ship datasets in the synthetic aperture radar (SAR) and optical domains to verify the effectiveness of the method. The results show that the proposed method outperforms baseline by around 2.95% average precision (AP50), which achieves an effective USD system by complementing the information between domain-invariant representations. [ABSTRACT FROM AUTHOR]

• The temperature distribution on weldment sidewall exhibits a certain degree of separability under different offset degrees of cladding layer. • A cladding layer offset recognition method based on infrared sensing and deep learning for WAAM is proposed. • The proposed method effectively solves the problem of arc light interference in the cladding layer offset recognition method based on visual perception. In the wire arc additive manufacturing (WAAM) procedure, the stability of the cladding layer's formation directly impacts the final dimensional precision of the WAAM weldment. To achieve high-precision forming of the WAAM weldment, online monitoring of the cladding layer offset is essential. This article explores the monitoring technology for the cladding layer offset in the WAAM process and introduces a cladding layer offset monitoring technique using infrared temperature measurement and deep learning. First, a temperature field measurement system for the WAAM weldment is set up using an infrared thermal imager. Next, a region of interest (ROI) is selected on the weldment sidewall. The optimised histogram of oriented gradient (HOG) algorithm is then employed to extract the temperature distribution characteristics of the ROI. The separability of the temperature field on the weldment sidewall is analysed under various offset degrees of the cladding layer. Finally, using the temperature image of the ROI as input, a cladding layer offset recognition model is developed using a convolutional neural network (CNN). This study also evaluates the effect of the selected ROI position on the recognition precision of the model for the cladding layer offset. Experimental results indicate that the recognition accuracy of the model for cladding layer offset, as presented in this article, exceeds 99%. Moreover, the model demonstrates strong robustness, further enriching the intelligent monitoring methods for the WAAM process. [ABSTRACT FROM AUTHOR]

Purpose: To develop two deep learning-based systems for diagnosing and localizing pneumothorax on portable supine chest X-rays (SCXRs). Methods: For this retrospective study, images meeting the following inclusion criteria were included: (1) patient age ≥ 20 years; (2) portable SCXR; (3) imaging obtained in the emergency department or intensive care unit. Included images were temporally split into training (1571 images, between January 2015 and December 2019) and testing (1071 images, between January 2020 to December 2020) datasets. All images were annotated using pixel-level labels. Object detection and image segmentation were adopted to develop separate systems. For the detection-based system, EfficientNet-B2, DneseNet-121, and Inception-v3 were the architecture for the classification model; Deformable DETR, TOOD, and VFNet were the architecture for the localization model. Both classification and localization models of the segmentation-based system shared the UNet architecture. Results: In diagnosing pneumothorax, performance was excellent for both detection-based (Area under receiver operating characteristics curve [AUC]: 0.940, 95% confidence interval [CI]: 0.907–0.967) and segmentation-based (AUC: 0.979, 95% CI: 0.963–0.991) systems. For images with both predicted and ground-truth pneumothorax, lesion localization was highly accurate (detection-based Dice coefficient: 0.758, 95% CI: 0.707–0.806; segmentation-based Dice coefficient: 0.681, 95% CI: 0.642–0.721). The performance of the two deep learning-based systems declined as pneumothorax size diminished. Nonetheless, both systems were similar or better than human readers in diagnosis or localization performance across all sizes of pneumothorax. Conclusions: Both deep learning-based systems excelled when tested in a temporally different dataset with differing patient or image characteristics, showing favourable potential for external generalizability. [ABSTRACT FROM AUTHOR]

• A lightweight and efficient BLS-based crack detection model is proposed to automatically classify and identify crack images. • The bottleneck structure of MobileNetV3 is introduced as the convolution module of the proposed broad learning system for feature extraction. • By training and validating with public and realistic crack image datasets, the proposed model achieves fast and efficient incremental learning as training samples increase. • The sliding window algorithm is introduced to efficiently recognize and locate cracks in the large-size image. Concrete surface crack detection is an important means to ensure infrastructure security, including bridges and tunnels. In recent years, computer vision has been widely adopted for concrete surface crack detection, and the intelligent crack detection method based on computer vision has achieved good results with the breakthrough of the deep learning algorithm. However, there are also some challenges, such as complex architecture, high computing burden, and time-consuming training. In this paper, a lightweight broad learning system detection method called MobileNetV3-BLS is proposed. First, the inverted residual structure with the linear bottleneck of MobileNetV3 is introduced as the convolution module of the proposed broad learning system for feature extraction. Then the extracted features are employed to randomly generate both mapping and enhancement nodes in a broad learning system, which are used to train the crack detection model. The proposed MobileNetV3-BLS can improve crack detection accuracy and training speed compared with other classical deep learning networks. On the other hand, the proposed method is a dynamic network of which the weights can be updated quickly to achieve incremental learning by adding new nodes and samples. [ABSTRACT FROM AUTHOR]

Coastal ship tracking is used in many applications, such as autonomous navigation, maritime rescue, and environmental monitoring. Many general object-tracking methods based on deep learning have been explored for ship tracking, but they often fail to accurately track ships in challenging scenarios, such as occlusion, scale variation, and motion blur. We propose a memory-guided perception network (MGPN) to address these issues. MGPN has two main innovative improvements. The dynamic memory mechanism (DMM) in the proposed method stores past features of the tracked target to enhance the model's feature fusion capability in the temporal dimension. Meanwhile, the hierarchical context-aware module (HCAM) enables the interaction of different scales, global and local information, to address the scale discrepancy of targets and improve the feature fusion capability in the spatial dimension. These innovations enhance the robustness of tracking and reduce inaccuracies in the bounding boxes. We conducted an in-depth ablation study to demonstrate the effectiveness of DMM and HCAM. Finally, influenced by the above two points, MGPN has achieved state-of-the-art performance on a large offshore ship tracking dataset, which contains challenging scenarios such as complex backgrounds, ship occlusion, and varying scales. [ABSTRACT FROM AUTHOR]

Indocyanine green (ICG) has been used in clinical practice for more than 40 years and its safety and preferential accumulation in tumors has been reported for various tumor types, including colon cancer. However, reports on clinical assessments of ICG-based molecular endoscopy imaging for precancerous lesions are scarce. We determined visualization ability of ICG fluorescence endoscopy in colitis-associated colon cancer using 30 lesions from an azoxymethane/dextran sulfate sodium (AOM/DSS) mouse model and 16 colon cancer patient tissue-samples. With a total of 60 images (optical, fluorescence) obtained during endoscopy observation of mouse colon cancer, we used deep learning network to predict four classes (Normal, Dysplasia, Adenoma, and Carcinoma) of colorectal cancer development. ICG could detect 100% of carcinoma, 90% of adenoma, and 57% of dysplasia, with little background signal at 30 min after injection via real-time fluorescence endoscopy. Correlation analysis with immunohistochemistry revealed a positive correlation of ICG with inducible nitric oxide synthase (iNOS; r > 0.5). Increased expression of iNOS resulted in increased levels of cellular nitric oxide in cancer cells compared to that in normal cells, which was related to the inhibition of drug efflux via the ABCB1 transporter down-regulation resulting in delayed retention of intracellular ICG. With artificial intelligence training, the accuracy of image classification into four classes using data sets, such as fluorescence, optical, and fluorescence/optical images was assessed. Fluorescence images obtained the highest accuracy (AUC of 0.8125) than optical and fluorescence/optical images (AUC of 0.75 and 0.6667, respectively). These findings highlight the clinical feasibility of ICG as a detector of precancerous lesions in real-time fluorescence endoscopy with artificial intelligence training and suggest that the mechanism of ICG retention in cancer cells is related to intracellular nitric oxide concentration. [ABSTRACT FROM AUTHOR]

Although the existing deep-learning-based hyperspectral image (HSI) denoising methods have achieved tremendous success, recovering high-quality HSIs in complex scenes that contain mixed noise is still challenging. Besides, these methods have not fully explored the local and global spatial–spectral information of HSIs. To address the above issues, a novel HSI mixed noise removal network called subspace projection attention and residual channel attention network (SPARCA-Net) is proposed. Specifically, we propose an orthogonal subspace projection attention (OSPA) module to adaptively learn to generate bases of the signal subspace and project the input into such space to remove noise. By leveraging the local and global spatial relations, OSPA is able to reconstruct the local structure of the feature maps more precisely. We further propose a residual channel attention (RCA) module to emphasize the interdependence between feature maps and exploit the global channel correlation of them, which could enhance the channel-wise adaptive learning. In addition, multiscale joint spatial–spectral input and residual learning strategies are employed to capture multiscale spatial–spectral features and reduce the degradation problem, respectively. Synthetic and real HSI data experiments demonstrated that the proposed HSI denoising network outperforms many of the advanced methods in both quantitative and qualitative assessments. [ABSTRACT FROM AUTHOR]

Objectives: Cardiovascular border (CB) analysis is the primary method for detecting and quantifying the severity of cardiovascular disease using posterior-anterior chest radiographs (CXRs). This study aimed to develop and validate a deep learning–based automatic CXR CB analysis algorithm (CB_auto) for diagnosing and quantitatively evaluating valvular heart disease (VHD). Methods: We developed CB_auto using 816 normal and 798 VHD CXRs. For validation, 640 normal and 542 VHD CXRs from three different hospitals and 132 CXRs from a public dataset were assigned. The reliability of the CB parameters determined by CB_auto was evaluated. To evaluate the differences between parameters determined by CB_auto and manual CB drawing (CB_hand), the absolute percentage measurement error (APE) was calculated. Pearson correlation coefficients were calculated between CB_hand and echocardiographic measurements. Results: CB parameters determined by CB_auto yielded excellent reliability (intraclass correlation coefficient > 0.98). The 95% limits of agreement for the cardiothoracic ratio were 0.00 ± 0.04% without systemic bias. The differences between parameters determined by CB_auto and CB_hand as defined by the APE were < 10% for all parameters except for carinal angle and left atrial appendage. In the public dataset, all CB parameters were successfully drawn in 124 of 132 CXRs (93.9%). All CB parameters were significantly greater in VHD than in normal controls (all p < 0.05). All CB parameters showed significant correlations (p < 0.05) with echocardiographic measurements. Conclusions: The CB_auto system empowered by deep learning algorithm provided highly reliable CB measurements that could be useful not only in daily clinical practice but also for research purposes. Key Points: • A deep learning–based automatic CB analysis algorithm for diagnosing and quantitatively evaluating VHD using posterior-anterior chest radiographs was developed and validated. • Our algorithm (CB_auto) yielded comparable reliability to manual CB drawing (CB_hand) in terms of various CB measurement variables, as confirmed by external validation with datasets from three different hospitals and a public dataset. • All CB parameters were significantly different between VHD and normal control measurements, and echocardiographic measurements were significantly correlated with CB parameters measured from normal control and VHD CXRs. [ABSTRACT FROM AUTHOR]

Purpose: We aimed to investigate the ability of MRI radiomics features-based machine learning (ML) models to classify the time since stroke onset (TSS), which could aid in stroke assessment and treatment options. Methods: This study involved 84 patients with acute ischemic stroke due to anterior circulation artery occlusion (51 in the training cohort and 33 in the independent test cohort). Region of infarct segmentation was manually outlined by 3D-slicer software. Image processing including registration, normalization and radiomics features calculation were done in R (version 3.6.1). A total of 4312 radiomic features from each image sequence were captured and used in six ML models to estimate stroke onset time for binary classification (≤ 4.5 h). Receiver-operating characteristic curve (ROC) and other parameters were calculated to evaluate the performance of the models in both training and test cohorts. Results: Twelve radiomics and six clinic features were selected to construct the ML models for TSS classification. The deep learning model-based DWI/ADC radiomic features performed the best for binary TSS classification in the independent test cohort, with an AUC of 0.754, accuracy of 0.788, sensitivity of 0.952, specificity of 0.500, positive predictive value of 0.769, and negative predictive value of 0.857, respectively. Furthermore, adding clinical information did not improve the performance of the DWI/ADC-based deep learning model. The TSS prediction models can be visited at: http://123.57.65.199:3838/deeptss/. Conclusions: A unique deep learning model based on DWI/ADC radiomic features was constructed for TSS classification, which could aid in decision making for thrombolysis in patients with unknown stroke onset. [ABSTRACT FROM AUTHOR]

• Cycle-consistent adversarial denoising for unsupervised cardiac CT data. • Wavelet-based preprocessing to disentangle noise components from images. • Effective noise reduction of 4% extreme-low-dose CT images without creating artifacts. • Extensive evaluation of denoised image quality on a data set of 149 patients. [Display omitted] In electrocardiography (ECG) gated cardiac CT angiography (CCTA), multiple images covering the entire cardiac cycle are taken continuously, so reduction of the accumulated radiation dose could be an important issue for patient safety. Although ECG-gated dose modulation (so-called ECG pulsing) is used to acquire many phases of CT images at a low dose, the reduction of the radiation dose introduces noise into the image reconstruction. To address this, we developed a high performance unsupervised deep learning method using noise disentanglement that can effectively learn the noise patterns even from extreme low dose CT images. For noise disentanglement, we use a wavelet transform to extract the high-frequency signals that contain the most noise. Since matched low-dose and high-dose cardiac CT data are impossible to obtain in practice, our neural network was trained in an unsupervised manner using cycleGAN for the extracted high frequency signals from the low-dose and unpaired high-dose CT images. Once the network is trained, denoised images are obtained by subtracting the estimated noise components from the input images. Image quality evaluation of the denoised images from only 4% dose CT images was performed by experienced radiologists for several anatomical structures. Visual grading analysis was conducted according to the sharpness level, noise level, and structural visibility. Also, the signal-to-noise ratio was calculated. The evaluation results showed that the quality of the images produced by the proposed method is much improved compared to low-dose CT images and to the baseline cycleGAN results. The proposed noise-disentangled cycleGAN with wavelet transform effectively removed noise from extreme low-dose CT images compared to the existing baseline algorithms. It can be an important denoising platform for low-dose CT. [ABSTRACT FROM AUTHOR]