Search

Purpose: In MRI, motion artifacts can significantly degrade image quality. Motion artifact correction methods using deep neural networks usually required extensive training on large datasets, making them time-consuming and resource-intensive. In this paper, an unsupervised deep learning-based motion artifact correction method for turbo-spin echo MRI is proposed using the deep image prior framework., Theory and Methods: The proposed approach takes advantage of the high impedance to motion artifacts offered by the neural network parameterization to remove motion artifacts in MR images. The framework consists of parameterization of MR image, automatic spatial transformation, and motion simulation model. The proposed method synthesizes motion-corrupted images from the motion-corrected images generated by the convolutional neural network, where an optimization process minimizes the objective function between the synthesized images and the acquired images., Results: In the simulation study of 280 slices from 14 subjects, the proposed method showed a significant increase in the averaged structural similarity index measure by 0.2737 in individual coil images and by 0.4550 in the root-sum-of-square images. In addition, the ablation study demonstrated the effectiveness of each proposed component in correcting motion artifacts compared to the corrected images produced by the baseline method. The experiments on real motion dataset has shown its clinical potential., Conclusion: The proposed method exhibited significant quantitative and qualitative improvements in correcting rigid and in-plane motion artifacts in MR images acquired using turbo spin-echo sequence., (© 2024 International Society for Magnetic Resonance in Medicine.)

Subject motion is a long-standing problem of magnetic resonance imaging (MRI), which can seriously deteriorate the image quality. Various prospective and retrospective methods have been proposed for MRI motion correction, among which deep learning approaches have achieved state-of-the-art motion correction performance. This survey paper aims to provide a comprehensive review of deep learning-based MRI motion correction methods. Neural networks used for motion artifacts reduction and motion estimation in the image domain or frequency domain are detailed. Furthermore, besides motion-corrected MRI reconstruction, how estimated motion is applied in other downstream tasks is briefly introduced, aiming to strengthen the interaction between different research areas. Finally, we identify current limitations and point out future directions of deep learning-based MRI motion correction., (© 2024. The Author(s), under exclusive licence to European Society for Magnetic Resonance in Medicine and Biology (ESMRMB).)

Background: Quantification of metabolites concentrations in institutional unit (IU) is important for inter-subject and long-term comparisons in the applications of magnetic resonance spectroscopy (MRS). Recently, deep learning (DL) algorithms have found a variety of applications on the process of MRS data. A quantification strategy compatible to DL base MRS spectral processing method is, therefore, useful., Materials and Methods: This study aims to investigate whether metabolite concentrations quantified using a convolutional neural network (CNN) based method, coupled with a scaling procedure that normalizes spectral signals for CNN input and linear regression, can effectively reflect variations in metabolite concentrations in IU across different brain regions with varying signal-to-noise ratios (SNR) and linewidths (LW). An error index based on standard error (SE) is proposed to indicate the confidence levels associated with metabolite predictions. In vivo MRS spectra were acquired from three brain regions of 43 subjects using a 3T system., Results: The metabolite concentrations in IU of five major metabolites, quantified using CNN and LCModel, exhibit similar ranges with Pearson's correlation coefficients ranging from 0.24 to 0.78. The SE of the metabolites shows a positive correlation with Cramer-Rao lower bound (CRLB) (r=0.46) and  absolute CRLB (r=0.81), calculated by multiplying CRLBs with the quantified metabolite content., Conclusion: In conclusion, the CNN based method with the proposed scaling procedures can be employed to quantify in vivo MRS spectra and derive metabolites concentrations in IU. The SE can be used as error index, indicating predicted uncertainties for metabolites and sharing information similar to the absolute CRLB., (© 2023. The Author(s), under exclusive licence to European Society for Magnetic Resonance in Medicine and Biology (ESMRMB).)

Purpose: Urolithiasis is a chronic condition that leads to repeated CT scans throughout the patient's life. The goal was to assess the diagnostic performance and image quality of submillisievert abdominopelvic computed tomography (CT) using deep learning-based image reconstruction (DLIR) in urolithiasis., Methods: 57 patients with suspected urolithiasis underwent both non-contrast low-dose (LD) and ULD abdominopelvic CT. Raw image data of ULD CT were reconstructed using hybrid iterative reconstruction (ASIR-V 70%) and high-strength-level DLIR (DLIR-H). The performance of ULD CT for the detection of urinary stones was assessed by two readers and compared with LD CT with ASIR-V 70% as a reference standard. Image quality was assessed subjectively and objectively., Results: 266 stones were detected in 38 patients. Mean effective dose was 0.59 mSv for ULD CT and 1.96 mSv for LD CT. For diagnostic performance, sensitivity and specificity were 89% and 94%, respectively, for ULDCT with DLIR-H. There was an almost perfect intra-observer concordance on ULD CT with DLIR-H versus LDCT with ASIR-V 70% (ICC = 0.90 and 0.90 for the two readers). Image noise was significantly lower and signal-to-noise ratio significantly higher with DLIR-H compared to ASIR-V 70%. Subjective image quality was also significantly better with ULDCT with DLIR-H., Conclusion: ULD CT with Deep Learning Image Reconstruction maintains a good diagnostic performance in urolithiasis, with better image quality than hybrid iterative reconstruction and a significant radiation dose reduction., (© 2024. The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.)

Elucidating unknown structures of proteins, such as metastable states, is critical in designing therapeutic agents. Protein structure exploration has been performed using advanced computational methods, especially molecular dynamics and Markov chain Monte Carlo simulations, which require untenably long calculation times and prior structural knowledge. Here, we developed an innovative method for protein structure determination called never give up protein exploration (NPEX) with deep reinforcement learning. The NPEX method leverages the soft actor-critic algorithm and the intrinsic reward system, effectively adding a bias potential without the need for prior knowledge. To demonstrate the method's effectiveness, we applied it to four models: a double well, a triple well, the alanine dipeptide, and the tryptophan cage. Compared with Markov chain Monte Carlo simulations, NPEX had markedly greater sampling efficiency. The significantly enhanced computational efficiency and lack of prior domain knowledge requirements of the NPEX method will revolutionize protein structure exploration., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Inc. All rights reserved.)

Background: Risk prediction plays a crucial role in planning for prevention, monitoring, and treatment. Electronic Health Records (EHRs) offer an expansive repository of temporal medical data encompassing both risk factors and outcome indicators essential for effective risk prediction. However, challenges emerge due to the lack of readily available gold-standard outcomes and the complex effects of various risk factors. Compounding these challenges are the false positives in diagnosis codes, and formidable task of pinpointing the onset timing in annotations., Objective: We develop a Semi-supervised Double Deep Learning Temporal Risk Prediction (SeDDLeR) algorithm based on extensive unlabeled longitudinal Electronic Health Records (EHR) data augmented by a limited set of gold standard labels on the binary status information indicating whether the clinical event of interest occurred during the follow-up period., Methods: The SeDDLeR algorithm calculates an individualized risk of developing future clinical events over time using each patient's baseline EHR features via the following steps: (1) construction of an initial EHR-derived surrogate as a proxy for the onset status; (2) deep learning calibration of the surrogate along gold-standard onset status; and (3) semi-supervised deep learning for risk prediction combining calibrated surrogates and gold-standard onset status. To account for missing onset time and heterogeneous follow-up, we introduce temporal kernel weighting. We devise a Gated Recurrent Units (GRUs) module to capture temporal characteristics. We subsequently assess our proposed SeDDLeR method in simulation studies and apply the method to the Massachusetts General Brigham (MGB) Biobank to predict type 2 diabetes (T2D) risk., Results: SeDDLeR outperforms benchmark risk prediction methods, including Semi-parametric Transformation Model (STM) and DeepHit, with consistently best accuracy across experiments. SeDDLeR achieved the best C-statistics ( 0.815, SE 0.023; vs STM +.084, SE 0.030, P-value .004; vs DeepHit +.055, SE 0.027, P-value .024) and best average time-specific AUC (0.778, SE 0.022; vs STM + 0.059, SE 0.039, P-value .067; vs DeepHit + 0.168, SE 0.032, P-value <0.001) in the MGB T2D study., Conclusion: SeDDLeR can train robust risk prediction models in both real-world EHR and synthetic datasets with minimal requirements of labeling event times. It holds the potential to be incorporated for future clinical trial recruitment or clinical decision-making., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Inc. All rights reserved.)

In the world, grapes are considered as the most significant fruit, and it comprises various nutrients, like Vitamin C and it is utilized to produce wines and raisins. The major six general grape leaf diseases and pests are brown spots, leaf blight, downy mildew, anthracnose, and black rot. However, the existing manual detection methods are time-consuming and require more efforts. In this paper, an effectual grape leaf disease finding and percentage of pesticide detection approach is devised usingan optimized deep learning scheme. Here, the input image is pre-processed and then, black spot segmentation is done using proposed Taylor Remora Optimization Procedure (TROA). The TROA is the combination of Taylor concept and Remora Optimization Algorithm (ROA). After that, the multi-classification of grape leaf disease is performed to classify the disease as Black rot, Black measles, Isariopsis leaf spot and healthy. Accordingly, the training process of the Deep Neuro-Fuzzy Optimizer (DNFN) is done using Sine Cosine Butterfly Optimization (SCBO). Then, pesticide classification is done using Deep Maxout Network (DMN) and the training of the DMN is done using the Monarch Anti Corona Optimization (MACO) algorithm. Finally, the pesticide percentage level detection is performed using Deep Belief Network (DBN), which is trained by the TROA. The devised scheme obtained highest accuracy of 0.9327, sensitivity of 0.9383, and 0.9429. Thus, this method can assist as an effectual decision provision system for assisting the farmers to find the percentage of pesticide affected in grape leaf diseases., (© 2023. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.)

Magnetic resonance imaging produces detailed anatomical and physiological images of the human body that can be used in the clinical diagnosis and treatment of diseases. However, MRI suffers its comparatively longer acquisition time than other imaging methods and is thus vulnerable to motion artifacts, which ultimately lead to likely failed or even wrong diagnosis. In order to perform faster reconstruction, deep learning-based methods along with traditional strategies such as parallel imaging and compressed sensing come into play in recent years in this field. Meanwhile, in order to better analyze the diseases, it is also often necessary to acquire images in the same region of interest under different modalities, which yield images with different contrast levels. However, most of these aforementioned methods tend to use single-modal images for reconstruction, neglecting the correlation and redundancy information embedded in MR images acquired with different modalities. While there are works on multi-modal reconstruction, the information is yet to be efficiently explored. In this paper, we propose an end-to-end neural network called MLMFNet, which helps the reconstruction of the target modality by using information from the auxiliary modality across feature channels and layers. Specifically, this is highlighted by three components: (I) An encoder based on UNet with a single-stream strategy that fuses auxiliary and target modalities; (II) a decoder that tends to multi-level features from all layers of the encoder, and (III) a channel attention module. Quantitative and qualitative analyses are performed on a public brain dataset and knee brain dataset, which show that the proposed method achieves satisfying results in MRI reconstruction within the multi-modal context, and also demonstrate its effectiveness and potential to be used in clinical practice., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023. Published by Elsevier Inc.)

High-grade serous ovarian cancer has a high degree of malignancy, and at detection, it is prone to infiltration of surrounding soft tissues, as well as metastasis to the peritoneum and lymph nodes, peritoneal seeding, and distant metastasis. Whether recurrence occurs becomes an important reference for surgical planning and treatment methods for this disease. Current recurrence prediction models do not consider the potential pathological relationships between internal tissues of the entire ovary. They use convolutional neural networks to extract local region features for judgment, but the accuracy is low, and the cost is high. To address this issue, this paper proposes a new lightweight deep learning algorithm model for predicting recurrence of high-grade serous ovarian cancer. The model first uses ghost convolution (Ghost Conv) and coordinate attention (CA) to establish ghost counter residual (SCblock) modules to extract local feature information from images. Then, it captures global information and integrates multi-level information through proposed layered fusion Transformer (STblock) modules to enhance interaction between different layers. The Transformer module unfolds the feature map to compute corresponding region blocks, then folds it back to reduce computational cost. Finally, each STblock module fuses deep and shallow layer depth information and incorporates patient's clinical metadata for recurrence prediction. Experimental results show that compared to the mainstream lightweight mobile visual Transformer (MobileViT) network, the proposed slicer visual Transformer (SlicerViT) network improves accuracy, precision, sensitivity, and F1 score, with only 1/6 of the computational cost and half the parameter count. This research confirms that the proposed algorithm model is more accurate and efficient in predicting recurrence of high-grade serous ovarian cancer. In the future, it can serve as an auxiliary diagnostic technique to improve patient survival rates and facilitate the application of the model in embedded devices.

Glaucoma infection is rapidly spreading globally and the number of glaucoma patients is expected to exceed 110 million by 2040. Early identification and detection of glaucoma is particularly important as it can easily lead to irreversible vision damage or even blindness if not treated with intervention in the early stages. Deep learning has attracted much attention in the field of computer vision and has been widely studied especially in the recognition and diagnosis of ophthalmic diseases. It is challenging to efficiently extract effective features for accurate grading of glaucoma in a limited dataset. Currently, in glaucoma recognition algorithms, 2D fundus images are mainly used to automatically identify the disease or not, but do not distinguish between early or late stages; however, in clinical practice, the treatment of early and late glaucoma is not the same, so it is more important to proceed to achieve accurate grading of glaucoma. This study uses a private dataset containing modal data, 2D fundus images, and 3D-OCT scanner images, to extract the effective features therein to achieve an accurate triple classification (normal, early, and moderately advanced) for optimal performance on various measures. In view of this, this paper proposes an automatic glaucoma classification method based on the attention mechanism and EfficientNetB3 network. The EfficientNetB3 network and ResNet34 network are built to extract and fuse 2D fundus images and 3D-OCT scanner images, respectively, to achieve accurate classification. The proposed auto-classification method minimizes feature redundancy while improving classification accuracy, and incorporates an attention mechanism in the two-branch model, which enables the convolutional neural network to focus its attention on the main features of the eye and discard the meaningless black background region in the image to improve the performance of the model. The auto-classification method combined with the cross-entropy function achieves the highest accuracy up to 97.83%. Since the proposed automatic grading method is effective and ensures reliable decision-making for glaucoma screening, it can be used as a second opinion tool by doctors, which can greatly reduce missed diagnosis and misdiagnosis by doctors, and buy more time for patient's treatment., Competing Interests: The authors have declared that no competing interests exist., (Copyright: © 2024 Zhang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.)

In this paper, a novel deep learning method Mamba-RAYOLO is presented, which can improve detection and classification in the processing and analysis of ECG images in real time by integrating three advanced modules. The feature extraction module in our work with a multi-branch structure during training can capture a wide range of features to ensure efficient inference and rich feature extraction. The attention mechanism module utilized in our proposed network can dynamically focus on the most relevant spatial and channel-wise features to improve detection accuracy and computational efficiency. Then, the extracted features can be refined for efficient spatial feature processing and robust feature fusion. Several sets of experiments have been carried out to test the validity of the proposed Mamba-RAYOLO and these indicate that our method has made significant improvements in the detection and classification of ECG images. The research offers a promising framework for more accurate and efficient medical ECG diagnostics.

Door access control systems are important to protect the security and integrity of physical spaces. Accuracy and speed are important factors that govern their performance. In this paper, we investigate a novel approach to identify users by measuring patterns of their interactions with a doorknob via an embedded accelerometer and gyroscope and by applying deep-learning-based algorithms to these measurements. Our identification results obtained from 47 users show an accuracy of 90.2%. When the sex of the user is used as an input feature, the accuracy is 89.8% in the case of male individuals and 97.0% in the case of female individuals. We study how the accuracy is affected by the sample duration, finding that is its possible to identify users using a sample of 0.5 s with an accuracy of 68.5%. Our results demonstrate the feasibility of using patterns of motor activity to provide access control, thus extending with it the set of alternatives to be considered for behavioral biometrics.

As a special type of multi-objective combinatorial optimization problems (MOCOPs), the multi-objective traveling salesman problem (MOTSP) plays an important role in practical fields such as transportation and robot control. However, due to the complexity of its solution space and the conflicts between different objectives, it is difficult to obtain satisfactory solutions in a short time. This paper proposes an end-to-end algorithm framework for solving MOTSP based on deep reinforcement learning (DRL). By decomposing strategies, solving MOTSP is transformed into solving multiple single-objective optimization subproblems. Through linear transformation, the features of the MOTSP are combined with the weights of the objective function. Subsequently, a modified graph pointer network (GPN) model is used to solve the decomposed subproblems. Compared with the previous DRL model, the proposed algorithm can solve all the subproblems using only one model without adding weight information as input features. Furthermore, our algorithm can output a corresponding solution for each weight, which increases the diversity of solutions. In order to verify the performance of our proposed algorithm, it is compared with four classical evolutionary algorithms and two DRL algorithms on several MOTSP instances. The comparison shows that our proposed algorithm outperforms the compared algorithms both in terms of training time and the quality of the resulting solutions., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Ltd. All rights reserved.)

Background: Volume of interest (VOI) segmentation is a crucial step for Radiomics analyses and radiotherapy (RT) treatment planning. Because it can be time-consuming and subject to inter-observer variability, we developed and tested a Deep Learning-based automatic segmentation (DLBAS) algorithm to reproducibly predict the primary gross tumor as VOI for Radiomics analyses in extremity soft tissue sarcomas (STS)., Methods: A DLBAS algorithm was trained on a cohort of 157 patients and externally tested on an independent cohort of 87 patients using contrast-enhanced MRI. Manual tumor delineations by a radiation oncologist served as ground truths (GTs). A benchmark study with 20 cases from the test cohort compared the DLBAS predictions against manual VOI segmentations of two residents (ERs) and clinical delineations of two radiation oncologists (ROs). The ROs rated DLBAS predictions regarding their direct applicability., Results: The DLBAS achieved a median dice similarity coefficient (DSC) of 0.88 against the GTs in the entire test cohort (interquartile range (IQR): 0.11) and a median DSC of 0.89 (IQR 0.07) and 0.82 (IQR 0.10) in comparison to ERs and ROs, respectively. Radiomics feature stability was high with a median intraclass correlation coefficient of 0.97, 0.95 and 0.94 for GTs, ERs, and ROs, respectively. DLBAS predictions were deemed clinically suitable by the two ROs in 35% and 20% of cases, respectively., Conclusion: The results demonstrate that the DLBAS algorithm provides reproducible VOI predictions for radiomics feature extraction. Variability remains regarding direct clinical applicability of predictions for RT treatment planning., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 The Authors. Published by Elsevier B.V. All rights reserved.)

Purpose: As agricultural technology continues to develop, the scale of planting and production of date fruit is increasing, which brings higher yields. However, the increasing yields also put a lot of pressure on the classification step afterward. Image recognition based on deep learning algorithms can help to identify and classify the date fruit species, even in natural light., Method: In this paper, a deep fusion model based on whale optimization and an artificial neural network for Arabian date classification is proposed. The dataset used in this study includes five classes of date fruit images (Barhi, Khalas, Meneifi, Naboot Saif, Sullaj). The process of designing each model can be divided into three phases. The first phase is feature extraction. The second phase is feature selection. The third phase is the training and testing phase. Finally, the best-performing model was selected and compared with the currently established models (Alexnet, Squeezenet, Googlenet, Resnet50)., Results: The experimental results show that, after trying different combinations of optimization algorithms and classifiers, the highest test accuracy achieved by DeepDate was 95.9%. It takes less time to achieve a balance between classification accuracy and time consumption. In addition, the performance of DeepDate is better than that of many deep transfer learning models such as Alexnet, Squeezenet, Googlenet, VGG-19, NasNet, and Inception-V3., Conclusion: The proposed DeepDate improves the accuracy and efficiency of classifying date fruits and achieves better results in classification metrics such as accuracy and F1. DeepDate provides a promising classification solution for date fruit classification with higher accuracy. To further advance the industry, it is recommended that stakeholders invest in technology transfer programs to bring advanced image recognition and AI tools to smaller producers, enhancing sustainability and productivity across the sector. Collaborations between agricultural technologists and growers could also foster more tailored solutions that address specific regional challenges in date fruit production., Competing Interests: NO authors have competing interests., (Copyright: © 2024 Khalifa et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.)

As part of the field of DNA methylation identification, this study tackles the challenge of enhancing recognition performance by introducing a specialized deep learning framework called DeepPGD. DNA methylation, a crucial biological modification, plays a vital role in gene expression analyses, cellular differentiation, and the study of disease progression. However, accurately and efficiently identifying DNA methylation sites remains a pivotal concern in the field of bioinformatics. The issue addressed in this paper is the presence of methylation in DNA, which is a binary classification problem. To address this, our research aimed to develop a deep learning algorithm capable of more precisely identifying these sites. The DeepPGD framework combined a dual residual structure involving Temporal convolutional networks (TCNs) and bidirectional long short-term memory (BiLSTM) networks to effectively extract intricate DNA structural and sequence features. Additionally, to meet the practical requirements of DNA methylation identification, extensive experiments were conducted across a variety of biological species. The experimental results highlighted DeepPGD's exceptional performance across multiple evaluation metrics, including accuracy, Matthews' correlation coefficient (MCC), and the area under the curve (AUC). In comparison to other algorithms in the same domain, DeepPGD demonstrated superior classification and predictive capabilities across various biological species datasets. This significant advancement in algorithmic prowess not only offers substantial technical support, but also holds potential for research and practical implementation within the DNA methylation identification domain. Moreover, the DeepPGD framework shows potential for application in genomics research, biomedicine, and disease diagnostics, among other fields.

Facial emotion recognition (FER) is largely utilized to analyze human emotion in order to address the needs of many real-time applications such as computer-human interfaces, emotion detection, forensics, biometrics, and human-robot collaboration. Nonetheless, existing methods are mostly unable to offer correct predictions with a minimum error rate. In this paper, an innovative facial emotion recognition framework, termed extended walrus-based deep learning with Botox feature selection network (EWDL-BFSN), was designed to accurately detect facial emotions. The main goals of the EWDL-BFSN are to identify facial emotions automatically and effectively by choosing the optimal features and adjusting the hyperparameters of the classifier. The gradient wavelet anisotropic filter (GWAF) can be used for image pre-processing in the EWDL-BFSN model. Additionally, SqueezeNet is used to extract significant features. The improved Botox optimization algorithm (IBoA) is then used to choose the best features. Lastly, FER and classification are accomplished through the use of an enhanced optimization-based kernel residual 50 (EK-ResNet50) network. Meanwhile, a nature-inspired metaheuristic, walrus optimization algorithm (WOA) is utilized to pick the hyperparameters of EK-ResNet50 network model. The EWDL-BFSN model was trained and tested with publicly available CK+ and FER-2013 datasets. The Python platform was applied for implementation, and various performance metrics such as accuracy, sensitivity, specificity, and F1-score were analyzed with state-of-the-art methods. The proposed EWDL-BFSN model acquired an overall accuracy of 99.37 and 99.25% for both CK+ and FER-2013 datasets and proved its superiority in predicting facial emotions over state-of-the-art methods.

The hippocampus is a critical component of the brain and is associated with many neurological disorders. It can be further subdivided into several subfields, and accurate segmentation of these subfields is of great significance for diagnosis and research. However, the structures of hippocampal subfields are irregular and have complex boundaries, and their voxel values are close to surrounding brain tissues, making the segmentation task highly challenging. Currently, many automatic segmentation tools exist for hippocampal subfield segmentation, but they suffer from high time costs and low segmentation accuracy. In this paper, we propose a new dual-branch segmentation network structure (DSnet) based on deep learning for hippocampal subfield segmentation. While traditional convolutional neural network-based methods are effective in capturing hierarchical structures, they struggle to establish long-term dependencies. The DSnet integrates the Transformer architecture and a hybrid attention mechanism, enhancing the network's global perceptual capabilities. Moreover, the dual-branch structure of DSnet leverages the segmentation results of the hippocampal region to facilitate the segmentation of its subfields. We validate the efficacy of our algorithm on the public Kulaga-Yoskovitz dataset. Experimental results indicate that our method is more effective in segmenting hippocampal subfields than conventional single-branch network structures. Compared to the classic 3D U-Net, our proposed DSnet improves the average Dice accuracy of hippocampal subfield segmentation by 0.57%., (© 2024. The Author(s).)

Purpose: The accurate age estimation of cadavers is essential for their identification. However, conventional methods fail to yield adequate age estimation especially in elderly cadavers. We developed a deep learning algorithm for age estimation on CT images of the vertebral column and checked its accuracy., Method: For the development of our deep learning algorithm, we included 1,120 CT data of the vertebral column of 140 patients for each of 8 age decades. The deep learning model of regression analysis based on Visual Geometry Group-16 (VGG16) was improved in its estimation accuracy by bagging. To verify its accuracy, we applied our deep learning algorithm to estimate the age of 219 cadavers who had undergone postmortem CT (PMCT). The mean difference and the mean absolute error (MAE), the standard error of the estimate (SEE) between the known- and the estimated age, were calculated. Correlation analysis using the intraclass correlation coefficient (ICC) and Bland-Altman analysis were performed to assess differences between the known- and the estimated age., Results: For the 219 cadavers, the mean difference between the known- and the estimated age was 0.30 years; it was 4.36 years for the MAE, and 5.48 years for the SEE. The ICC (2,1) was 0.96 (95 % confidence interval: 0.95-0.97, p < 0.001). Bland-Altman analysis showed that there were no proportional or fixed errors (p = 0.08 and 0.41)., Conclusions: Our deep learning algorithm for estimating the age of 219 cadavers on CT images of the vertebral column was more accurate than conventional methods and highly useful., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier B.V. All rights reserved.)

Purpose: The objective of this study was to develop a deep learning model, using the ConvNeXt algorithm, that can effectively differentiate between ovarian endometriosis cysts (OEC) and benign mucinous cystadenomas (MC) by analyzing ultrasound images. The performance of the model in the diagnostic differentiation of these two conditions was also evaluated., Methods: A retrospective analysis was conducted on OEC and MC patients who had sought medical attention at the Fourth Affiliated Hospital of Harbin Medical University between August 2018 and May 2023. The diagnosis was established based on postoperative pathology or the characteristics of aspirated fluid guided by ultrasound, serving as the gold standard. Ultrasound images were collected and subjected to screening and preprocessing procedures. The data set was randomly divided into training, validation, and testing sets in a ratio of 5:3:2. Transfer learning was utilized to determine the initial weights of the ConvNeXt deep learning algorithm, which were further adjusted by retraining the algorithm using the training and validation ultrasound images to establish a new deep learning model. The weights that yielded the highest accuracy were selected to evaluate the diagnostic performance of the model using the validation set. Receiver operating characteristic (ROC) curves were generated, and the area under the curve (AUC) was calculated. Additionally, sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, negative likelihood ratio, and odds ratio were calculated. Decision curve analysis (DCA) curves were plotted., Results: The study included 786 ultrasound images from 184 patients diagnosed with either OEC or MC. The deep learning model achieved an AUC of 0.90 (95 % CI: 0.85-0.95) in accurately distinguishing between the two conditions, with a sensitivity of 90 % (95 % CI: 84 %-95 %), specificity of 90 % (95 % CI: 77 %-97 %), a positive predictive value of 96 % (95 % CI: 91 %-99 %), a negative predictive value of 77 % (95 % CI: 63 %-88 %), a positive likelihood ratio of 9.27 (95 % CI: 3.65-23.56), and a negative likelihood ratio of 0.11 (95 % CI: 0.06-0.19). The DCA curve demonstrated the practical clinical utility of the model., Conclusions: The deep learning model developed using the ConvNeXt algorithm exhibits high accuracy (90 %) in distinguishing between OEC and MC. This model demonstrates excellent diagnostic performance and clinical utility, providing a novel approach for the clinical differentiation of these two conditions., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier B.V. All rights reserved.)

Chronic ankle instability (CAI) is a major public health concern and adversely affects people's mobility and quality of life. Traditional assessment methods are subjective and qualitative by means of clinician observation and patient self-reporting, which may lead to inaccurate assessment and reduce the effectiveness of treatment in clinical practice. Gait analysis becomes a commonly used approach for monitoring human motion behaviors, which can be applied to specific diagnosis and assessment of CAI. However, it is still challenging to recognize the pathological gait pattern for CAI subjects. In this paper, we propose an integrated deep learning framework to solve the CAI recognition problem using kinematic data. Specifically, inspired by the biomechanics of human body system, we create a simple graph neural network (GNN), termed GaitNet, that operates on a spatial domain and exploits interactions among 3-D joint coordinates. We also develop an attention reinforcement learning (ARL) model that determines attention weights of frames on a temporal domain, which is combined with GaitNet for prediction. The effectiveness of our method is validated on the kinematic NEU-CAI dataset which is collected in our institution using a stereophotogrammetric system. According to extensive experiments, we demonstrate that the selected key phases (i.e., sequences of frames with high attentions) significantly increase the predictability of the proposed biomechanics-based GNN model to differentiate between CAI cohort and control cohort. Moreover, we show a significant prediction accuracy improvement (20%-25%) by our approach compared to state-of-the-art machine learning and deep learning methods.

The primary challenges in tea production under multiple stress exposures have negatively affected its global market sustainability, so introducing an infield fast technique for monitoring tea leaves' stresses has tremendous urgent needs. Therefore, this study aimed to propose an efficient method for the detection of stress symptoms based on a portable smartphone with deep learning models. Firstly, a database containing over 10,000 images of tea garden canopies in complex natural scenes was developed, which included healthy (no stress) and three types of stress (tea anthracnose (TA), tea blister blight (TB) and sunburn (SB)). Then, YOLOv5m and YOLOv8m algorithms were adapted to discriminate the four types of stress symptoms; where the YOLOv8m algorithm achieved better performance in the identification of healthy leaves (98%), TA (92.0%), TB (68.4%) and SB (75.5%). Furthermore, the YOLOv8m algorithm was used to construct a model for differentiation of disease severity of TA, and a satisfactory result was obtained with the accuracy of mild, moderate, and severe TA infections were 94%, 96%, and 91%, respectively. Besides, we found that CNN kernels of YOLOv8m could efficiently extract the texture characteristics of the images at layer 2, and these characteristics can clearly distinguish different types of stress symptoms. This makes great contributions to the YOLOv8m model to achieve high-precision differentiation of four types of stress symptoms. In conclusion, our study provided an effective system to achieve low-cost, high-precision, fast, and infield diagnosis of tea stress symptoms in complex natural scenes based on smartphone and deep learning algorithms., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Masson SAS. All rights reserved.)

Parkinson's disease is a common mental disease in the world, especially in the middle-aged and elderly groups. Today, clinical diagnosis is the main diagnostic method of Parkinson's disease, but the diagnosis results are not ideal, especially in the early stage of the disease. In this paper, a Parkinson's auxiliary diagnosis algorithm based on a hyperparameter optimization method of deep learning is proposed for the Parkinson's diagnosis. The diagnosis system uses ResNet50 to achieve feature extraction and Parkinson's classification, mainly including speech signal processing part, algorithm improvement part based on Artificial Bee Colony algorithm (ABC) and optimizing the hyperparameters of ResNet50 part. The improved algorithm is called Gbest Dimension Artificial Bee Colony algorithm (GDABC), proposing "Range pruning strategy" which aims at narrowing the scope of search and "Dimension adjustment strategy" which is to adjust gbest dimension by dimension. The accuracy of the diagnosis system in the verification set of Mobile Device Voice Recordings at King's College London (MDVR-CKL) dataset can reach more than 96%. Compared with current Parkinson's sound diagnosis methods and other optimization algorithms, our auxiliary diagnosis system shows better classification performance on the dataset within limited time and resources.

Applying affective computing techniques to recognize fear and combining them with portable signal monitors makes it possible to create real-time detection systems that could act as bodyguards when users are in danger. With this aim, this paper presents a fear recognition method based on physiological signals obtained from wearable devices. The procedure involves creating two-dimensional feature maps from the raw signals, using data augmentation and feature selection algorithms, followed by deep learning-based classification models, taking inspiration from those used in image processing. This proposal has been validated with two different datasets, achieving, in WEMAC, WESAD 3-classes, and WESAD 2-classes, F1-score results of 78.13%, 88.07%, and 99.60%, respectively, and 79.90%, 89.12%, and 99.60% in accuracy. Furthermore, the paper demonstrates the feasibility of implementing the proposed method on the Coral Edge TPU device, prepared to make inferences on the edge.

Over the years the number of research publications per year is growing exponentially. Finding research papers of quality from the massive literature of relevant articles is a challenging and time-consuming task. The approaches in the latest literature address citation recommendation by utilizing large bibliographic information and use machine learning and deep learning methods for the task. These techniques clearly require a large amount of training data as well as machines with high processing power. To overcome these issues, we propose a novel method by modifying the popular hyperlink induced topic search (HITS), a web page ranking algorithm, as citation recommendation using hyperlink induced topic search (CR-HITS) that works on a directed and weighted heterogeneous bibliographic network containing diverse types of nodes and edges. We define effective scoring schemes for nodes and edges based on basic bibliographic information like citations of papers, number of publications of an author, etc. Given a few seed papers, the citation recommendation algorithm CR-HITS is run on small neighborhoods of the seed papers and hence the time taken by the execution is very small to yield the final recommendations. To the best of our knowledge, HITS has been used for the first time for the citation recommendation problem. We perform extensive experimentation on DBLP (version-11) and ACM (version-9) datasets and compare the results with many baseline methods in terms of MAP, MRR, and recall@N measures. The performance of the proposed algorithms is superior with respect to the MAP metric and matches the second best for the other two metrics. Since the top two algorithms use deep learning methods and use much larger bibliographic information including abstracts of the papers, we claim that our approach utilizes very low resources, yet yields recommendations that are very close to the top recommendations. [ABSTRACT FROM AUTHOR]

Deep learning, a subset of machine learning that utilizes neural networks, has seen significant advancements in recent years. These advancements have led to breakthroughs in a wide range of fields, from natural language processing to computer vision, and have the potential to revolutionize many industries or organizations. They have also demonstrated exceptional performance in the identification and mapping of seagrass images. However, these deep learning models, particularly the popular Convolutional Neural Networks (CNNs) require architectural engineering and hyperparameter tuning. This paper proposes a Deep Neuroevolutionary (DNE) model that can automate the architectural engineering and hyperparameter tuning of CNNs models by developing and using a novel metaheuristic algorithm, named 'Boosted Atomic Orbital Search (BAOS)'. The proposed BAOS is an improved version of the recently proposed Atomic Orbital Search (AOS) algorithm which is based on the principle of atomic model and quantum mechanics. The proposed algorithm leverages the power of the Lévy flight technique to boost the performance of the AOS algorithm. The proposed DNE algorithm (BAOS-CNN) is trained, evaluated and compared with six popular optimisation algorithms on a patch-based multi-species seagrass dataset. This proposed BAOS-CNN model achieves the highest overall accuracy (97.48%) among the seven evolutionary-based CNN models. The proposed model also achieves the state-of-the-art overall accuracy of 92.30% and 93.5% on the publicly available four classes and five classes version of the 'DeepSeagrass' dataset, respectively. This multi-species seagrass dataset is available at: https://ro.ecu.edu.au/datasets/141/., Competing Interests: The authors have declared that no competing interests exist., (Copyright: © 2024 Noman et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.)

Physical aggression is a serious and widespread problem in society, affecting people worldwide. It impacts nearly every aspect of life. While some studies explore the root causes of violent behavior, others focus on urban planning in high-crime areas. Real-time violence detection, powered by artificial intelligence, offers a direct and efficient solution, reducing the need for extensive human supervision and saving lives. This paper is a continuation of a systematic mapping study and its objective is to provide a comprehensive and up-to-date review of AI-based video violence detection, specifically in physical assaults. Regarding violence detection, the following have been grouped and categorized from the review of the selected papers: 21 challenges that remain to be solved, 28 datasets that have been created in recent years, 21 keyframe extraction methods, 16 types of algorithm inputs, as well as a wide variety of algorithm combinations and their corresponding accuracy results. Given the lack of recent reviews dealing with the detection of violence in video, this study is considered necessary and relevant.

Accurate delineation of key waveforms in an ECG is a critical step in extracting relevant features to support the diagnosis and treatment of heart conditions. Although deep learning based methods using segmentation models to locate P, QRS, and T waves have shown promising results, their ability to handle arrhythmias has not been studied in any detail. In this paper we investigate the effect of arrhythmias on delineation quality and develop strategies to improve performance in such cases. We introduce a U-Net-like segmentation model for ECG delineation with a particular focus on diverse arrhythmias. This is followed by a post-processing algorithm which removes noise and automatically determines the boundaries of P, QRS, and T waves. Our model has been trained on a diverse dataset and evaluated against the LUDB and QTDB datasets to show strong performance, with F1-scores exceeding 99% for QRS and T waves, and over 97% for P waves in the LUDB dataset. Furthermore, we assess various models across a wide array of arrhythmias and observe that models with a strong performance on standard benchmarks may still perform poorly on arrhythmias that are underrepresented in these benchmarks, such as tachycardias. We propose solutions to address this discrepancy., Competing Interests: I have read the journal’s policy and the authors of this manuscript have the following competing interests: CJ, TP, WK and OvK received financial support through NRF grants 2022R1A5A6000840 as well as RS-2022-00165404, NRF2023005562, funded by the Korean Government. SHK, SYO, JHJ, JSH, WJK and MJC received financial support through grants 1711174270, RS-2021-KD000008 funded by the Korean government. In addition, SHK, SYO, JHJ, JSH, WJK and MJC are Stockholders of Medifarmsoft Co., Ltd., (Copyright: © 2024 Joung et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.)

Deep learning models have demonstrated remarkable performance in the classification of motor imagery BCI systems. However, these models exhibit sensitivity to challenging trials, often called hard trials, leading to performance degradation. In this paper, we address this issue by proposing two novel methods for identifying and mitigating the impact of hard trials on model performance. The first method leverages model prediction scores to discern hard trials. The second approach employs a quantitative explainable artificial intelligence (XAI) approach, enabling a more transparent and interpretable means of hard trial identification. The identified hard trials are removed from the entire motor imagery training and validation dataset, and the deep learning model is further re-trained using the dataset without hard trials. To evaluate the efficacy of these proposed methods, experiments were conducted on the Open BMI dataset. The results for hold-out analysis show that the proposed quantitative XAI- based hard trial removal method has statistically improved the average classification accuracy of the baseline deep CNN model from 63.77% to 68.70%, with p -value =7.66e-11for the subject-specific MI classification. Additionally, analyzing the scalp map representing the average relevance scores of correctly classified trials compared to a misclassified trial provides a deeper insight into identifying hard trials. The results indicate that the proposed quantitaive-based XAI approach outperformes the prediction-score-based approach in hard trial identification., (© 2024 IOP Publishing Ltd.)

In recent years, fatigue driving has become the main cause of traffic accidents, leading to increased attention towards fatigue detection systems. However, the pooling and strided convolutional operations in fatigue detection algorithm based on traditional deep learning methods may led to the loss of some useful information. This paper proposed a novel τ -shaped convolutional network ( τ Net ) aiming to address this issue. Unlike traditional network structures, τ Net incorporates the operations of upsampling features and concatenating high- and low-level features, enabling full utilization of useful information. Moreover, considering that the fatigue state is a mental state involving temporal evolution, we proposed the novel long short-term memory (LSTM)- τ -shaped convolutional network (LSTM- τ Net ), a parallel structure composed of LSTM and τ Net for fatigue detection, where τ Net extracts time-invariant features with location information, and LSTM extracts long temporal dependencies. We compared LSTM- τ Net with six competing methods based on two datasets. Results showed that the proposed algorithm achieved higher classification accuracy than the other methods, with 94.25% on EEG data (binary classification) and 82.19% on EOG data (triple classification). Additionally, the proposed algorithm exhibits low computational cost, good training stability, and robustness against insufficient training. Therefore, it is promising for further implementation of fatigue online detection systems., (© 2024. International Federation for Medical and Biological Engineering.)

Magnetic Resonance Imaging (MRI) reconstruction has made significant progress with the introduction of Deep Learning (DL) technology combined with Compressed Sensing (CS). However, most existing methods require large fully sampled training datasets to supervise the training process, which may be unavailable in many applications. Current unsupervised models also show limitations in performance or speed and may face unaligned distributions during testing. This paper proposes an unsupervised method to train competitive reconstruction models that can generate high-quality samples in an end-to-end style. Firstly teacher models are trained by filling the re-undersampled images and compared with the undersampled images in a self-supervised manner. The teacher models are then distilled to train another cascade model that can leverage the entire undersampled k-space during its training and testing. Additionally, we propose an adaptive distillation method to re-weight the samples based on the variance of teachers, which represents the confidence of the reconstruction results, to improve the quality of distillation. Experimental results on multiple datasets demonstrate that our method significantly accelerates the inference process while preserving or even improving the performance compared to the teacher model. In our tests, the distilled models show 5%-10% improvements in PSNR and SSIM compared with no distillation and are 10 times faster than the teacher.

Deep learning models often need sufficient supervision (i.e., labelled data) in order to be trained effectively. By contrast, humans can swiftly learn to identify important anatomy in medical images like MRI and CT scans, with minimal guidance. This recognition capability easily generalises to new images from different medical facilities and to new tasks in different settings. This rapid and generalisable learning ability is largely due to the compositional structure of image patterns in the human brain, which are not well represented in current medical models. In this paper, we study the utilisation of compositionality in learning more interpretable and generalisable representations for medical image segmentation. Overall, we propose that the underlying generative factors that are used to generate the medical images satisfy compositional equivariance property, where each factor is compositional (e.g., corresponds to human anatomy) and also equivariant to the task. Hence, a good representation that approximates well the ground truth factor has to be compositionally equivariant. By modelling the compositional representations with learnable von-Mises-Fisher (vMF) kernels, we explore how different design and learning biases can be used to enforce the representations to be more compositionally equivariant under un-, weakly-, and semi-supervised settings. Extensive results show that our methods achieve the best performance over several strong baselines on the task of semi-supervised domain-generalised medical image segmentation. Code will be made publicly available upon acceptance at https://github.com/vios-s.

Background: Metastatic femoral tumors may lead to pathological fractures during daily activities. A CT-based finite element analysis of a patient's femurs was shown to assist orthopedic surgeons in making informed decisions about the risk of fracture and the need for a prophylactic fixation. Improving the accuracy of such analyses ruqires an automatic and accurate segmentation of the tumors and their automatic inclusion in the finite element model. We present herein a deep learning algorithm (nnU-Net) to automatically segment lytic tumors within the femur., Method: A dataset consisting of fifty CT scans of patients with manually annotated femoral tumors was created. Forty of them, chosen randomly, were used for training the nnU-Net, while the remaining ten CT scans were used for testing. The deep learning model's performance was compared to two experienced radiologists., Findings: The proposed algorithm outperformed the current state-of-the-art solutions, achieving dice similarity scores of 0.67 and 0.68 on the test data when compared to two experienced radiologists, while the dice similarity score for inter-individual variability between the radiologists was 0.73., Interpretation: The automatic algorithm may segment lytic femoral tumors in CT scans as accurately as experienced radiologists with similar dice similarity scores. The influence of the realistic tumors inclusion in an autonomous finite element algorithm is presented in (Rachmil et al., "The Influence of Femoral Lytic Tumors Segmentation on Autonomous Finite Element Analyses", Clinical Biomechanics, 112, paper 106192, (2024))., Competing Interests: Declaration of competing interest None., (Copyright © 2024 Elsevier Ltd. All rights reserved.)

Snakebite poses a significant health threat in numerous tropical and subtropical nations, with around 5.4 million cases reported annually, which results in 1.8-2.7 million instances of envenomation, underscoring its critical impact on public health. The 'BIG FOUR' group comprises the primary committers responsible for most snake bites in India. Effective management of snakebite victims is essential for prognosis, emphasizing the need for preventive measures to limit snakebite-related deaths. The proposed initiative seeks to develop a transfer learning-based image classification algorithm using DenseNet to identify venomous and non-venomous snakes automatically. The study comprehensively evaluates the image classification results, employing accuracy, F1-score, Recall, and Precision metrics. DenseNet emerges as a potent tool for multiclass snake image classification, achieving a notable accuracy rate of 86%. The proposed algorithm intends to be incorporated into an AI-based snake-trapping device with artificial prey made with tungsten wire and vibration motors to mimic heat and vibration signatures, enhancing its appeal to snakes. The proposed algorithm in this research holds promise as a primary tool for preventing snake bites globally, offering a path toward automated snake capture without human intervention. These findings are significant in preventing snake bites and advancing snakebite mitigation strategies., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Ltd. All rights reserved.)

Thyroid cancer incidences endure to increase even though a large number of inspection tools have been developed recently. Since there is no standard and certain procedure to follow for the thyroid cancer diagnoses, clinicians require conducting various tests. This scrutiny process yields multi-dimensional big data and lack of a common approach leads to randomly distributed missing (sparse) data, which are both formidable challenges for the machine learning algorithms. This paper aims to develop an accurate and computationally efficient deep learning algorithm to diagnose the thyroid cancer. In this respect, randomly distributed missing data stemmed singularity in learning problems is treated and dimensionality reduction with inner and target similarity approaches are developed to select the most informative input datasets. In addition, size reduction with the hierarchical clustering algorithm is performed to eliminate the considerably similar data samples. Four machine learning algorithms are trained and also tested with the unseen data to validate their generalization and robustness abilities. The results yield 100% training and 83% testing preciseness for the unseen data. Computational time efficiencies of the algorithms are also examined under the equal conditions., (© The Author(s) 2024. Published by Oxford University Press.)

The identification research of hydrogenation catalyst information has always been one of the most important businesses in the chemical industry. In order to aid researchers in efficiently screening high-performance catalyst carriers and tackle the pressing challenge at hand, it is imperative to find a solution for the intelligent recognition of hydrogenation catalyst images. To address the issue of low recognition accuracy caused by adhesion and stacking of hydrogenation catalysts, An image recognition algorithm of hydrogenation catalyst based on FPNC Net was proposed in this paper. In the present study, Resnet50 backbone network was used to extract the features, and spatially-separable convolution kernel was used to extract the multi-scale features of catalyst fringe. In addition, to effectively segment the adhesive regions of stripes, FPN (Feature Pyramid Network) is added to the backbone network for deep and shallow feature fusion. Introducing an attention module to adaptively adjust weights can effectively highlight the target features of the catalyst. The experimental results showed that the FPNC Net model achieved an accuracy of 94.2% and an AP value improvement of 19.37% compared to the original CenterNet model. The improved model demonstrates a significant enhancement in detection accuracy, indicating a high capability for detecting hydrogenation catalyst targets., Competing Interests: The authors have declared that no competing interests exist., (Copyright: © 2024 Hou et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.)

Ultrasound-guided percutaneous interventions have numerous advantages over traditional techniques. Accurate needle placement in the target anatomy is crucial for successful intervention, and reliable visual information is essential to achieve this. However, previous studies have revealed several challenges, such as the variability in needle echogenicity and the common misalignment of the ultrasound beam and the needle. Advanced techniques have been developed to optimize needle visualization, including hardware-based and image-processing-based methods. This paper proposes a novel strategy of integrating ultrasound-based deep learning approaches into an optical navigation system to enhance needle visualization and improve tip positioning accuracy. Both the tracking and detection algorithms are optimized utilizing optical tracking information. The information is introduced into the tracking network to define the search patch update strategy and form a trajectory reference to correct tracking results. In the detection network, the original image is processed according to the needle insertion position and current position given by the optical localization system to locate a coarse region, and the depth-score criterion is adopted to optimize detection results. Extensive experiments demonstrate that our approach achieves promising tip tracking and detection performance with tip localization errors of 1.11 ± 0.59 mm and 1.17 ± 0.70 mm, respectively. Moreover, we establish a paired dataset consisting of ultrasound images and their corresponding spatial tip coordinates acquired from the optical tracking system and conduct real puncture experiments to verify the effectiveness of the proposed methods. Our approach significantly improves needle visualization and provides physicians with visual guidance for posture adjustment.

In computer-aided medical diagnosis, obtaining labeled medical image data is expensive, while there is a high demand for model interpretability. However, most deep learning models currently require a large amount of data and lack interpretability. To address these challenges, this paper proposes a novel data augmentation method for medical image segmentation. The uniqueness and advantages of this method lie in the utilization of gradient-weighted class activation mapping to extract data efficient features, which are then fused with the original image. Subsequently, a new channel weight feature extractor is constructed to learn the weights between different channels. This approach achieves non-destructive data augmentation effects, enhancing the model's performance, data efficiency, and interpretability. Applying the method of this paper to the Hyper-Kvasir dataset, the intersection over union (IoU) and Dice of the U-net were improved, respectively; and on the ISIC-Archive dataset, the IoU and Dice of the DeepLabV3+ were also improved respectively. Furthermore, even when the training data is reduced to 70 %, the proposed method can still achieve performance that is 95 % of that achieved with the entire dataset, indicating its good data efficiency. Moreover, the data-efficient features used in the method have interpretable information built-in, which enhances the interpretability of the model. The method has excellent universality, is plug-and-play, applicable to various segmentation methods, and does not require modification of the network structure, thus it is easy to integrate into existing medical image segmentation method, enhancing the convenience of future research and applications.

Glaucoma is a chronic neurodegenerative disease that can result in irreversible vision loss if not treated in its early stages. The cup-to-disc ratio is a key criterion for glaucoma screening and diagnosis, and it is determined by dividing the area of the optic cup (OC) by that of the optic disc (OD) in fundus images. Consequently, the automatic and accurate segmentation of the OC and OD is a pivotal step in glaucoma detection. In recent years, numerous methods have resulted in great success on this task. However, most existing methods either have unsatisfactory segmentation accuracy or high time costs. In this paper, we propose a lightweight deep-learning architecture for the simultaneous segmentation of the OC and OD, where we have adopted fuzzy learning and a multi-layer perceptron to simplify the learning complexity and improve segmentation accuracy. Experimental results demonstrate the superiority of our proposed method as compared to most state-of-the-art approaches in terms of both training time and segmentation accuracy.

Training an accurate classifier for EEG-based brain-computer interface (BCI) requires EEG data from a large number of users, whereas protecting their data privacy is a critical consideration. Federated learning (FL) is a promising solution to this challenge. This paper proposes Federated classification with local Batch-specific batch normalization and Sharpness-aware minimization (FedBS) for privacy protection in EEG-based motor imagery (MI) classification. FedBS utilizes local batch-specific batch normalization to reduce data discrepancies among different clients, and sharpness-aware minimization optimizer in local training to improve model generalization. Experiments on three public MI datasets using three popular deep learning models demonstrated that FedBS outperformed six state-of-the-art FL approaches. Remarkably, it also outperformed centralized training, which does not consider privacy protection at all. In summary, FedBS protects user EEG data privacy, enabling multiple BCI users to participate in large-scale machine learning model training, which in turn improves the BCI decoding accuracy.

Sleep staging serves as a fundamental assessment for sleep quality measurement and sleep disorder diagnosis. Although current deep learning approaches have successfully integrated multimodal sleep signals, enhancing the accuracy of automatic sleep staging, certain challenges remain, as follows: 1) optimizing the utilization of multi-modal information complementarity, 2) effectively extracting both long- and short-range temporal features of sleep information, and 3) addressing the class imbalance problem in sleep data. To address these challenges, this paper proposes a two-stream encode-decoder network, named TSEDSleepNet, which is inspired by the depth sensitive attention and automatic multi-modal fusion (DSA2F) framework. In TSEDSleepNet, a two-stream encoder is used to extract the multiscale features of electrooculogram (EOG) and electroencephalogram (EEG) signals. And a self-attention mechanism is utilized to fuse the multiscale features, generating multi-modal saliency features. Subsequently, the coarser-scale construction module (CSCM) is adopted to extract and construct multi-resolution features from the multiscale features and the salient features. Thereafter, a Transformer module is applied to capture both long- and short-range temporal features from the multi-resolution features. Finally, the long- and short-range temporal features are restored with low-layer details and mapped to the predicted classification results. Additionally, the Lovász loss function is applied to alleviate the class imbalance problem in sleep datasets. Our proposed method was tested on the Sleep-EDF-39 and Sleep-EDF-153 datasets, and it achieved classification accuracies of 88.9% and 85.2% and Macro-F1 scores of 84.8% and 79.7%, respectively, thus outperforming conventional traditional baseline models. These results highlight the efficacy of the proposed method in fusing multi-modal information. This method has potential for application as an adjunct tool for diagnosing sleep disorders.

Deep learning methods have advanced quickly in brain imaging analysis over the past few years, but they are usually restricted by the limited labeled data. Pre-trained model on unlabeled data has presented promising improvement in feature learning in many domains, such as natural language processing. However, this technique is under-explored in brain network analysis. In this paper, we focused on pre-training methods with Transformer networks to leverage existing unlabeled data for brain functional network classification. First, we proposed a Transformer-based neural network, named as BrainNPT, for brain functional network classification. The proposed method leveraged token as a classification embedding vector for the Transformer model to effectively capture the representation of brain networks. Second, we proposed a pre-training framework for BrainNPT model to leverage unlabeled brain network data to learn the structure information of brain functional networks. The results of classification experiments demonstrated the BrainNPT model without pre-training achieved the best performance with the state-of-the-art models, and the BrainNPT model with pre-training strongly outperformed the state-of-the-art models. The pre-training BrainNPT model improved 8.75% of accuracy compared with the model without pre-training. We further compared the pre-training strategies and the data augmentation methods, analyzed the influence of the parameters of the model, and explained the trained model.

Early diagnosis and intervention of depression promote complete recovery, with its traditional clinical assessments depending on the diagnostic scales, clinical experience of doctors and patient cooperation. Recent researches indicate that functional near-infrared spectroscopy (fNIRS) based on deep learning provides a promising approach to depression diagnosis. However, collecting large fNIRS datasets within a standard experimental paradigm remains challenging, limiting the applications of deep networks that require more data. To address these challenges, in this paper, we propose an fNIRS-driven depression recognition architecture based on cross-modal data augmentation (fCMDA), which converts fNIRS data into pseudo-sequence activation images. The approach incorporates a time-domain augmentation mechanism, including time warping and time masking, to generate diverse data. Additionally, we design a stimulation task-driven data pseudo-sequence method to map fNIRS data into pseudo-sequence activation images, facilitating the extraction of spatial-temporal, contextual and dynamic characteristics. Ultimately, we construct a depression recognition model based on deep classification networks using the imbalance loss function. Extensive experiments are performed on the two-class depression diagnosis and five-class depression severity recognition, which reveal impressive results with accuracy of 0.905 and 0.889, respectively. The fCMDA architecture provides a novel solution for effective depression recognition with limited data.

The olfactory system enables humans to smell different odors, which are closely related to emotions. The high temporal resolution and non-invasiveness of Electroencephalogram (EEG) make it suitable to objectively study human preferences for odors. Effectively learning the temporal dynamics and spatial information from EEG is crucial for detecting odor-induced emotional valence. In this paper, we propose a deep learning architecture called Temporal Attention with Spatial Autoencoder Network (TASA) for predicting odor-induced emotions using EEG. TASA consists of a filter-bank layer, a spatial encoder, a time segmentation layer, a Long Short-Term Memory (LSTM) module, a multi-head self-attention (MSA) layer, and a fully connected layer. We improve upon the previous work by utilizing a two-phase learning framework, using the autoencoder module to learn the spatial information among electrodes by reconstructing the given input with a latent representation in the spatial dimension, which aims to minimize information loss compared to spatial filtering with CNN. The second improvement is inspired by the continuous nature of the olfactory process; we propose to use LSTM-MSA in TASA to capture its temporal dynamics by learning the intercorrelation among the time segments of the EEG. TASA is evaluated on an existing olfactory EEG dataset and compared with several existing deep learning architectures to demonstrate its effectiveness in predicting olfactory-triggered emotional responses. Interpretability analyses with DeepLIFT also suggest that TASA learns spatial-spectral features that are relevant to olfactory-induced emotion recognition.

Diagnosing multi-stage diseases typically requires doctors to consider multiple data sources, including clinical symptoms, physical signs, biochemical test results, imaging findings, pathological examination data, and even genetic data. When applying machine learning modeling to predict and diagnose multi-stage diseases, several challenges need to be addressed. Firstly, the model needs to handle multimodal data, as the data used by doctors for diagnosis includes image data, natural language data, and structured data. Secondly, privacy of patients' data needs to be protected, as these data contain the most sensitive and private information. Lastly, considering the practicality of the model, the computational requirements should not be too high. To address these challenges, this paper proposes a privacy-preserving federated deep learning diagnostic method for multi-stage diseases. This method improves the forward and backward propagation processes of deep neural network modeling algorithms and introduces a homomorphic encryption step to design a federated modeling algorithm without the need for an arbiter. It also utilizes dedicated integrated circuits to implement the hardware Paillier algorithm, providing accelerated support for homomorphic encryption in modeling. Finally, this paper designs and conducts experiments to evaluate the proposed solution. The experimental results show that in privacy-preserving federated deep learning diagnostic modeling, the method in this paper achieves the same modeling performance as ordinary modeling without privacy protection, and has higher modeling speed compared to similar algorithms. [ABSTRACT FROM AUTHOR]

Accurate and timely forecasting of critical components is pivotal in intelligent transportation systems and traffic management, crucially mitigating congestion and enhancing safety. This paper aims to comprehensively review deep learning algorithms and classical models employed in traffic forecasting. Spanning diverse traffic datasets, the study encompasses various scenarios, offering a nuanced understanding of traffic forecasting methods. Reviewing 111 seminal research works since the 1980s, encompassing both deep learning and classical models, the paper begins by detailing the data sources utilized in transportation systems. Subsequently, it delves into the theoretical underpinnings of prevalent deep learning algorithms and classical models prevalent in traffic forecasting. Furthermore, it investigates the application of these algorithms and models in forecasting key traffic characteristics, informed by their utility in transport and traffic analyses. Finally, the study elucidates the merits and drawbacks of proposed models through applied research in traffic forecasting. Findings indicate that while deep learning algorithms and classic models serve as valuable tools, their suitability varies across contexts, necessitating careful consideration in future studies. The study underscores research opportunities in road traffic forecasting, providing a comprehensive guide for future endeavors in this domain. [ABSTRACT FROM AUTHOR]

This paper proposes a deep learning-based intelligent detection device for insulation pull rod defects, addressing the issues of low detection accuracy, poor timeliness of intelligent analysis, and the difficulty in preserving detection results. Firstly, by constructing the pull rod defects dataset and training the YOLOv5s network, along with commonly used object detection algorithms in industrial defect detection, the feasibility of deep learning networks for insulation pull rod defects detection is explored. Secondly, the trained model is combined to build an intelligent detection device for pull rod defects, integrating insulation pull rod image acquisition and defect detection into a unified system. The research results demonstrate that the YOLOv5s network can quickly and accurately detect pull rod defects. On the test set constructed in this paper, the detection performance metric mAP@0.5:0.95 of the trained model reached 54.7%. Specifically, the mAP@0.5 score was 86.9% at a threshold of 0.5. The detection speed FPS reached 169.5, significantly improving the detection efficiency and accuracy compared to traditional object detection algorithms. By establishing an organic connection between the image hardware acquisition device and the deep learning network, the existing problems of inefficient detection and difficult storage of detection results in pull rod defects detection methods are effectively addressed. This research provides new insights for detecting insulation pull rod defects. [ABSTRACT FROM AUTHOR]

Lane detection is an essential task for autonomous vehicles. Deep learning-based lane detection methods are leading development in this sector. This paper proposes an algorithm named Deep Learning-based Lane Detection (DLbLD), a Convolutional Neural Network (CNN)-based lane detection algorithm. The presented paradigm deploys CNN to detect line features in the image block, predict a point on the lane line part, and project all the detected points for each frame into one-dimensional form before applying K-mean clustering to assign points to related lane lines. Extensive tests on different benchmarks were done to evaluate the performance of the proposed algorithm. The results demonstrate that the introduced DLbLD scheme achieves state-of-the-art performance, where F1 scores of 97.19 and 79.02 have been recorded for TuSimple and CU-Lane benchmarks, respectively. Nevertheless, results indicate the high accuracy of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Unmanned aerial vehicle (UAV)-based object detection methods are widely used in traffic detection due to their high flexibility and extensive coverage. In recent years, with the increasing complexity of the urban road environment, UAV object detection algorithms based on deep learning have gradually become a research hotspot. However, how to further improve algorithmic efficiency in response to the numerous and rapidly changing road elements, and thus achieve high-speed and accurate road object detection, remains a challenging issue. Given this context, this paper proposes the high-efficiency multi-object detection algorithm for UAVs (HeMoDU). HeMoDU reconstructs a state-of-the-art, deep-learning-based object detection model and optimizes several aspects to improve computational efficiency and detection accuracy. To validate the performance of HeMoDU in urban road environments, this paper uses the public urban road datasets VisDrone2019 and UA-DETRAC for evaluation. The experimental results show that the HeMoDU model effectively improves the speed and accuracy of UAV object detection. [ABSTRACT FROM AUTHOR]

Fruit recognition and location are the premises of robot automatic picking. YOLOv3 has been used to detect different fruits in complex environment. However, for the object with definite features, the complex network structure will increase the computing time and may cause overfitting. Therefore, this paper has carried out a lightweight design for the YOLOv3. This paper proposed an improved T-Net to detect tomato images. Firstly, the T-Net reduces the residual network layers. This paper changed the number of cycles in each group of the residual unit to 1, 2, 2, 1, and 1. Second, two feature layers with different scales are selected according to the features of tomatoes. Meanwhile, the convolutional layer at the neck has been reduced by two layers. Finally, the location and approximate diameter of the ripe tomato are obtained by combining the node information of the Intel D435i camera and T-Net in the Robot Operation System. T-Net obtains mean average precision (mAP) of 99.2%, F1-score of 98.9%, precision of 99.0%, and recall of 98.8% at a detection rate of 104.2 FPS. The proposed T-Net has outperformed the YOLOv3 with 0.4%, 0.1%, and 0.2% increase in precision, mAP, and F1-score. The detection speed of T-Net is 1.8 times faster than YOLOv3. The mean errors of the center coordinates and diameter of the tomato are 8.5 mm and 2.5 mm, respectively. This model provides a method for efficient real-time detection and location of tomatoes. [ABSTRACT FROM AUTHOR]