Importing Diffusion and Re-Designed Backward Process for Image De-Raining.

In recent years, with the increasing demand for high-quality images in various fields, more and more attention has been focused on noise removal techniques for image processing. The effective elimination of unwanted noise plays a crucial role in improving image quality. To meet this challenge, many noise removal methods have been proposed, among which the diffusion model has become one of the focuses of many researchers. In order to make the restored image closer to the real image and retain more features of the image, this paper proposes a DIR-SDE method with reference to the diffusion models of IR-SDE and IDM, which improve the feature retention of the image in the de-raining process, and then improve the realism of the image for the image de-raining task. In this study, IR-SDE was used as the base structure of the diffusion model, IR-SDE was improved, and DINO-ViT was combined to enhance the image features. During the diffusion process, the image features were extracted using DINO-ViT, and these features were fused with the original images to enhance the learning effect of the model. The model was also trained and validated with the Rain100H dataset. Compared with the IR-SDE method, it improved 0.003 in the SSIM, 0.003 in the LPIPS, and 1.23 in the FID. The experimental results show that the diffusion model proposed in this study can effectively improve the image restoration performance.

Explainable machine learning model for identifying key gut microbes and metabolites biomarkers associated with myasthenia gravis.

Diagnostic markers for myasthenia gravis (MG) are limited; thus, innovative approaches are required for supportive diagnosis and personalized care. Gut microbes are associated with MG pathogenesis; however, few studies have adopted machine learning (ML) to identify the associations among MG, gut microbiota, and metabolites. In this study, we developed an explainable ML model to predict biomarkers for MG diagnosis. We enrolled 19 MG patients and 10 non-MG individuals. Stool samples were collected and microbiome assessment was performed using 16S rRNA sequencing. Untargeted metabolic profiling was conducted to identify fecal amplicon significant variants (ASVs) and metabolites. We developed an explainable ML model in which the top ASVs and metabolites are combined to identify the best predictive performance. This model uses the SHapley Additive exPlanations method to generate both global and personalized explanations. Fecal microbe-metabolite composition differed significantly between groups. The key bacterial families were Lachnospiraceae and Ruminococcaceae, and the top three features were Lachnospiraceae, inosine, and methylhistidine. An ML model trained with the top 1 % ASVs and top 15 % metabolites combined outperformed all other models. Personalized explanations revealed different patterns of microbe-metabolite contributions in patients with MG. The integration of the microbiota-metabolite features and the development of an explainable ML framework can accurately identify MG and provide personalized explanations, revealing the associations between gut microbiota, metabolites, and MG. An online calculator employing this algorithm was developed that provides a streamlined interface for MG diagnosis screening and conducting personalized evaluations.

Development of a Machine Learning Algorithm to Correlate Lumbar Disc Height on X-rays with Disc Bulging or Herniation.

Lumbar disc bulging or herniation (LDBH) is one of the major causes of spinal stenosis and related nerve compression, and its severity is the major determinant for spine surgery. MRI of the spine is the most important diagnostic tool for evaluating the need for surgical intervention in patients with LDBH. However, MRI utilization is limited by its low accessibility. Spinal X-rays can rapidly provide information on the bony structure of the patient. Our study aimed to identify the factors associated with LDBH, including disc height, and establish a clinical diagnostic tool to support its diagnosis based on lumbar X-ray findings. In this study, a total of 458 patients were used for analysis and 13 clinical and imaging variables were collected. Five machine-learning (ML) methods, including LASSO regression, MARS, decision tree, random forest, and extreme gradient boosting, were applied and integrated to identify important variables for predicting LDBH from lumbar spine X-rays. The results showed L4-5 posterior disc height, age, and L1-2 anterior disc height to be the top predictors, and a decision tree algorithm was constructed to support clinical decision-making. Our study highlights the potential of ML-based decision tools for surgeons and emphasizes the importance of L1-2 disc height in relation to LDBH. Future research will expand on these findings to develop a more comprehensive decision-supporting model.

Assessing the length of hospital stay for patients with myasthenia gravis based on the data mining MARS approach.

Predicting the length of hospital stay for myasthenia gravis (MG) patients is challenging due to the complex pathogenesis, high clinical variability, and non-linear relationships between variables. Considering the management of MG during hospitalization, it is important to conduct a risk assessment to predict the length of hospital stay. The present study aimed to successfully predict the length of hospital stay for MG based on an expandable data mining technique, multivariate adaptive regression splines (MARS). Data from 196 MG patients' hospitalization were analyzed, and the MARS model was compared with classical multiple linear regression (MLR) and three other machine learning (ML) algorithms. The average hospital stay duration was 12.3 days. The MARS model, leveraging its ability to capture non-linearity, identified four significant factors: disease duration, age at admission, MGFA clinical classification, and daily prednisolone dose. Cut-off points and correlation curves were determined for these risk factors. The MARS model outperformed the MLR and the other ML methods (including least absolute shrinkage and selection operator MLR, classification and regression tree, and random forest) in assessing hospital stay length. This is the first study to utilize data mining methods to explore factors influencing hospital stay in patients with MG. The results highlight the effectiveness of the MARS model in identifying the cut-off points and correlation for risk factors associated with MG hospitalization. Furthermore, a MARS-based formula was developed as a practical tool to assist in the measurement of hospital stay, which can be feasibly supported as an extension of clinical risk assessment.

Machine learning strategy for identifying altered gut microbiomes for diagnostic screening in myasthenia gravis.

Myasthenia gravis (MG) is a neuromuscular junction disease with a complex pathophysiology and clinical variation for which no clear biomarker has been discovered. We hypothesized that because changes in gut microbiome composition often occur in autoimmune diseases, the gut microbiome structures of patients with MG would differ from those without, and supervised machine learning (ML) analysis strategy could be trained using data from gut microbiota for diagnostic screening of MG. Genomic DNA from the stool samples of MG and those without were collected and established a sequencing library by constructing amplicon sequence variants (ASVs) and completing taxonomic classification of each representative DNA sequence. Four ML methods, namely least absolute shrinkage and selection operator, extreme gradient boosting (XGBoost), random forest, and classification and regression trees with nested leave-one-out cross-validation were trained using ASV taxon-based data and full ASV-based data to identify key ASVs in each data set. The results revealed XGBoost to have the best predicted performance. Overlapping key features extracted when XGBoost was trained using the full ASV-based and ASV taxon-based data were identified, and 31 high-importance ASVs (HIASVs) were obtained, assigned importance scores, and ranked. The most significant difference observed was in the abundance of bacteria in the Lachnospiraceae and Ruminococcaceae families. The 31 HIASVs were used to train the XGBoost algorithm to differentiate individuals with and without MG. The model had high diagnostic classification power and could accurately predict and identify patients with MG. In addition, the abundance of Lachnospiraceae was associated with limb weakness severity. In this study, we discovered that the composition of gut microbiomes differed between MG and non-MG subjects. In addition, the proposed XGBoost model trained using 31 HIASVs had the most favorable performance with respect to analyzing gut microbiomes. These HIASVs selected by the ML model may serve as biomarkers for clinical use and mechanistic study in the future. Our proposed ML model can identify several taxonomic markers and effectively discriminate patients with MG from those without with a high accuracy, the ML strategy can be applied as a benchmark to conduct noninvasive screening of MG.

Fusion-Extracted Features by Deep Networks for Improved COVID-19 Classification with Chest X-ray Radiography.

Convolutional neural networks (CNNs) have shown promise in accurately diagnosing coronavirus disease 2019 (COVID-19) and bacterial pneumonia using chest X-ray images. However, determining the optimal feature extraction approach is challenging. This study investigates the use of fusion-extracted features by deep networks to improve the accuracy of COVID-19 and bacterial pneumonia classification with chest X-ray radiography. A Fusion CNN method was developed using five different deep learning models after transferred learning to extract image features (Fusion CNN). The combined features were used to build a support vector machine (SVM) classifier with a RBF kernel. The performance of the model was evaluated using accuracy, Kappa values, recall rate, and precision scores. The Fusion CNN model achieved an accuracy and Kappa value of 0.994 and 0.991, with precision scores for normal, COVID-19, and bacterial groups of 0.991, 0.998, and 0.994, respectively. The results indicate that the Fusion CNN models with the SVM classifier provided reliable and accurate classification performance, with Kappa values no less than 0.990. Using a Fusion CNN approach could be a possible solution to enhance accuracy further. Therefore, the study demonstrates the potential of deep learning and fusion-extracted features for accurate COVID-19 and bacterial pneumonia classification with chest X-ray radiography.

Autonomous Driving Control Based on the Technique of Semantic Segmentation.

Advanced Driver Assistance Systems (ADAS) are only applied to relatively simple scenarios, such as highways. If there is an emergency while driving, the driver should take control of the car to deal properly with the situation at any time. Obviously, this incurs the uncertainty of safety. Recently, in the literature, several studies have been proposed for the above-mentioned issue via Artificial Intelligence (AI). The achievement is exactly the aim that we look forward to, i.e., the autonomous vehicle. In this paper, we realize the autonomous driving control via Deep Reinforcement Learning (DRL) based on the CARLA (Car Learning to Act) simulator. Specifically, we use the ordinary Red-Green-Blue (RGB) camera and semantic segmentation camera to observe the view in front of the vehicle while driving. Then, the captured information is utilized as the input for different DRL models so as to evaluate the performance, where the DRL models include DDPG (Deep Deterministic Policy Gradient) and RDPG (Recurrent Deterministic Policy Gradient). Moreover, we also design an appropriate reward mechanism for these DRL models to realize efficient autonomous driving control. According to the results, only the RDPG strategies can finish the driving mission with the scenario that does not appear/include in the training scenario, and with the help of the semantic segmentation camera, the RDPG control strategy can further improve its efficiency.

Prevalence and Risk Factors of Zoonotic Dermatophyte Infection in Pet Rabbits in Northern Taiwan.

Dermatophytes are the group of keratinophilic fungi that cause superficial cutaneous infection, which traditionally belong to the genera Trichophyton, Microsporum, and Epidermophyton. Dermatophyte infection is not only a threat to the health of small animals, but also an important zoonotic and public health issue because of the potential transmission from animals to humans. Rabbit dermatophytosis is often clinically identified; however, limited information was found in Asia. The aims of this study are to investigate the prevalence and to evaluate the risk factors of dermatophytosis in pet rabbits in Northern Taiwan. Between March 2016 and October 2018, dander samples of pet rabbits were collected for fungal infection examination by Wood's lamp, microscopic examination (KOH preparation), fungal culture, and PCR assay (molecular identification). Z test and Fisher's exact test were performed to evaluate the potential risk factors, and logistic regression analysis was then performed to build the model of risk factors related to dermatophyte infection. Of the collected 250 dander samples of pet rabbits, 29 (11.6%) samples were positive for dermatophytes by molecular identification. In those samples, 28 samples were identified as the T. mentagrophytes complex and 1 sample was identified as M. canis. Based on the results of the Firth's bias reduction logistic analyses, animal source (rabbits purchased from pet shops) and number of rearing rabbits (three rabbits or more) were shown as the main risks for dermatophyte infection in the pet rabbits in Taiwan. The results of the present study elucidate the prevalence of rabbit dermatophyte infection, pathogens, and risk factors in Taiwan, and provide an important reference for the prevention and control of rabbit dermatophytosis.

Utilization of Decision Tree Algorithms for Supporting the Prediction of Intensive Care Unit Admission of Myasthenia Gravis: A Machine Learning-Based Approach.

Myasthenia gravis (MG), an acquired autoimmune-related neuromuscular disorder that causes muscle weakness, presents with varying severity, including myasthenic crisis (MC). Although MC can cause significant morbidity and mortality, specialized neuro-intensive care can produce a good long-term prognosis. Considering the outcomes of MG during hospitalization, it is critical to conduct risk assessments to predict the need for intensive care. Evidence and valid tools for the screening of critical patients with MG are lacking. We used three machine learning-based decision tree algorithms, including a classification and regression tree, C4.5, and C5.0, for predicting intensive care unit (ICU) admission of patients with MG. We included 228 MG patients admitted between 2015 and 2018. Among them, 88.2% were anti-acetylcholine receptors antibody positive and 4.7% were anti-muscle-specific kinase antibody positive. Twenty clinical variables were used as predictive variables. The C5.0 decision tree outperformed the other two decision tree and logistic regression models. The decision rules constructed by the best C5.0 model showed that the Myasthenia Gravis Foundation of America clinical classification at admission, thymoma history, azathioprine treatment history, disease duration, sex, and onset age were significant risk factors for the development of decision rules for ICU admission prediction. The developed machine learning-based decision tree can be a supportive tool for alerting clinicians regarding patients with MG who require intensive care, thereby improving the quality of care.

Assessment of Sarcopenia and Obesity in Patients with Myasthenia Gravis Using Dual-Energy X-ray Absorptiometry: A Cross-Sectional Study.

Sarcopenia and obesity can negatively impact quality of life and cause chronic fragility, and are associated with neuromuscular diseases, including myasthenia gravis (MG). The long-term consequences of body composition changes in chronic MG remain unknown; we therefore evaluated changes in body composition, including sarcopenia, obesity, lean body mass, and the prevalence of sarcopenic obesity in patients. In this cross-sectional study, 35 patients with MG (mean age: 56.1 years) and 175 matched controls were enrolled. Body fat mass and skeletal muscle mass were measured using whole body dual-energy X-ray absorptiometry. Patients with MG exhibited a higher prevalence of obesity and higher android adiposity and total body fat percentage than those of controls. Although the prevalence of sarcopenia and sarcopenic obesity did not increase with age, there was a decrease in arm and android muscle mass in patients with MG compared with controls. Lower muscle mass percentages were correlated with increased age and MG severity, but not with corticosteroid use. Thus, MG is associated with increased risk for obesity and decreased muscle mass with aging, regardless of corticosteroid use. Therefore, accurate diagnosis of body composition changes in MG could facilitate the application of appropriate therapies to promote health, improve quality of life, and prevent fragility.

Clinical Predictors of Prolonged Hospital Stay in Patients with Myasthenia Gravis: A Study Using Machine Learning Algorithms.

Myasthenia gravis (MG) is an autoimmune disorder that causes muscle weakness. Although the management is well established, some patients are refractory and require prolonged hospitalization. Our study is aimed to identify the important factors that predict the duration of hospitalization in patients with MG by using machine learning methods. A total of 21 factors were chosen for machine learning analyses. We retrospectively reviewed the data of patients with MG who were admitted to hospital. Five machine learning methods, including stochastic gradient boosting (SGB), least absolute shrinkage and selection operator (Lasso), ridge regression (Ridge), eXtreme gradient boosting (XGboost), and gradient boosting with categorical features support (Catboost), were used to construct models for identify the important factors affecting the duration of hospital stay. A total of 232 data points of 204 hospitalized MG patients admitted were enrolled into the study. The MGFA classification, treatment of high-dose intravenous corticosteroid, age at admission, treatment with intravenous immunoglobulins, and thymoma were the top five significant variables affecting prolonged hospitalization. Our findings from machine learning will provide physicians with information to evaluate the potential risk of MG patients having prolonged hospital stay. The use of high-dose corticosteroids is associated with prolonged hospital stay and to be used cautiously in MG patients.

Changes in Physical Fitness and Body Composition Associated with Physical Exercise in Patients with Myasthenia Gravis: A Longitudinal Prospective Study.

There is a lack of guidelines for physical exercise in patients with myasthenia gravis (MG). A few pilot studies have shown that exercise can be safely applied to patients with MG. However, how physical exercise affects body composition, disease function, and disease severity remains unknown. In this prospective study, we enrolled 34 patients with MG with stable condition and evaluated the disease severity, physical fitness parameters, and body composition (measured using whole-body dual-energy X-ray absorptiometry (DXA)), before and after conducting a 24-week physical exercise regimen of aerobic and resistance strength training. The outcomes were measured by DXA, quantitative MG (QMG) score, quality of life score, handgrip strength and walking speed. During the training regimen, participants were free to decide how many exercise sessions per week and regularly reported their weekly exercise time. The physical exercise program was well tolerated by the participants, the parameters of the QMG score and handgrip strength improved, and participants' body composition did not change significantly. The high exercise group experienced greater deterioration in muscle mass in the arms, but exhibited a greater improvement in forced vital capacity, walking speed, and symptom severity. The group with low QMG scores improved more in terms of physical fitness, including walking speed. These findings indicate that physical exercise is well tolerated by patients with MG, and is accompanied by improved muscular and physical functions. We propose that physical exercise is safe, effective, and appropriate for patients with well-regulated MG.

Emboli stroke following migration of carotid foreign body: A case report.

Foreign body embolization can cause intracranial artery occlusion with ischemic stroke. Reported etiologies include post cerebrovascular interventions, migration of esophageal foreign body and neck trauma. We reported a case with punctured wound at left neck, X-ray and computed tomography revealed a foreign body located in the carotid region. The patient eventually developed stroke symptoms in the next day after operation. Non-contrast brain Computer tomography at that time revealed that porcelain fragment located at the suprasellar area, and infarction of the left anterior basal ganglion. Our patient is the first reported case having an embolic stroke secondary to distal migration of a foreign body from the carotid artery after neck trauma. We call attention to this rare neurologic complication of neck trauma with foreign body retention. Appropriate and prompt identification of concurrent vascular injuries with retention of foreign body is strongly advised in neck trauma patients.

Nondestructive characterization of bone tissue scaffolds for clinical scenarios.

OBJECTIVES: This study aimed to develop a simple and efficient numerical modeling approach for characterizing strain and total strain energy in bone scaffolds implanted in patient-specific anatomical sites. MATERIALS AND METHODS: A simplified homogenization technique was developed to substitute a detailed scaffold model with the same size and equivalent orthotropic material properties. The effectiveness of the proposed modeling approach was compared with two other common homogenization methods based on periodic boundary conditions and the Hills-energy theorem. Moreover, experimental digital image correlation (DIC) measurements of full-field surface strain were conducted to validate the numerical results. RESULTS: The newly proposed simplified homogenization approach allowed for fairly accurate prediction of strain and total strain energy in tissue scaffolds implanted in a large femur mid-shaft bone defect subjected to a simulated in-vivo loading condition. The maximum discrepancy between the total strain energy obtained from the simplified homogenization approach and the one obtained from detailed porous scaffolds was 8.8%. Moreover, the proposed modeling technique could significantly reduce the computational cost (by about 300 times) required for simulating an in-vivo bone scaffolding scenario as the required degrees of freedom (DoF) was reduced from about 26 million for a detailed porous scaffold to only 90,000 for the homogenized solid counterpart in the analysis. CONCLUSIONS: The simplified homogenization approach has been validated by correlation with the experimental DIC measurements. It is fairly efficient and comparable with some other common homogenization techniques in terms of accuracy. The proposed method is implicating to different clinical applications, such as the optimal selection of patient-specific fixation plates and screw system.

10-Point CT-ASPECTS-based reperfusion therapy for unknown onset stroke.

Under the time-based criteria, patients with unknown onset stroke (UOS) are ineligible for reperfusion therapies. However, previous studies suggest that some patients with UOS may benefit from reperfusion. Several imaging modalities have been suggested to select patients for intervention, but the optimal imaging criteria are still controversial. Herein we present a series of four cases using 10-point CT-ASPECTS to support our decision of reperfusion therapy. We decided based on history, symptoms, and the 10-point CT-ASPECTS alone. Each patient's history suggested that the stroke just took place. All four patients had apparent clinical symptoms, with 10-point CT-ASPECTS. All of them had a reduction in their NIHSS after the reperfusion therapy. 10-point CT-ASPECTS could be used to support the presumption that the stroke just happens in patients with UOS. Further study is warranted to elucidate the value of CT-ASPECTS for UOS patients.

Micro-CT based modelling for characterising injection-moulded porous titanium implants.

Design of prosthetic implants to ensure rapid and stable osseointegration remains a significant challenge, and continuous efforts have been directed to new implant materials, structures and morphology. This paper aims to develop and characterise a porous titanium dental implant fabricated by metallic powder injection-moulding. The surface morphology of the specimens was first examined with a scanning electron microscope (SEM), followed by microscopic computerised tomography (µ-CT) scanning to capture its 3D microscopic features non-destructively. The nature of porosity and pore sizes were determined statistically. A homogenisation technique based on the Hills-energy theorem was adopted to evaluate its directional elastic moduli, and the conservation of mass theorem was employed to quantify the oxygen diffusivity for bio-transportation feature. This porous medium was found to have pore sizes varying from 50 to 400 µm and the average porosity of 46.90 ± 1.83%. The anisotropic principal elastic moduli were found fairly close to the upper range of cortical bone, and the directional diffusivities could potentially enable radial osseous tissue ingrowth and vascularisation. This porous titanium successfully reduces the elastic modulus mismatch between implant and bone for dental and orthopaedic applications, and provides improved capacity for transporting oxygen, nutrient and waste for pre-vascular network formation. Copyright © 2016 John Wiley & Sons, Ltd.

Bioinspired lightweight cellular materials--understanding effects of natural variation on mechanical properties.

Cuttlebone is a natural marine cellular material possessing the exceptional mechanical properties of high compressive strength, high porosity and high permeability. This combination of properties is exceedingly desirable in biomedical applications, such as bone tissue scaffolds. In light of recent studies, which converted raw cuttlebone into hydroxyapatite tissue scaffolds, the impact of morphological variations in the microstructure of this natural cellular material on the effective mechanical properties is explored in this paper. Two extensions of the finite element-based homogenization method are employed to account for deviations from the assumption of periodicity. Firstly, a representative volume element (RVE) of cuttlebone is systematically varied to reflect the large range of microstructural configurations possibly among different cuttlefish species. The homogenization results reveal the critical importance of pillar formation and aspect ratio (height/width of RVE) on the effective bulk and shear moduli of cuttlebone. Secondly, multi-cell analysis domains (or multiple RVE domains) permit the introduction of random variations across neighboring cells. Such random variations decrease the bulk modulus whilst displaying minimal impact on the shear modulus. Increasing the average size of random variations increases the effect on bulk modulus. Also, the results converge rapidly as the size of the analysis domain is increased, meaning that a relatively small multi-cell domain can provide a reasonable approximation of the effective properties for a given set of random variation parameters. These results have important implications for the proposed use of raw cuttlebone as an engineering material. They also highlight some potential for biomimetic design capabilities for materials inspired by the cuttlebone microstructure, which may be applicable in biomedical applications such as bone tissue scaffolds.

A fully coherent electron beam from a noble-metal covered W(111) single-atom emitter.

In quantum mechanics, a wavefunction contains two factors: the amplitude and the phase. Only when the probing beam is fully phase coherent, can complete information be retrieved from a particle beam based experiment. Here we use the electron beam field emitted from a noble-metal covered W(111) single-atom tip to image single-walled carbon nanotubes (SWNTs) in an electron point projection microscope (PPM). The interference fringes of an SWNT bundle exhibit a very high contrast and the fringe pattern extends throughout the entire beam width. This indicates good phase correlation at all points transverse to the propagation direction. Application of these sources can significantly improve the performance and expand the capabilities of current electron beam based techniques. New instrumentation based on the full spatial coherence may allow determination of the three-dimensional atomic structures of nonperiodic nanostructures and make many advanced experiments possible.