Search

Objective: Cavernous sinus invasion (CSI) plays a pivotal role in determining management in pituitary adenomas. The study aimed to develop a Convolutional Neural Network (CNN) model to diagnose CSI in multiple centers. Methods: A total of 729 cases were retrospectively obtained in five medical centers with (n = 543) or without CSI (n = 186) from January 2011 to December 2021. The CNN model was trained using T1-enhanced MRI from two pituitary centers of excellence (n = 647). The other three municipal centers (n = 82) as the external testing set were imported to evaluate the model performance. The area-under-the-receiver-operating-characteristic-curve values (AUC-ROC) analyses were employed to evaluate predicted performance. Gradient-weighted class activation mapping (Grad-CAM) was used to determine models' regions of interest. Results: The CNN model achieved high diagnostic accuracy (0.89) in identifying CSI in the external testing set, with an AUC-ROC value of 0.92 (95% CI, 0.88–0.97), better than CSI clinical predictor of diameter (AUC-ROC: 0.75), length (AUC-ROC: 0.80), and the three kinds of dichotomizations of the Knosp grading system (AUC-ROC: 0.70–0.82). In cases with Knosp grade 3A (n = 24, CSI rate, 0.35), the accuracy the model accounted for 0.78, with sensitivity and specificity values of 0.72 and 0.78, respectively. According to the Grad-CAM results, the views of the model were confirmed around the sellar region with CSI. Conclusions: The deep learning model is capable of accurately identifying CSI and satisfactorily able to localize CSI in multicenters. [ABSTRACT FROM AUTHOR]

Despite demonstrating exceptional inversion production for synthetic data, the application of deep learning (DL) inversion methods to invert realistic magnetotelluric (MT) measurements, which are inevitably contaminated by noise in acquisition, poses a significant challenge. Hence, to facilitate DL inversion for realistic MT measurements, this work explores developing a noise-robust MT DL inversion method by generating targeted noisy training datasets and constructing a physics-informed neural network. Different from most previous works that only considered the noise of one fixed distribution and level, we propose three noise injection strategies and compare their combinations to mitigate the adverse effect of measurement noise on MT DL inversion results: (1) add synthetic relative noise obeying Gaussian distribution; (2) propose a multiwindow Savitzky–Golay (MWSG) filtering scheme to extract potential and possible noise from the target field data and then introduce them into training data; (3) create an augmented training dataset based on the former two strategies. Moreover, we employ the powerful Swin Transformer as the backbone network to construct a U-shaped DL model (SwinTUNet), based on which a physics-informed SwinTUNet (PISwinTUNet) is implemented to further enhance its generalization ability. In synthetic examples, the proposed noise injection strategies demonstrate impressive inversion effects, regardless of whether they are contaminated by familiar or unfamiliar noise. In a field example, the combination of three strategies drives PISwinTUNet to produce considerably faithful reconstructions for subsurface resistivity structures and outperform the classical deterministic Occam inversions. The experimental results show that the proposed noise-robust DL inversion method based on the noise injection strategies and physics-informed DL architecture holds great promise in processing MT field data. [ABSTRACT FROM AUTHOR]

In the 2030s, a new era of gravitational wave (GW) observations will dawn as multiple space-based GW detectors, such as the Laser Interferometer Space Antenna, Taiji, and TianQin, will open the millihertz window for GW astronomy. These detectors are poised to detect a multitude of GW signals emitted by different sources. It is a challenging task for GW data analysis to recover the parameters of these sources at a low computational cost. Generally, the matched filtering approach entails exploring an extensive parameter space for all resolvable sources, incurring a substantial cost owing to the generation of GW waveform templates. To alleviate the challenge, we make an attempt to perform parameter inference for coalescing massive black hole binaries (MBHBs) using deep learning. The model trained in this work has the capability to produce 50,000 posterior samples for the redshifted total mass, mass ratio, coalescence time, and luminosity distance of an MBHB in about twenty seconds. Our model can serve as an effective data pre-processing tool, reducing the volume of parameter space by more than four orders of magnitude for MBHB signals with a signal-to-noise ratio larger than 100. Moreover, the model exhibits robustness when handling input data that contain multiple MBHB signals. [ABSTRACT FROM AUTHOR]

Background: MR scans used in radiotherapy can be partially truncated due to the limited field of view (FOV), affecting dose calculation accuracy in MR‐based radiation treatment planning. Purpose: We proposed a novel Compensation‐cycleGAN (Comp‐cycleGAN) by modifying the cycle‐consistent generative adversarial network (cycleGAN), to simultaneously create synthetic CT (sCT) images and compensate the missing anatomy from the truncated MR images. Methods: Computed tomography (CT) and T1 MR images with complete anatomy of 79 head‐and‐neck patients were used for this study. The original MR images were manually cropped 10–25 mm off at the posterior head to simulate clinically truncated MR images. Fifteen patients were randomly chosen for testing and the rest of the patients were used for model training and validation. Both the truncated and original MR images were used in the Comp‐cycleGAN training stage, which enables the model to compensate for the missing anatomy by learning the relationship between the truncation and known structures. After the model was trained, sCT images with complete anatomy can be generated by feeding only the truncated MR images into the model. In addition, the external body contours acquired from the CT images with full anatomy could be an optional input for the proposed method to leverage the additional information of the actual body shape for each test patient. The mean absolute error (MAE) of Hounsfield units (HU), peak signal‐to‐noise ratio (PSNR), and structural similarity index (SSIM) were calculated between sCT and real CT images to quantify the overall sCT performance. To further evaluate the shape accuracy, we generated the external body contours for sCT and original MR images with full anatomy. The Dice similarity coefficient (DSC) and mean surface distance (MSD) were calculated between the body contours of sCT and original MR images for the truncation region to assess the anatomy compensation accuracy. Results: The average MAE, PSNR, and SSIM calculated over test patients were 93.1 HU/91.3 HU, 26.5 dB/27.4 dB, and 0.94/0.94 for the proposed Comp‐cycleGAN models trained without/with body‐contour information, respectively. These results were comparable with those obtained from the cycleGAN model which is trained and tested on full‐anatomy MR images, indicating the high quality of the sCT generated from truncated MR images by the proposed method. Within the truncated region, the mean DSC and MSD were 0.85/0.89 and 1.3/0.7 mm for the proposed Comp‐cycleGAN models trained without/with body contour information, demonstrating good performance in compensating the truncated anatomy. Conclusions: We developed a novel Comp‐cycleGAN model that can effectively create sCT with complete anatomy compensation from truncated MR images, which could potentially benefit the MRI‐based treatment planning. [ABSTRACT FROM AUTHOR]

Due to the strong capability of building complex nonlinear mapping without involving linearization theory and high prediction efficiency; the deep learning (DL) technique applied to solve geophysical inverse problems has been a subject of growing interest. Currently, most DL-based inversion approaches are fully data-driven (namely standard deep learning), the performance of which largely depends on the training sample sets. However, due to the heavy burden of time and computational resources, it can be challenging to supply such a massive and exhaustive training dataset for generic realistic exploration scenarios and to perform network training. In this work, based on the recent advances in physics-based networks, the physical laws of magnetotelluric (MT) wave propagation is incorporated into a purely data-driven DL approach (PlainDNN) and thus builds a physics-driven DL MT inversion scheme (PhyDNN). In this scheme, the forward operator modeling MT wave propagation is integrated into the network training loop, in the form of minimizing a hybrid loss objective function composed of the data-driven model misfit and physics-based data misfit, to guide the network training. Consequently, the proposed PhyDNN method will take the advantage of the fully data-driven DL and conventional physics-based deterministic methods, allowing it to deal with complex realistic exploration scenarios. Quantitative and qualitative analysis results demonstrate that the PhyDNN can honor the physical laws of the MT inverse problem, and with other conditions unchanged, the PhyDNN outperforms the PlainDNN and the classical deterministic Occam inversion method. When processing field data, the PhyDNN method yields considerably impressive inversion results compared to the Occam method, and the corresponding simulated MT responses agree well with the real measurements, which confirms the effectiveness and applicability of the PhyDNN method. [ABSTRACT FROM AUTHOR]

A keloid results from abnormal wound healing, which has different blood perfusion and growth states among patients. Active monitoring and treatment of actively growing keloids at the initial stage can effectively inhibit keloid enlargement and has important medical and aesthetic implications. LSCI (laser speckle contrast imaging) has been developed to obtain the blood perfusion of the keloid and shows a high relationship with the severity and prognosis. However, the LSCI-based method requires manual annotation and evaluation of the keloid, which is time consuming. Although many studies have designed deep-learning networks for the detection and classification of skin lesions, there are still challenges to the assessment of keloid growth status, especially based on small samples. This retrospective study included 150 untreated keloid patients, intensity images, and blood perfusion images obtained from LSCI. A newly proposed workflow based on cascaded vision transformer architecture was proposed, reaching a dice coefficient value of 0.895 for keloid segmentation by 2% improvement, an error of 8.6 ± 5.4 perfusion units, and a relative error of 7.8% ± 6.6% for blood calculation, and an accuracy of 0.927 for growth state prediction by 1.4% improvement than baseline. [ABSTRACT FROM AUTHOR]

Current groundwater models face significant challenges in their implementation due to heavy computational burdens. To overcome this, our work proposes a cost-effective emulator that efficiently and accurately forecasts the impact of abstraction in an aquifer. Our approach uses a deep neural operator (DeepONet) framework to learn operators that map between infinite-dimensional function spaces via deep neural networks. The goal is to infer the distribution of hydraulic heads in a confined aquifer in the presence of a pumping well. We successfully tested the DeepONet framework on multiple problems, including forward time-dependent problems, an inverse analysis, and a nonlinear system. Additionally, we propose a novel extension of the DeepONet-based architecture to generate accurate predictions for varied hydraulic conductivity fields and pumping well locations that are unseen during training. Our emulator's predictions match the target data with excellent performance, demonstrating that the proposed model can act as an efficient and fast tool to support a range of tasks that require repetitive forward numerical simulations or inverse simulations of groundwater flow problems. Overall, our work provides a promising avenue for developing cost-effective and accurate groundwater models. • We propose, for the first time, a neural operator (DeepONet) as a surrogate model to efficiently and accurately emulate the impact of abstraction in the distribution of hydraulic heads in confined, heterogeneous aquifers. • The proposed method can efficiently learn solutions of both the forward and inverse operator. • By proposing a modification to the original formulation of DeepONet, we have achieved precise predictions for varied groundwater head distributions, accommodating the random positioning of wells in the heterogeneous aquifer. • Our experiments demonstrate DeepONet's capability in tackling tasks demanding repetitive forward numerical simulations, even in time-dependent and nonlinear setups. [ABSTRACT FROM AUTHOR]

Cystic lesions are common lesions of the sellar region with various pathological types, including pituitary apoplexy, Rathke's cleft cyst, cystic craniopharyngioma, etc. Suggested surgical approaches are not unique when dealing with different cystic lesions. However, cystic lesions with different pathological types were hard to differentiate on MRI with the naked eye by doctors. This study aimed to distinguish different pathological types of cystic lesions in the sellar region using preoperative magnetic resonance imaging (MRI). Radiomics and deep learning approaches were used to extract features from gadolinium-enhanced MRIs of 399 patients enrolled at Peking Union Medical College Hospital over the past 15 years. Paired imaging differentiations were performed on four subtypes, including pituitary apoplexy, cystic pituitary adenoma (cysticA), Rathke's cleft cyst, and cystic craniopharyngioma. Results showed that the model achieved an average AUC value of 0.7685. The model based on a support vector machine could distinguish cystic craniopharyngioma from Rathke's cleft cyst with the highest AUC value of 0.8584. However, distinguishing cystic apoplexy from pituitary apoplexy was difficult and almost unclassifiable with any algorithms on any feature set, with the AUC value being only 0.6641. Finally, the proposed methods achieved an average Accuracy of 0.7532, which outperformed the traditional clinical knowledge-based method by about 8%. Therefore, in this study, we first fill the gap in the existing literature and provide a non-invasive method for accurately differentiating between these lesions, which could improve preoperative diagnosis accuracy and help to make surgery plans in clinical work. [ABSTRACT FROM AUTHOR]

Image inpainting algorithms have a wide range of applications, which can be used for object removal in digital images. With the development of semantic level image inpainting technology, this brings great challenges to blind image forensics. In this case, many conventional methods have been proposed which have disadvantages such as high time complexity and low robustness to postprocessing operations. Therefore, this paper proposes a mask regional convolutional neural network (Mask R-CNN) approach for patch-based inpainting detection. According to the current research, many deep learning methods have shown the capacity for segmentation tasks when labeled datasets are available, so we apply a deep neural network to the domain of inpainting forensics. This deep learning model can distinguish and obtain different features between the inpainted and noninpainted regions. To reduce the missed detection areas and improve detection accuracy, we also adjust the sizes of the anchor scales due to the inpainting images and replace the original nonmaximum suppression single threshold with an improved nonmaximum suppression (NMS). The experimental results demonstrate this intelligent method has better detection performance over recent approaches of image inpainting forensics. [ABSTRACT FROM AUTHOR]

The coalescences of massive black hole binaries are one of the main targets of space-based gravitational wave observatories. Such gravitational wave sources are expected to be accompanied by electromagnetic emissions. Low latency detection of the massive black hole mergers provides a start point for a global-fit analysis to explore the large parameter space of signals simultaneously being present in the data but at great computational cost. To alleviate this issue, we present a deep learning method for rapidly searching for signals of massive black hole binaries in gravitational wave data. Our model is capable of processing a year of data, simulated from the LISA data challenge, in only several seconds, while identifying all coalescences of massive black hole binaries with no false alarms. We further demonstrate that the model shows robust resistance to a wide range of generalization cases, including various waveform families and updated instrumental configurations. This method offers an effective approach that combines advances in artificial intelligence to open a new pathway for space-based gravitational wave observations. [ABSTRACT FROM AUTHOR]

• A detection neural network is proposed for the cannibalism behavior of juvenile fish. • The network improves the detection accuracy by adding an attention mechanism and changing the feature fusion network. • The network improves the detection speed by using lightweight modules. • Experiments show that the algorithm can detect the cannibalism behavior in aquaculture scenes in real time. Special behavior detection of fish is an effective way to ensure fish welfare and improve the intelligent level of aquaculture. However, due to the influence of the breeding environment and physiological factors, fish are prone to cannibalism during the juvenile period. In order to effectively monitor the behavior of the fry, this article proposes a real-time detection scheme based on the improved You Only Look Once (YOLO)-v5 to detect the cannibalism of grouper fry in recirculating aquaculture system(RAS). The specific improvements are as follows: (1) Multi-head attention mechanism is used to obtain global information in the last block of the backbone network (2) According to the idea of BIfpn, nodes that contribute little in the feature network which combines different features the neck is deleted. At the same time, feature fusion is added for nodes whose input and output are in the same layer (3) The lightweight general upsampling operator Carafe is used to replace the original upsampling. The experimental results show that the improved YOLOv5 model achieves 97% accuracy in the detection of cannibalism behavior of juvenile fish, effectively solving the problems of small targets, severe occlusion and motion blur in the culture environment. Meanwhile, compared with the YOLOv5s model, the improved model has 12.6% and 14% improvement in detection accuracy and processing speed, respectively, and achieves the requirements of high accuracy and real-time detection. In the actual aquaculture environment, farmers can take corresponding measures according to the detection results, which can effectively improve the survival rate and economic benefits of aquaculture. [ABSTRACT FROM AUTHOR]

Synthetic aperture radar (SAR) is a powerful tool for monitoring sea states in terms of the significant wave height (SWH). Regarding the specific wave mode, to date, the previous empirical models for estimating SWH from SAR data rely on single polarization. In the emerging deep learning era, few published quad-polarized SAR SWH retrieval algorithms have been based on machine learning technique, and whether quad-polarimetry improves the skill of wave height estimation remains a question. Here we propose a deep residual convolutional neural network-based SAR SWH retrieval algorithm in quad-polarization. By collocating WaveWatch III sea state hindcasts and all available archives of quad-polarized Chinese Gaofen-3 SAR imagettes in wave mode, a database with approximately 30,000 matchups was employed to establish our deeply-learned network. The GaoFen-3 significant wave height retrievals were validated against the hindcast dataset independent of training along with altimeter observations. The result of good consistency in terms of a root mean square error of 0.32 m (under sea state conditions of approximately 0.5–7.0 m) outperforms the existing Gaofen-3 wave height retrieval algorithms. Additionally, this paper introduces a discussion about the contribution of polarizations by comparing SWH derived from single-, dual- and quad-polarized deep convolutional neural networks. Single-polarized Gaofen-3 SAR data are found to be sufficient to provide accurate estimates compared to quad-polarization via a deep learning model under moderate sea conditions. Exploitation of SAR quad-polarimetry information will improve SAR wave height retrievals under high sea conditions. • We propose residual networks for estimating wave heights from Gaofen-3. • Our approach outperformed the existing Gaofen-3 wave height retrieval methods. • Single polarization is sufficient for accurately estimating moderate wave heights. • Quad-polarized data enhance the SAR wave estimating capabilities for high states. [ABSTRACT FROM AUTHOR]

Trapped lee waves, and resultant turbulent rotors downstream, present a hazard for aviation and land‐based transport. Though high‐resolution numerical weather prediction models can represent such phenomena, there is currently no simple and reliable automated method for detecting the extent and characteristics of these waves in model output. Spectral transform methods have traditionally been used to detect and characterise regions of wave activity in model and observational data; however, these methods can be slow and have their limitations. Machine‐learning (ML) techniques offer a new and potentially fruitful method of tackling this problem. We demonstrate that a deep‐learning model can be trained to accurately recognise and label coherent regions of lee waves from vertical velocity data on a single level from a high‐resolution numerical weather prediction (NWP) model. Using transfer learning, wave characteristics (wavelength, orientation, and amplitude) can be extracted from the trained segmentation model. The use of synthetic wave fields with prescribed wave characteristics makes this transfer learning possible without the need to characterise real complex wave fields. Addition of noise to the synthetic data makes the models more robust when applied to more complex and noisy NWP data. The collection of trained models produced provides a valuable tool to investigate the prevalence and nature of lee wave activity, as well as a new way for forecasters to detect resolved waves. The deep‐learning model was more capable and quicker at detecting and characterising lee waves than a spectral technique was. This work is just one example of how already established ML techniques can be used to detect and characterise complex weather phenomena from NWP model output and observational data, and how the careful use of synthetic data can reduce the requirements for large volumes of hand‐labelled training data for ML models. [ABSTRACT FROM AUTHOR]

• A detection and tracking model combining Yolov5 and Siamrpn++ was proposed. • It achieved high-precision target detection in the case of occlusion and deformation. • Multi-objective tracking task was accomplished using a single case algorithm. • Detection and tracking of abnormal fish will improve fish welfare. In recirculating aquaculture system, the abnormal behavior of fish is usually caused by poor water quality, hypoxia or diseases. Delayed recognition of this behavior will lead to large number of fish deaths. Thus, real-time detection and tracking of fish that behaviors abnormally is an effective way to promote the fish welfare and to improve the survival rate as well as economic benefits of aquaculture. However, due to the high-density breeding, the targets in the fish images are often quite small and in occlusion, which causes high false detection and target loss rate. This article proposes a combined end-to-end neural network to detect and track the abnormal behavior of porphyry seabream. The detection algorithm passes the initial value of the target into the tracking algorithm, and the tracking algorithm tracks subsequent frames to achieve end-to-end abnormal fish behavior detection and achieve high-speed and accurate tracking of abnormal behavior individuals. In the target detection part, YOLOV5s is improved by incorporating multi-level features and adding feature mapping. Compared with the original network, the detection precision A P 50 : 95 is increased by 8.8% while A P 50 reaches 99.4%. In the target tracking part, this paper achieves multi-target tracking of abnormal fish based on single-target tracking algorithm SiamRPN++. The tracking precision is 76.7%. By combining the two approaches, individual fish with abnormal behavior can be detected precisely and tracked in real time. [ABSTRACT FROM AUTHOR]

Liver tumor semantic segmentation is a crucial task in medical image analysis that requires multiple MRI modalities. This paper proposes a novel coarse-to-fine fusion segmentation approach to detect and segment small liver tumors of various sizes. To enhance the segmentation accuracy of small liver tumors, the method incorporates a detection module and a CSR (convolution-SE-residual) module, which includes a convolution block, an SE (squeeze and excitation) module, and a residual module for fine segmentation. The proposed method demonstrates superior performance compared to conventional single-stage end-to-end networks. A private liver MRI dataset comprising 218 patients with a total of 3605 tumors, including 3273 tumors smaller than 3.0 cm, were collected for the proposed method. There are five types of liver tumors identified in this dataset: hepatocellular carcinoma (HCC); metastases of the liver; cholangiocarcinoma (ICC); hepatic cyst; and liver hemangioma. The results indicate that the proposed method outperforms the single segmentation networks 3D UNet and nnU-Net as well as the fusion networks of 3D UNet and nnU-Net with nnDetection. The proposed architecture was evaluated on a test set of 44 images, with an average Dice similarity coefficient (DSC) and recall of 86.9% and 86.7%, respectively, which is a 1% improvement compared to the comparison method. More importantly, compared to existing methods, our proposed approach demonstrates state-of-the-art performance in segmenting small objects with sizes smaller than 10 mm, achieving a Dice score of 85.3% and a malignancy detection rate of 87.5%. [ABSTRACT FROM AUTHOR]

Background: Accurate segmentation of pelvic bones is an initial step to achieve accurate detection and localisation of pelvic bone metastases. This study presents a deep learning-based approach for automated segmentation of normal pelvic bony structures in multiparametric magnetic resonance imaging (mpMRI) using a 3D convolutional neural network (CNN). Methods: This retrospective study included 264 pelvic mpMRI data obtained between 2018 and 2019. The manual annotations of pelvic bony structures (which included lumbar vertebra, sacrococcyx, ilium, acetabulum, femoral head, femoral neck, ischium, and pubis) on diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) images were used to create reference standards. A 3D U-Net CNN was employed for automatic pelvic bone segmentation. Additionally, 60 mpMRI data from 2020 were included and used to evaluate the model externally. Results: The CNN achieved a high Dice similarity coefficient (DSC) average in both testing (0.80 [DWI images] and 0.85 [ADC images]) and external (0.79 [DWI images] and 0.84 [ADC images]) validation sets. Pelvic bone volumes measured with manual and CNN-predicted segmentations were highly correlated (R2 value of 0.84–0.97) and in close agreement (mean bias of 2.6–4.5 cm3). A SCORE system was designed to qualitatively evaluate the model for which both testing and external validation sets achieved high scores in terms of both qualitative evaluation and concordance between two readers (ICC = 0.904; 95% confidence interval: 0.871–0.929). Conclusions: A deep learning-based method can achieve automated pelvic bone segmentation on DWI and ADC images with suitable quantitative and qualitative performance. [ABSTRACT FROM AUTHOR]

Image inpainting can effectively repair damaged areas, but it can also be a way of image tampering when it is used to remove meaningful content from an image. Therefore, this paper focuses on the research of inpainting forensics, and proposes a multi-task deep learning method. In order to enhance the learning of texture features, the corresponding local binary pattern channels are added to the input of the network. Furthermore, considering that the multi-task object detection network Mask R-CNN cannot fully utilize the features of all scale feature information during the FPN feature extraction phase, the network in this paper combines Feature Pyramid Networks and back connections to extract more features. This network model can detect not only the images tampered by traditional inpainting methods, but also the images inpainted by deep learning methods. Experimental results on two large public data sets demonstrate the superior performance of the proposed method. [ABSTRACT FROM AUTHOR]

Sea state estimation from wide-swath and frequent-revisit scatterometers, which are providing ocean winds in the routine, is an attractive challenge. In this study, state-of-the-art deep learning technology is successfully adopted to develop an algorithm for deriving significant wave height from Advanced Scatterometer (ASCAT) aboard MetOp-A. By collocating three years (2016–2018) of ASCAT measurements and WaveWatch III sea state hindcasts at a global scale, huge amount data points (>8 million) were employed to train the multi-hidden-layer deep learning model, which has been established to map the inputs of thirteen sea state related ASCAT observables into the wave heights. The ASCAT significant wave height estimates were validated against hindcast dataset independent on training, showing good consistency in terms of root mean square error of 0.5 m under moderate sea condition (1.0–5.0 m). Additionally, reasonable agreement is also found between ASCAT derived wave heights and buoy observations from National Data Buoy Center for the proposed algorithm. Results are further discussed with respect to sea state maturity, radar incidence angle along with the limitations of the model. Our work demonstrates the capability of scatterometers for monitoring sea state, thus would advance the use of scatterometers, which were originally designed for winds, in studies of ocean waves. [ABSTRACT FROM AUTHOR]

Aiming at the problems of poor self-adaptive ability in traditional feature extraction methods and weak generalization ability in single classifier under big data, an internal parameter-optimized Deep Belief Network (DBN) method based on grasshopper optimization algorithm (GOA) is proposed. First, the minimum Root Mean Square Error (RMSE) in the network training is taken as the fitness function, in which GOA is used to search for the optimal parameter combination of DBN. After that the learning rate and the number of batch learning in DBN which have great influence on the training error would be properly selected. At the same time, the optimal structure distribution of DBN is given through comparison. Then, FFT and linear normalization are introduced to process the original vibration signal of the gearbox, preprocess the data from multiple sensors and construct the input samples for DBN. Finally, combining with deep learning featured by powerful self-adaptive feature extraction and nonlinear mapping capabilities, the obtained samples are input into DBN for training, and the fault diagnosis model for gearbox based on DBN would be established. After several tests with the remaining samples, the diagnosis rate of the model could reach over 99.5%, which is far better than the traditional fault diagnosis method based on feature extraction and pattern recognition. The experimental results show that this method could effectively improve the self-adaptive feature extraction ability of the model as well as its accuracy of fault diagnosis, which has better generalization performance. [ABSTRACT FROM AUTHOR]

Objective: Accurate segmentation of liver tumors, which could help physicians make appropriate treatment decisions and assess the effectiveness of surgical treatment, is crucial for the clinical diagnosis of liver cancer. In this study, we propose a 4-dimensional (4D) deep learning model based on 3D convolution and convolutional long short-term memory (C-LSTM) for hepatocellular carcinoma (HCC) lesion segmentation. Methods: The proposed deep learning model utilizes 4D information on dynamic contrast enhanced (DCE) magnetic resonance imaging (MRI) images to assist liver tumor segmentation. Specifically, a shallow U-net based 3D CNN module was designed to extract 3D spatial domain features from each DCE phase, followed by a 4-layer C-LSTM network module for time domain information exploitation. The combined information of multi-phase DCE images and the manner by which tissue imaging features change on multi-contrast images allow the network to more effectively learn the characteristics of HCC, resulting in better segmentation performance. Results: The proposed model achieved a Dice score of 0.825± 0.077, a Hausdorff distance of 12.84± 8.14 mm, and a volume similarity of 0.891± 0.080 for liver tumor segmentation, which outperformed the 3D U-net model, RA-UNet model and other models in the ablation study in both internal and external test sets. Moreover, the performance of the proposed model is comparable to the nnU-Net model, which showed state-of-the-art performance in many segmentation tasks, with significantly reduced prediction time. Conclusion: The proposed 3D convolution and C-LSTM based model can achieve accurate segmentation of HCC lesions. [ABSTRACT FROM AUTHOR]

To build a prostate cancer (PCa) risk prediction model based on common clinical indicators to provide a theoretical basis for the diagnosis and treatment of PCa and to evaluate the value of artificial intelligence (Al) technology under healthcare data platforms. After preprocessing of the data from Population Health Data Archive, smuothly clipped absolute deviation (SCAD) was used to select features. Random forest (RF), support vector machine (SVM), back propagation neural network (BP), and convolutional neural network (CNN) were used to predict the risk of PCa, among which BP and CNN were used on the enhanced data by SMOTE. The performances of models were compared using area under the curve (AUC) of the receiving operating characteristic curve. After the optimal model was selected, we used the Shiny to develop an online calculator for PCa risk prediction based on predictive indicators. Inorganic phosphorus, triglycerides, and calcium were closely related to PCa in addition to the volume of fragmented tissue and free prostate-specific antigen (PSA). Among the four models, RF had the best performance in predicting PCa (accuracy: 96.80%; AUC: 0.975, 95% CI : 0.964-0.986). Followed by BP (accuracy: 85.36%; AUC: 0.892, 95% CI : 0.849-0.934) and SVM (accuracy: 82.67%; AUC: 0.824, 95% CI : 0.805-0.844). CNN performed worse (accuracy: 72.37%; AUC: 0.724, 95% CI : 0.670-0.779). An online platform for PCa risk prediction was developed based on the RF model and the predictive indicators. This study revealed the application value of traditional machine learning and deep learning models in disease risk prediction under healthcare data platform, proposed new ideas for PCa risk prediction in patients suspected for PCa and had undergone core needle biopsy. Besides, the online calculation may enhance the practicability of Al prediction technology and facilitate medical diagnosis. [ABSTRACT FROM AUTHOR]

Automatic classification of different species of fish is important for the comprehension of marine ecology, fish behaviour analysis, aquaculture management, and fish health monitoring. In recent years, many automatic classification methods have been developed, among which machine vision-based classification methods are widely used with the advantages of being fast and non-destructive. In addition, the successful application of rapidly emerging deep learning techniques in machine vision has brought new opportunities for fish classification. This paper provides an overview of machine vision models applied in the field of fish classification, followed by a detailed discussion of specific applications of various classification methods. Furthermore, the challenges and future research directions in the field of fish classification are discussed. This paper would help researchers and practitioners to understand the applicability of machine vision in fish classification and encourage them to develop advanced algorithms and models to address the complex problems that exist in fish classification practice. [ABSTRACT FROM AUTHOR]

Creating synthetic CT (sCT) from magnetic resonance (MR) images enables MR-based treatment planning in radiation therapy. However, the MR images used for MR-guided adaptive planning are often truncated in the boundary regions due to the limited field of view and the need for sequence optimization. Consequently, the sCT generated from these truncated MR images lacks complete anatomic information, leading to dose calculation error for MR-based adaptive planning. We propose a novel structure-completion generative adversarial network (SC-GAN) to generate sCT with full anatomic details from the truncated MR images. To enable anatomy compensation, we expand input channels of the CT generator by including a body mask and introduce a truncation loss between sCT and real CT. The body mask for each patient was automatically created from the simulation CT scans and transformed to daily MR images by rigid registration as another input for our SC-GAN in addition to the MR images. The truncation loss was constructed by implementing either an auto-segmentor or an edge detector to penalize the difference in body outlines between sCT and real CT. The experimental results show that our SC-GAN achieved much improved accuracy of sCT generation in both truncated and untruncated regions compared to the original cycleGAN and conditional GAN methods. • An automatic workflow to generate synthetic CT from truncated MR images. • Achieve anatomic compensation using a novel truncation loss function. • Improve synthetic CT image quality in both truncated and non-truncated regions. • Facilitate accurate dose calculation for MR-guided online adaptive planning. [ABSTRACT FROM AUTHOR]

The accurate identification of irreversible infarction and salvageable tissue is important in planning the treatments for acute ischemic stroke (AIS) patients. Computed tomographic perfusion (CTP) can be used to evaluate the ischemic core and deficit, covering most of the territories of anterior circulation, but many community hospitals and primary stroke centers do not have the capability to perform CTP scan in emergency situation. This study aimed to identify AIS lesions from widely available non-contrast computed tomography (NCCT) and CT angiography (CTA) using deep learning. A total of 345AIS patients from our emergency department were included. A multi-scale 3D convolutional neural network (CNN) was used as the predictive model with inputs of NCCT, CTA, and CTA+ (8 s delay after CTA) images. An external cohort with 108 patients was included to further validate the generalization performance of the proposed model. Strong correlations with CTP-RAPID segmentations (r = 0.84 for core, r = 0.83 for deficit) were observed when NCCT, CTA, and CTA+ images were all used in the model. The diagnostic decisions according to DEFUSE3 showed high accuracy when using NCCT, CTA, and CTA+ (0.90±0.04), followed by the combination of NCCT and CTA (0.87±0.04), CTA-alone (0.76±0.06), and NCCT-alone (0.53±0.09). [ABSTRACT FROM AUTHOR]

Background and objective. Early detection of hepatocellular carcinoma (HCC) is crucial for clinical management. Current studies have reported large HCC detections using automatic algorithms, but there is a lack of research on automatic detection of small HCCs (sHCCs). This study is to investigate the feasibility of automatic detection of sHCC (≤2 cm) based on pattern matching and deep learning (PM-DL) model. Materials and methods. A retrospective study included 5376 image sets from 56 cirrhosis patients (28 sHCC patients with 32 pathologically confirmed lesions and 28 non-HCC cirrhosis patients) in the training-validation cohort to build and validate the model through five-fold cross-validation. In addition, an external test cohort including 6144 image sets from 64 cirrhosis patients (32 sHCC patients with 38 lesions and 32 non-HCC cirrhosis patients) was applied to further verify the generalization ability of the model. The proposed PM-DL model consisted of three main steps: 3D co-registration and liver segmentation, screening of suspicious lesions on diffusion-weighted imaging images based on pattern matching algorithm, and identification/segmentation of sHCC lesions on dynamic contrast-enhanced images with convolutional neural network. Results. The PM-DL model achieved a sensitivity of 89.74% and a positive predictive value of 85.00% in the external test cohort for per-lesion analysis. No significant difference was observed in volumes (P = 0.13) and the largest sizes (P = 0.89) between manually delineated and segmented lesions. The DICE coefficient reached 0.77 ± 0.16. Similar performances were identified in the validation cohort. Moreover, the PM-DL model outperformed Liver Imaging Reporting and Data System (LI-RADS) in sensitivity (probable HCCs: LR-5 or LR-4, P = 0.18; definite HCCs: LR-5, P < 0.001), with a similar high specificity for per-patient analysis. Conclusion. The PM-DL model may be feasible for accurate automatic detection of sHCC in cirrhotic liver. [ABSTRACT FROM AUTHOR]

Quantitative susceptibility mapping (QSM) has demonstrated great potential in quantifying tissue susceptibility in various brain diseases. However, the intrinsic ill-posed inverse problem relating the tissue phase to the underlying susceptibility distribution affects the accuracy for quantifying tissue susceptibility. Recently, deep learning has shown promising results to improve accuracy by reducing the streaking artifacts. However, there exists a mismatch between the observed phase and the theoretical forward phase estimated by the susceptibility label. In this study, we proposed a model-based deep learning architecture that followed the STI (susceptibility tensor imaging) physical model, referred to as MoDL-QSM. Specifically, MoDL-QSM accounts for the relationship between STI-derived phase contrast induced by the susceptibility tensor terms (χ 13 , χ 23 and χ 33) and the acquired single-orientation phase. The convolutional neural networks are embedded into the physical model to learn a regularization term containing prior information. χ 33 and phase induced by χ 13 and χ 23 terms were used as the labels for network training. Quantitative evaluation metrics were compared with recently developed deep learning QSM methods. The results showed that MoDL-QSM achieved superior performance, demonstrating its potential for future applications. [Display omitted] [ABSTRACT FROM AUTHOR]

• A deep learning-based phase reconstruction scheme is proposed for high-resolution multi-shot (MSH) DWI reconstruction. • Higher SNR (especially with high acceleration rates), fewer aliasing artifacts, lower ghost-to-signal-ratio (GSR), higher tracked fiber counts, and finer fiber delineation can be obtained with MSH-DWI reconstruction compared to conventional MUSE. • Single-shot (SSH) DWIs were used for training, making the proposed method readily applicable for routine clinical exams. A phase correction method for high-resolution multi-shot (MSH) diffusion weighted imaging (DWI) is proposed. The efficacy and generalization capability of the method were validated on both healthy volunteers and patients. Conventionally, inter-shot phase variations for MSH echo-planar imaging (EPI) DWI are corrected by model-based algorithms. However, many acquisition imperfections are hard to measure accurately for conventional model-based methods, making the phase estimation and artifacts suppression unreliable. We propose a deep learning multiplexed sensitivity-encoding (DL-MUSE) framework to improve the phase estimations based on convolutional neural network (CNN) reconstruction. Aliasing-free single-shot (SSH) DW images, which have been used routinely in clinical settings, were used for training before the aliasing correction of MSH-DWI images. A dual-channel U-net comprising multiple convolutional layers was used for the phase estimation of MSH-EPI. The network was trained on a dataset containing 30 healthy volunteers and tested on another dataset of 52 healthy subjects and 15 patients with lesions or tumors with different shot numbers (4, 6 and 8). To further validate the generalization capability of our network, we acquired a dataset with different numbers of shots, TEs, partial Fourier factors, resolutions, ETLs, FOVs, coil numbers, and image orientations from two sites. We also compared the reconstruction performance of our proposed method with that of the conventional MUSE and SSH-EPI qualitatively and quantitatively. Our results show that DL-MUSE is capable of correcting inter-shot phase errors with high and robust performance. Compared to conventional model-based MUSE, our method, by applying deep learning-based phase corrections, showed reduced distortion, noise level, and signal loss in high b -value DWIs. The improvements of image quality become more evident as the shot number increases from 4 to 8, especially in those central regions of the images, where g -factor artifacts are severe. Furthermore, the proposed method could provide the information about the orientation of the white matter with better consistency and achieve finer fibers delineation compared to the SSH-EPI method. Besides, the experiments on volunteers and patients from two different sites demonstrated the generalizability of our proposed method preliminarily. A deep learning-based reconstruction algorithm for MSH-EPI images, which helps improve image quality greatly, was proposed. Results from healthy volunteers and tumor patients demonstrated the feasibility and generalization performances of our method for high-resolution MSH-EPI DWI, which can be used for routine clinical applications as well as neuroimaging research. [ABSTRACT FROM AUTHOR]

Deep learning has demonstrated superior performance in image reconstruction compared to most conventional iterative algorithms. However, their effectiveness and generalization capability are highly dependent on the sample size and diversity of the training data. Deep learning-based reconstruction requires multi-coil raw k-space data, which are not collected by routine scans. On the other hand, large amounts of magnitude images are readily available in hospitals. Hence, we proposed the MAGnitude Images to Complex K-space (MAGIC-K) Net to generate multi-coil k-space data from existing magnitude images and a limited number of required raw k-space data to facilitate the reconstruction. Compared to some basic data augmentation methods applying global intensity and displacement transformations to the source images, the MAGIC-K Net can generate more realistic intensity variations and displacements from pairs of anatomical Digital Imaging and Communications in Medicine (DICOM) images. The reconstruction performance was validated in 30 healthy volunteers and 6 patients with different types of tumors. The experimental results demonstrated that the high-resolution Diffusion Weighted Image (DWI) reconstruction benefited from the proposed augmentation method. The MAGIC-K Net enabled the deep learning network to reconstruct images with superior performance in both healthy and tumor patients, qualitatively and quantitatively. [ABSTRACT FROM AUTHOR]

Accurate grading of liver fibrosis can effectively assess the severity of liver disease and help doctors make an appropriate diagnosis. This study aimed to perform the automatic staging of hepatic fibrosis on patients with hepatitis B, who underwent gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid (Gd-EOB-DTPA)-enhanced magnetic resonance imaging with dynamic radiomics analysis. The proposed dynamic radiomics model combined imaging features from multi-phase dynamic contrast-enhanced (DCE) images and time-domain information. Imaging features were extracted from the deep learning-based segmented liver volume, and time-domain features were further explored to analyze the variation in features during contrast enhancement. Model construction and evaluation were based on a 132-case data set. The proposed model achieved remarkable performance in significant fibrosis (fibrosis stage S1 vs. S2–S4; accuracy (ACC) = 0.875, area under the curve (AUC) = 0.867), advanced fibrosis (S1–S2 vs. S3–S4; ACC = 0.825, AUC = 0.874), and cirrhosis (S1–S3 vs. S4; ACC = 0.850, AUC = 0.900) classifications in the test set. It was more dominant compared with the conventional single-phase or multi-phase DCE-based radiomics models, normalized liver enhancement, and some serological indicators. Time-domain features were found to play an important role in the classification models. The dynamic radiomics model can be applied for highly accurate automatic hepatic fibrosis staging. [ABSTRACT FROM AUTHOR]