Average Similarity Research Articles

Employing Siamese Networks as Quantitative Biomarker for Assessing the Effect of Dolphin-Assisted Therapy on Pediatric Cerebral Palsy.

This study explores the potential of using a Siamese Network as a biomarker for assessing the effectiveness of Dolphin-Assisted Therapy (DAT) in children with Spastic Cerebral Palsy (SCP). The problem statement revolves around the need for objective measures to evaluate the impact of DAT on patients with SCP, considering the subjective nature of traditional assessment methods. The methodology involves training a Siamese network, a type of neural network designed to compare similarities between inputs, using data collected from SCP patients undergoing DAT sessions. The study employed Event-Related Potential (ERP) and Fast Fourier Transform (FFT) analyses to examine cerebral activity and brain rhythms, proposing the use of SNN to compare electroencephalographic (EEG) signals of children with cerebral palsy before and after Dolphin-Assisted Therapy. Testing on samples from four children yielded a high average similarity index of 0.9150, indicating consistent similarity metrics before and after therapy. The network is trained to learn patterns and similarities between pre- and post-therapy evaluations, in order to identify biomarkers indicative of therapy effectiveness. Notably, the Siamese Network's architecture ensures that comparisons are made within the same feature space, allowing for more accurate assessments. The results of the study demonstrate promising findings, indicating different patterns in the output of the Siamese Network that correlate with improvements in symptoms of SCP post-DAT. Confirming these observations will require large, longitudinal studies but such findings would suggest that the Siamese Network could have utility as a biomarker in monitoring treatment responses for children with SCP who undergo DAT and offer them more objective as well as quantifiable manners of assessing therapeutic interventions. Great discrepancies in neuronal voltage perturbations, 7.9825 dB on average at the specific samples compared to the whole dataset (6.2838 dB), imply a noted deviation from resting activity. These findings indicate that Dolphin-Assisted Therapy activates particular brain regions specifically during the intervention.

Spatial-aware contrastive learning for cross-domain medical image registration.

With the rapid advancement of medical imaging technologies, precise image analysis and diagnosis play a crucial role in enhancing treatment outcomes and patient care. Computed tomography (CT) and magnetic resonance imaging (MRI), as pivotal technologies in medical imaging, exhibit unique advantages in bone imaging and soft tissue contrast, respectively. However, cross-domain medical image registration confronts significant challenges due to the substantial differences in contrast, texture, and noise levels between different imaging modalities. The purpose of this study is to address the major challenges encountered in the field of cross-domain medical image registration by proposing a spatial-aware contrastive learning approach that effectively integrates shared information from CT and MRI images. Our objective is to optimize the feature space representation by employing advanced reconstruction and contrastive loss functions, overcoming the limitations of traditional registration methods when dealing with different imaging modalities. Through this approach, we aim to enhance the model's ability to learn structural similarities across domain images, improve registration accuracy, and provide more precise imaging analysis tools for clinical diagnosis and treatmentplanning. With prior knowledge that different domains of images (CT and MRI) share same content-style information, we extract equivalent feature spaces from both images, enabling accurate cross-domain point matching. We employ a structure resembling that of an autoencoder, augmented with designed reconstruction and contrastive losses to fulfill our objectives. We also propose region mask to solve the conflict between spatial correlation and distinctiveness, to obtain a better representation space. Our research results demonstrate the significant superiority of the proposed spatial-aware contrastive learning approach in the domain of cross-domain medical image registration. Quantitatively, our method achieved an average Dice similarity coefficient (DSC) of 85.68%, target registration error (TRE) of 1.92 mm, and mean Hausdorff distance (MHD) of 1.26 mm, surpassing current state-of-the-art methods. Additionally, the registration processing time was significantly reduced to 2.67 s on a GPU, highlighting the efficiency of our approach. The experimental outcomes not only validate the effectiveness of our method in improving the accuracy of cross-domain image registration but also prove its adaptability across different medical image analysis scenarios, offering robust support for enhancing diagnostic precision and patient treatmentoutcomes. The spatial-aware contrastive learning approach proposed in this paper introduces a new perspective and solution to the domain of cross-domain medical image registration. By effectively optimizing the feature space representation through carefully designed reconstruction and contrastive loss functions, our method significantly improves the accuracy and stability of registration between CT and MRI images. The experimental results demonstrate the clear advantages of our approach in enhancing the accuracy of cross-domain image registration, offering significant application value in promoting precise diagnosis and personalized treatment planning. In the future, we look forward to further exploring the application of this method in a broader range of medical imaging datasets and its potential integration with other advanced technologies, contributing more innovations to the field of medical image analysis andprocessing.

EEMSNet: Eagle-Eye Multi-Scale Supervised Network for cardiac segmentation

Cardiac segmentation plays a crucial role in computer-aided diagnosis of cardiac diseases. Nevertheless, the task of cardiac segmentation is inherently arduous due to the complex cardiac anatomy, and the intensity heterogeneity introduced during the acquisition of magnetic resonance images. To address these challenges, this paper presents an Eagle-Eye Multi-Scale Supervision Network. The network improves on the Full Convolutional Transformer network and introduces an innovative Eagle-Eye Multi-Scale Supervised module, which is used exclusively for edge supervision during the training process. The edge-supervised branch supervises the structure of cardiac edges at different scales within the input image pyramid structure. During training, the Eagle-Eye Multi-Scale Supervised module is based on multi-scale inputs. It uses edge information from Ground Truth to guide the model in learning the structural and detailed features of cardiac edges, ensuring the model captures edges accurately. Furthermore, considering that most cardiac segmentation networks typically overlook boundary profile information, our designed loss function places particular emphasis on the preservation of boundary profiles after the fusion of multi-scale edge features. Experimental results demonstrate that our proposed the Eagle-Eye Multi-Scale Supervision Network exhibits outstanding performance on the Automatic Cardiac Diagnosis Challenge dataset, achieving a notable average Dice Similarity Coefficient of 93.21%. Particularly, the Dice Similarity Coefficient value for right ventricle segmentation stands out at an impressive 93.52%. These findings signify the exceptional accuracy and effectiveness of our approach in addressing cardiac image segmentation tasks.

A New Strategy: Remaining Useful Life Prediction of Wind Power Bearings Based on Deep Learning under Data Missing Conditions

With its formidable nonlinear mapping capabilities, deep learning has been widely applied in bearing remaining useful life (RUL) prediction. Given that equipment in actual work is subject to numerous disturbances, the collected data tends to exhibit random missing values. Furthermore, due to the dynamic nature of wind turbine environments, LSTM models relying on manually set parameters exhibit certain limitations. Considering these factors can lead to issues with the accuracy of predictive models when forecasting the remaining useful life (RUL) of wind turbine bearings. In light of this issue, a novel strategy for predicting the remaining life of wind turbine bearings under data scarcity conditions is proposed. Firstly, the average similarity (AS) is introduced to reconstruct the discriminator of the Generative Adversarial Imputation Nets (GAIN), and the adversarial process between the generative module and the discriminant is strengthened. Based on this, the dung beetle algorithm (DBO) is used to optimize multiple parameters of the long-term and short-term memory network (LSTM), and the complete data after filling is used as the input data of the optimized LSTM to realize the prediction of the remaining life of the wind power bearing. The effectiveness of the proposed method is verified by the full-life data test of bearings. The results show that, under the condition of missing data, the new strategy of AS-GAIN-LSTM is used to predict the RUL of wind turbine bearings, which has a more stable prediction performance.

Artificial intelligence-based evaluation of carotid artery compressibility via point-of-care ultrasound in determining the return of spontaneous circulation during cardiopulmonary resuscitation

AimThis study introduces RealCAC-Net, an artificial intelligence (AI) system, to quantify carotid artery compressibility (CAC) and determine the return of spontaneous circulation (ROSC) during cardiopulmonary resuscitation. MethodsA prospective study based on data from a South Korean emergency department from 2022 to 2023 investigated carotid artery compressibility in adult patients with cardiac arrest using a novel AI model, RealCAC-Net. The data comprised 11,958 training images from 161 cases and 15,080 test images from 134 cases. RealCAC-Net processes images in three steps: TransUNet-based segmentation, the carotid artery compressibility measurement algorithm for improved segmentation and CAC calculation, and CAC-based classification from 0 (indicating a circular shape) to 1 (indicating high compression). The accuracy of the ROSC classification model was tested using metrics such as the dice similarity coefficient, intersection-over-union, precision, recall, and F1 score. ResultsRealCAC-Net, which applied the carotid artery compressibility measurement algorithm, performed better than the baseline model in cross-validation, with an average dice similarity coefficient of 0.90, an intersection-over-union of 0.84, and a classification accuracy of 0.96. The test set achieved a classification accuracy of 0.96 and an F1 score of 0.97, demonstrating its efficacy in accurately identifying ROSC in cardiac arrest situations. ConclusionsRealCAC-Net enabled precise CAC quantification for ROSC determination during cardiopulmonary resuscitation. Future research should integrate this AI-enhanced ultrasound approach to revolutionize emergency care.

Medical Image Classification using a Many to Many Relation, Multilayered Fuzzy Systems and AI

One of the research gaps in the medical sciences is the study of orphan diseases or rare diseases, due to limited data availability of rare diseases. Our previous study addressed this successfully by developing an Artificial Intelligence (AI)-based medical image classification method using a multilayer fuzzy approach (MFA), for detecting and classifying image abnormalities for large and very small datasets. A fuzzy system is an AI system used to handle imprecise data. There are more than three types of fuzziness in any image data set: 1) due to a projection of a 3D object on a 2D surface, 2) due to the digitalization of the scan, and 3) conversion of the digital image to grayscale, and more. Thus, this was referred to in the previous study as a multilayer fuzzy system, since fuzziness arises from multiple sources. The method used in MFA involves comparing normal images containing abnormalities with the same kind of image without abnormalities, yielding a similarity measure percentage that, when subtracted from a hundred, reveals the abnormality. However, relying on a single standard image in the MFA reduces efficiency, since images vary in contrast, lighting, and patient demographics, impacting similarity percentages. To mitigate this, the current study focused on developing a more robust medical image classification method than MFA, using a many-to-many relation and a multilayer fuzzy approach (MCM) that employs multiple diverse standard images to compare with the abnormal image. For each abnormal image, the average similarity was calculated across multiple normal images, addressing issues encountered with MFA, and enhancing versatility. In this study, an AI-based method of image analysis automation that utilizes fuzzy systems was applied to a cancer data set for the first time. MCM proved to be highly efficient in detecting the abnormality in all types of images and sample sizes and surpassed the gold standard, the convolutional neural network (CNN), in detecting the abnormality in images from a very small data set. Moreover, MCM detects and classifies abnormality without any training, validation, or testing steps for large and small data sets. Hence, MCM may be used to address one of the research gaps in medicine, which detects, quantifies, and classifies images related to rare diseases with small data sets. This has the potential to assist a physician with early detection, diagnosis, monitoring, and treatment planning of several diseases, especially rare diseases.

Quantitative susceptibility mapping based basal ganglia segmentation via AGSeg: leveraging active gradient guiding mechanism in deep learning.

With better visual contrast and the ability for magnetic susceptibility quantification analysis, quantitative susceptibility mapping (QSM) has emerged as an important magnetic resonance imaging (MRI) method for basal ganglia studies. Precise segmentation of basal ganglia is a prerequisite for quantification analysis of tissue magnetic susceptibility, which is crucial for subsequent disease diagnosis and surgical planning. The conventional method of localizing and segmenting basal ganglia heavily relies on layer-by-layer manual annotation by experts, resulting in a tedious amount of workload. Although several morphology registration and deep learning based methods have been developed to automate segmentation, the voxels around the nuclei boundary remain a challenge to distinguish due to insufficient tissue contrast. This paper proposes AGSeg, an active gradient guidance-based susceptibility and magnitude information complete (MIC) network for real-time and accurate basal ganglia segmentation. Various datasets, including clinical scans and data from healthy volunteers, were collected across multiple centers with different magnetic field strengths (3T/5T/7T), with a total of 210 three-dimensional (3D) susceptibility measurements. Manual segmentations following fixed rules for anatomical borders annotated by experts were used as ground truth labels. The proposed network took QSM maps and Magnitude images as two individual inputs, of which the features are selectively enhanced in the proposed magnitude information complete (MIC) module. AGSeg utilized a dual-branch architecture, with Seg-branch aiming to generate a proper segmentation map and Grad-branch to reconstruct the gradient map of regions of interest (ROIs). With the support of the newly designed active gradient module (AGM) and gradient guiding module (GGM), the Grad-branch provided attention guidance for the Seg-branch, facilitating it to focus on the boundary of target nuclei. Ablation studies were conducted to assess the functionality of the proposed modules. Significant performance decrement was observed after ablating relative modules. AGSeg was evaluated against several existing methods on both healthy and clinical data, achieving an average Dice similarity coefficient (DSC) =0.874 and average 95% Hausdorff distance (HD95) =2.009. Comparison experiments indicated that our model had superior performance on basal ganglia segmentation and better generalization ability over existing methods. The AGSeg outperformed all implemented comparison deep learning algorithms with average DSC enhancement ranging from 0.036 to 0.074. The current work integrates a deep learning-based method into automated basal ganglia segmentation. The high processing speed and segmentation robustness of AGSeg contribute to the feasibility of future surgery planning and intraoperative navigation. Experiments show that leveraging active gradient guidance mechanisms and magnitude information completion can facilitate the segmentation process. Moreover, this approach also offers a portable solution for other multi-modality medical image segmentation tasks.

Genotypic Identification of Trees Using DNA Barcodes and Microbiome Analysis of Rhizosphere Microbial Communities.

DNA barcodes can provide accurate identification of plants. We used previously reported DNA primers targeting the internal transcribed spacer (ITS1) region of the nuclear ribosomal cistron, internal transcribed spacer (ITS2), and chloroplast trnL (UAA) intron to identify four trees at Bergen Community College. Two of the four trees were identified as Acer rubrum and Fagus sylvatica. However, Quercus was only identified at the genus level, and the fourth tree did not show similar identification between barcodes. Next-generation sequencing of 16S rRNA genes showed that the predominant bacterial communities in the rhizosphere mainly consisted of the Pseudomonadota, Actinomycetota, Bacteroidota, and Acidobacteriota. A. rubrum showed the most diverse bacterial community while F. sylvatica was less diverse. The genus Rhodoplanes showed the highest relative bacterial abundance in all trees. Fungal ITS sequence analysis demonstrated that the communities predominantly consisted of the Ascomycota and Basidiomycota. Quercus showed the highest fungi diversity while F. sylvatica showed the lowest. Russula showed the highest abundance of fungi genera. Average similarity values in the rhizosphere for fungi communities at the phylum level were higher than for bacteria. However, at the genus level, bacterial communities showed higher similarities than fungi. Similarity values decreased at lower taxonomical levels for both bacteria and fungi, indicating each tree has selected for specific bacterial and fungal communities. This study confirmed the distinctiveness of the microbial communities in the rhizosphere of each tree and their importance in sustaining and supporting viability and growth but also demonstrating the limitations of DNA barcoding with the primers used in this study to identify genus and species for some of the trees. The optimization of DNA barcoding will require additional DNA sequences to enhance the resolution and identification of trees at the study site.

The association between discourse production and schizotypal personality traits

Discourse abnormalities are a prominent feature in schizophrenia spectrum disorders, including poor lexical diversity, and have been found to differentiate patients from healthy subjects. However, discourse processing in individuals with high schizotypal personality traits is less understood and is often overshadowed by research on clinical psychotic symptoms. In the present study, we examined schizotypal traits at a non-clinical threshold and their association with lexical diversity and discourse coherence using two automated Natural Language Processing (NLP) tools - Type-Token-Ratio (TTR) measures from the Tool for the Automatic Analysis of Lexical Diversity (TAALED) and discourse coherence using sentence-level average cosign similarity with FastText to assess sentence similarity. 276 college students completed the full assessment including the Schizotypal Personality Questionnaire- Brief Revised (SPQ-BR) and recorded a speech sample describing a detailed painting. Results revealed that high schizotypal traits, specifically positive traits, were associated with lower lexical diversity and higher sentence similarity. Our findings suggest that even when clinically significant symptoms are not present, discourse abnormalities are present in healthy populations with high ST. The stronger association with positive traits suggests that theories of perseveration and top-down processing may warrant further investigation in schizophrenia-spectrum disorders.

An automated in vitro wound healing microscopy image analysis approach utilizing U-net-based deep learning methodology

BackgroundThe assessment of in vitro wound healing images is critical for determining the efficacy of the therapy-of-interest that may influence the wound healing process. Existing methods suffer significant limitations, such as user dependency, time-consuming nature, and lack of sensitivity, thus paving the way for automated analysis approaches.MethodsHereby, three structurally different variations of U-net architectures based on convolutional neural networks (CNN) were implemented for the segmentation of in vitro wound healing microscopy images. The developed models were fed using two independent datasets after applying a novel augmentation method aimed at the more sensitive analysis of edges after the preprocessing. Then, predicted masks were utilized for the accurate calculation of wound areas. Eventually, the therapy efficacy-indicator wound areas were thoroughly compared with current well-known tools such as ImageJ and TScratch.ResultsThe average dice similarity coefficient (DSC) scores were obtained as 0.958∼\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\sim$$\\end{document}0.968 for U-net-based deep learning models. The averaged absolute percentage errors (PE) of predicted wound areas to ground truth were 6.41%, 3.70%, and 3.73%, respectively for U-net, U-net++, and Attention U-net, while ImageJ and TScratch had considerable averaged error rates of 22.59% and 33.88%, respectively.ConclusionsComparative analyses revealed that the developed models outperformed the conventional approaches in terms of analysis time and segmentation sensitivity. The developed models also hold great promise for the prediction of the in vitro wound area, regardless of the therapy-of-interest, cell line, magnification of the microscope, or other application-dependent parameters.

Heterogeneity of morphometric similarity networks in health and schizophrenia.

Morphometric similarity is a recently developed neuroimaging phenotype of inter-regional connectivity by quantifying the similarity of a region to other regions based on multiple MRI parameters. Altered average morphometric similarity has been reported in psychotic disorders at the group level, with considerable heterogeneity across individuals. We used normative modeling to address cross-sectional and longitudinal inter-individual heterogeneity of morphometric similarity in health and schizophrenia. Morphometric similarity for 62 cortical regions was obtained from baseline and follow-up T1-weighted scans of healthy individuals and patients with chronic schizophrenia. Cortical regions were classified into seven predefined brain functional networks. Using Bayesian Linear Regression and taking into account age, sex, image quality and scanner, we trained and validated normative models in healthy controls from eleven datasets (n = 4310). Individual deviations from the norm (z-scores) in morphometric similarity were computed for each participant for each network and region at both timepoints. A z-score ≧ than 1.96 was considered supra-normal and a z-score ≦ -1.96 infra-normal. As a longitudinal metric, we calculated the change over time of the total number of infra- or supra-normal regions per participant. At baseline, patients with schizophrenia had decreased morphometric similarity of the default mode network and increased morphometric similarity of the somatomotor network when compared with healthy controls. The percentage of patients with infra- or supra-normal values for any region at baseline and follow-up was low (<6%) and did not differ from healthy controls. Mean intra-group changes over time in the total number of infra- or supra-normal regions were small in schizophrenia and healthy control groups (<1) and there were no significant between-group differences. In a case-control setting, a decrease of morphometric similarity within the default mode network may be a robust finding implicated in schizophrenia. However, normative modeling suggests that significant reductions and changes over time of regional morphometric similarity are evident only in a minority of patients.

Precision dose prediction for breast cancer patients undergoing IMRT: The Swin-UMamba-Channel Model

BackgroundRadiation therapy is one of the crucial treatment modalities for cancer. An excellent radiation therapy plan relies heavily on an outstanding dose distribution map, which is traditionally generated through repeated trials and adjustments by experienced physicists. However, this process is both time-consuming and labor-intensive, and it comes with a degree of subjectivity. Now, with the powerful capabilities of deep learning, we are able to predict dose distribution maps more accurately, effectively overcoming these challenges. MethodsIn this study, we propose a novel Swin-UMamba-Channel prediction model specifically designed for predicting the dose distribution of patients with left breast cancer undergoing radiotherapy after total mastectomy. This model integrates anatomical position information of organs and ray angle information, significantly enhancing prediction accuracy. Through iterative training of the generator (Swin-UMamba) and discriminator, the model can generate images that closely match the actual dose, assisting physicists in quickly creating DVH curves and shortening the treatment planning cycle. Our model exhibits excellent performance in terms of prediction accuracy, computational efficiency, and practicality, and its effectiveness has been further verified through comparative experiments with similar networks. ResultsThe results of the study indicate that our model can accurately predict the clinical dose of breast cancer patients undergoing intensity-modulated radiation therapy (IMRT). The predicted dose range is from 0 to 50 Gy, and compared with actual data, it shows a high accuracy with an average Dice similarity coefficient of 0.86. Specifically, the average dose change rate for the planning target volume ranges from 0.28 % to 1.515 %, while the average dose change rates for the right and left lungs are 2.113 % and 0.508 %, respectively. Notably, due to their small sizes, the heart and spinal cord exhibit relatively higher average dose change rates, reaching 3.208 % and 1.490 %, respectively. In comparison with similar dose studies, our model demonstrates superior performance. Additionally, our model possesses fewer parameters, lower computational complexity, and shorter processing time, further enhancing its practicality and efficiency. These findings provide strong evidence for the accuracy and reliability of our model in predicting doses, offering significant technical support for IMRT in breast cancer patients. ConclusionThis study presents a novel Swin-UMamba-Channel dose prediction model, and its results demonstrate its precise prediction of clinical doses for the target area of left breast cancer patients undergoing total mastectomy and IMRT. These remarkable achievements provide valuable reference data for subsequent plan optimization and quality control, paving a new path for the application of deep learning in the field of radiation therapy.

Reconstructing images of objects: method for reconstructing images from digital off-axis holograms based on a generative adversarial neural network

The reconstruction of object images that are located in 3D scene cross-sections using digital holography is described. The potential of generative adversarial networks for reconstructing cross-sections of 3D scenes composed of multiple layers of off-axis objects from holograms is investigated. Such scenes consist of a series of sections with objects that are not aligned with the camera’s axis. Digital holograms were used to reconstruct images of cross-sectional views of 3D scenes. It has been shown that the use of neural networks increases the speed and reconstruction quality, and reduces the image noise. A method for reconstructing images of objects using digital off-axis holograms and a generative adversarial neural network is proposed. The proposed method was tested on both numerically simulated and experimentally captured digital holograms. It was able to successfully reconstruct up to 8 cross-sections of a 3D scene from a single hologram. It was obtained that an average structural similarity index measure was equal to at least 0.73. Based on optically registered holograms, the method allowed us to reconstruct object image cross-sections of a 3D scene with a structural similarity index measure over cross-sections of a 3D scene of equal to 0.83. Therefore, the proposed technique provides the possibility for high-quality object image reconstruction and could be utilized in the analysis of micro- and macroobjects, including medical and biological applications, metrology, characterization of materials, surfaces, and volume media.

Deep learning decryption approach for asymmetric computer-generated holography (CGH) cryptosystem.

Deep-learning-based optical image decryption has attracted attention due to its remarkable advantages of keyless managements. Here, a high-fidelity deep learning (DL) decryption strategy is proposed, aiming for the asymmetric DRPE-based CGH cryptosystem, which is combined with phase truncation technique and chaotic iris phase masks. First, a mass of ciphertext and plaintext image pairs are generated to create a dataset. Then, a deep neural network, namely ACGHC-Net (network for the asymmetric DRPE-based CGH cryptosystem), is designed and trained in a supervised learning manner. After the model training and tuning, the ACGHC-Net can quickly and accurately decrypt the ciphertext images. The average cross-correlation coefficient (CC) of the decrypted images achieves 0.998, the average structural similarity (SSIM) 0.895, and the average peak signal-to-noise ratio (PSNR) 31.090 dB. Furthermore, we conducted anti-noise and anti-clipping analysis on the ACGHC-Net. The results prove that the proposed ACGHC-Net can successfully decrypt the encrypted complex grayscale images, and has good anti-noise and anti-cropping robustness for the asymmetric DRPE-based CGH cryptosystem. The proposed method will be expected to further boost keyless decryption in image encryption systems.

Global Field Time-Frequency Representation-Based Discriminative Similarity Analysis of Passive Auditory ERPs for Diagnosis of Disorders of Consciousness.

Behavioural diagnosis of patients with disorders of consciousness (DOC) is challenging and prone to inaccuracies. Consequently, there have been increased efforts to develop bedside assessment based on EEG and event-related potentials (ERPs) that are more sensitive to the neural factors supporting conscious awareness. However, individual detection of residual consciousness using these techniques is less established. Here, we hypothesize that the cross-state similarity (defined as the similarity between healthy and impaired conscious states) of passive brain responses to auditory stimuli can index the level of awareness in individual DOC patients. To this end, we introduce the global field time-frequency representation-based discriminative similarity analysis (GFTFR-DSA). This method quantifies the average cross-state similarity index between an individual patient and our constructed healthy templates using the GFTFR as an EEG feature. We demonstrate that the proposed GFTFR feature exhibits superior within-group consistency in 34 healthy controls over traditional EEG features such as temporal waveforms. Second, we observed the GFTFR-based similarity index was significantly higher in patients with a minimally conscious state (MCS, 40 patients) than those with unresponsive wakefulness syndrome (UWS, 54 patients), supporting our hypothesis. Finally, applying a linear support vector machine classifier for individual MCS/UWS classification, the model achieved a balanced accuracy and F1 score of 0.77. Overall, our findings indicate that combining discriminative and interpretable markers, along with automatic machine learning algorithms, is effective for the differential diagnosis in patients with DOC. Importantly, this approach can, in principle, be transferred into any ERP of interest to better inform DOC diagnoses.

Application of KL distance-based intelligent recommendation method to fund recommendation for users with investment behavior in Asia Region

Over the past decade, the spectacular growth of many economies in Asia has amazed the economics profession and Asian people's quality of life has been significantly improved. Therefore, people's savings are increasing. The booming financial market attracts more people to invest in funds for its relatively low risk and high returns. This research first analyzed the widely used recommendation algorithms. Then it adopted the K-medoids clustering algorithm based on KL divergence to make recommendations. The recommendations were also based on the user-item rating matrix. First, the improved Kullback-Leibler distance was used to calculate the distance and similarity between the funds. Next, the data set of n funds were split by the K-medoids clustering technique into k clusters to obtain funds that were similar to the target fund. Finally, the prediction accuracy of the KL-KM recommendation algorithm was compared with those of traditional recommendation algorithms. When the number of recommendations changed from 1 to 5, the average minimum KL distance varied from 0.16, 0.22, 0.29, 0.31, to 0.39. The average KL similarity is 0.85, 0.83, 0.79, 0.74, and 0.72, respectively, and the average absolute error is 0.83, 0.82, 0.82, 0.83, and 0.82, respectively. The root mean square error (RMSE) of the proposed algorithm is 0.93, 0.94, 0.93, 0.94 and 0.94, respectively. Therefore, the proposed recommendation model has a better recommendation performance than existing models to meet users' demands. This algorithm generates recommendations of similar funds based on customers' purchase history. Moreover, principal component analysis is used to simplify the large data set of indicator values into a smaller set while still maintaining significant patterns and trends, thereby improving the accuracy and reducing the complexity of the model.

Research on case retrieval strategy based on drawing die of automobile coverings

Abstract To improve the accuracy and efficiency of case retrieval in the process of intelligent design of automobile panel drawing die, this paper proposes a case retrieval strategy based on model reasoning and an improved k-nearest neighbor (KNN) method. Through the functional representation and modular description of the existing knowledge of automobile panel drawing die design, the problems to be solved are compared with the model, and the preliminary screening is completed. Then the nearest neighbor algorithm is used to realize the retrieval, and the subjective and objective weight assignment method is used to optimize the retrieval strategy. The subjective weight uses the improved three-scale analytic hierarchy process to avoid the subjectivity increase caused by the judgment matrix’s artificial adjustment. The objective weight uses the grey wolf optimization algorithm, and the dynamic adaptive calculation of the average similarity is designed to enhance the reliability of the fitness function. Finally, the case retrieval of the drawing die during the design of a certain type of panel die is taken as an example to test. The retrieval strategy can accurately complete the retrieval of historically similar cases, which verifies the effectiveness of the proposed instance retrieval strategy.

Uncertainty estimation and evaluation of deformation image registration based convolutional neural networks

Objective. Fast and accurate deformable image registration (DIR), including DIR uncertainty estimation, is essential for safe and reliable clinical deployment. While recent deep learning models have shown promise in predicting DIR with its uncertainty, challenges persist in proper uncertainty evaluation and hyperparameter optimization for these methods. This work aims to develop and evaluate a model that can perform fast DIR and predict its uncertainty in seconds. Approach. This study introduces a novel probabilistic multi-resolution image registration model utilizing convolutional neural networks to estimate a multivariate normal distributed dense displacement field (DDF) in a multimodal image registration problem. To assess the quality of the DDF distribution predicted by the model, we propose a new metric based on the Kullback–Leibler divergence. The performance of our approach was evaluated against three other DIR algorithms (VoxelMorph, Monte Carlo dropout, and Monte Carlo B-spline) capable of predicting uncertainty. The evaluation of the models included not only the quality of the deformation but also the reliability of the estimated uncertainty. Our application investigated the registration of a treatment planning computed tomography (CT) to follow-up cone beam CT for daily adaptive radiotherapy. Main results. The hyperparameter tuning of the models showed a trade-off between the estimated uncertainty’s reliability and the deformation’s accuracy. In the optimal trade-off, our model excelled in contour propagation and uncertainty estimation (p <0.05) compared to existing uncertainty estimation models. We obtained an average dice similarity coefficient of 0.89 and a KL-divergence of 0.15. Significance. By addressing challenges in DIR uncertainty estimation and evaluation, our work showed that both the DIR and its uncertainty can be reliably predicted, paving the way for safe deployment in a clinical environment.

Cardiac MRI segmentation using shifted-window multilayer perceptron mixer networks

Objectives. In this work, we proposed a deep-learning segmentation algorithm for cardiac magnetic resonance imaging to aid in contouring of the left ventricle, right ventricle, and Myocardium (Myo). Approach. We proposed a shifted window multilayer perceptron (Swin-MLP) mixer network which is built upon a 3D U-shaped symmetric encoder-decoder structure. We evaluated our proposed network using public data from 100 individuals. The network performance was quantitatively evaluated using 3D volume similarity between the ground truth contours and the predictions using Dice score coefficient, sensitivity, and precision as well as 2D surface similarity using Hausdorff distance (HD), mean surface distance (MSD) and residual mean square distance (RMSD). We benchmarked the performance against two other current leading edge networks known as Dynamic UNet and Swin-UNetr on the same public dataset. Results. The proposed network achieved the following volume similarity metrics when averaged over three cardiac segments: Dice = 0.952 ± 0.017, precision = 0.948 ± 0.016, sensitivity = 0.956 ± 0.022. The average surface similarities were HD = 1.521 ± 0.121 mm, MSD = 0.266 ± 0.075 mm, and RMSD = 0.668 ± 0.288 mm. The network shows statistically significant improvement in comparison to the Dynamic UNet and Swin-UNetr algorithms for most volumetric and surface metrics with p-value less than 0.05. Overall, the proposed Swin-MLP mixer network demonstrates better or comparable performance than competing methods. Significance. The proposed Swin-MLP mixer network demonstrates more accurate segmentation performance compared to current leading edge methods. This robust method demonstrates the potential to streamline clinical workflows for multiple applications.

PAC-UNet: Parallel Dual Self-Attention with Convolution for Meniscal MRI Image Segmentation

Meniscal tears are common sports-injury disorders, and thus accurate segmentation of the meniscus is a worthwhile means of medical diagnosis. However, due to the limited research on image segmentation of the meniscus, most previous works have poor segmentation effects on the meniscus, which affects the adoption of assisted medical diagnosis of meniscus. To address this issue, we propose an improved neural network model, termed as the PAC-UNet model, which introduces a parallel dual self-attention with depth-wise convolution for the meniscus image segmentation. Firstly, in the encoder stage, the parallel block is used to eliminate the negative impact of weak modeling ability caused by weight sharing on shared dimensions, so as to better learn the semantic information. Then, in the decoder stage, the parallel block and patch expansion are used to restore the predicted results of meniscus segmentation by performing self-attention and up-sampling on the features. By analyzing the outcomes of experiments on the meniscus dataset, which were collected from the hospital, it is found that the proposed network performed better than other medical image segmentation networks, where the average Dice Similarity Coefficient (DSC), Intersection Over Union (IOU), precision, and Hausdorff_95 of the medial meniscus and lateral meniscus are up to 89.01%, 80.81%, 90.5%, and 2.43, respectively.