Preoperative differentiation of complicated and uncomplicated appendicitis is challenging. The research goal was to construct a new intelligent diagnostic rule that is accurate, fast, noninvasive, and cost-effective, distinguishing between complicated and uncomplicated appendicitis. Overall, 298 patients with acute appendicitis from the Wenzhou Central Hospital were recruited, and information on their demographic characteristics, clinical findings, and laboratory data was retrospectively reviewed and applied in this study. First, the most significant variables, including C-reactive protein (CRP), heart rate, body temperature, and neutrophils discriminating complicated from uncomplicated appendicitis, were identified using random forest analysis. Second, an improved grasshopper optimization algorithm-based support vector machine was used to construct the diagnostic model to discriminate complicated appendicitis (CAP) from uncomplicated appendicitis (UAP). The resultant optimal model can produce an average of 83.56% accuracy, 81.71% sensitivity, 85.33% specificity, and 0.6732 Matthews correlation coefficients. Based on existing routinely available markers, the proposed intelligent diagnosis model is highly reliable. Thus, the model can potentially be used to assist doctors in making correct clinical decisions.

Alzheimer's disease (AD) is a severe neurodegenerative disease with slow course of onset and deterioration with time. With the speedup of global aging, AD has become a disease that seriously threatens the physical health of the elderly; therefore, the effective prevention and treatments of AD is an extremely important area of study for researchers and clinicians. Rapid technological developments have promoted the analysis of various kinds of complex data sets using machine learning methods. The common machine learning algorithms, such as Lasso, SVM and Random Forest, are very important in AD research. To help accelerate AD-related research, we review some recent research progress on Alzheimer's disease, including database, image analysis, gene expression, etc., which can provide AD researchers with more comprehensive research methods.

In recent years, colorectal cancer has become one of the most significant diseases that endanger human health. Deep learning methods are increasingly important for the classification of colorectal histopathology images. However, existing approaches focus more on end-to-end automatic classification using computers rather than human-computer interaction. In this paper, we propose an IL-MCAM framework. It is based on attention mechanisms and interactive learning. The proposed IL-MCAM framework includes two stages: automatic learning (AL) and interactivity learning (IL). In the AL stage, a multi-channel attention mechanism model containing three different attention mechanism channels and convolutional neural networks is used to extract multi-channel features for classification. In the IL stage, the proposed IL-MCAM framework continuously adds misclassified images to the training set in an interactive approach, which improves the classification ability of the MCAM model. We carried out a comparison experiment on our dataset and an extended experiment on the HE-NCT-CRC-100K dataset to verify the performance of the proposed IL-MCAM framework, achieving classification accuracies of 98.98% and 99.77%, respectively. In addition, we conducted an ablation experiment and an interchangeability experiment to verify the ability and interchangeability of the three channels. The experimental results show that the proposed IL-MCAM framework has excellent performance in the colorectal histopathological image classification tasks.

Correct rider oscillation and position are the basics for a good horseback riding performance. In this paper, we propose a framework for the automatic analysis of athletes behaviour based on cluster analysis. Two groups of athletes (riders vs non-riders) were assigned to a horseback riding simulator exercise. The participants exercised four different incremental horse oscillation frequencies. This paper studies the postural coordination, by computing the different discrete relative phases of head-horse, elbow-horse and trunk-horse oscillations. Two clustering algorithms are then applied to automatically identify the change of rider and non-rider behaviour in terms of postural coordination. The results showed that the postural coordination was influenced by the level of rider expertise. More diverse behaviour was observed for non-riders. At the opposite, riders produced lower postural displacements and deployed more efficient postural control. The postural coordination for both groups was also influenced by the oscillation frequencies.

PURPOSE: Absorbed dose calculation in magnetic resonance-guided radiation therapy (MRgRT) is commonly based on pseudo CT (pCT) images. This study investigated the feasibility of unsupervised pCT generation from MRI using a cycle generative adversarial network (CycleGAN) and a heterogenous multicentric dataset. A dosimetric analysis in three-dimensional conformal radiotherapy (3DCRT) planning was also performed. MATERIAL AND METHODS: Overall, 87 T1-weighted and 102 T2-weighted MR images alongside with their corresponding computed tomography (CT) images of brain cancer patients from multiple centers were used. Initially, images underwent a number of preprocessing steps, including rigid registration, novel CT Masker, N4 bias field correction, resampling, resizing, and rescaling. To overcome the gradient vanishing problem, residual blocks and mean squared error (MSE) loss function were utilized in the generator and in both networks (generator and discriminator), respectively. The CycleGAN was trained and validated using 70 T1 and 80 T2 randomly selected patients in an unsupervised manner. The remaining patients were used as a holdout test set to report final evaluation metrics. The generated pCTs were validated in the context of 3DCRT. RESULTS: The CycleGAN model using masked T2 images achieved better performance with a mean absolute error (MAE) of 61.87 ± 22.58 HU, peak signal to noise ratio (PSNR) of 27.05 ± 2.25 (dB), and structural similarity index metric (SSIM) of 0.84 ± 0.05 on the test dataset. T1-weighted MR images used for dosimetric assessment revealed a gamma index of 3%, 3 mm, 2%, 2 mm and 1%, 1 mm with acceptance criteria of 98.96% ± 1.1%, 95% ± 3.68%, 90.1% ± 6.05%, respectively. The DVH differences between CTs and pCTs were within 2%. CONCLUSIONS: A promising pCT generation model capable of handling heterogenous multicenteric datasets was proposed. All MR sequences performed competitively with no significant difference in pCT generation. The proposed CT Masker proved promising in improving the model accuracy and robustness. There was no significant difference between using T1-weighted and T2-weighted MR images for pCT generation.

Biomedical image segmentation is essential for computerized medical image analysis. Deep learning algorithms allow us to design state-of-the-art models for solving segmentation problems. The U-Net and its variants have provided positive results across various datasets. However, the existing networks have the same receptive field at each level and the models are supervised only at the shallow level. Considering these two ideas, we have proposed the D3MSU-Net where the field of view in each level is varied depending upon the depth of the resolution layer and the model is supervised at each resolution level. We have evaluated our network in eight benchmark datasets such as Electron Microscopy, Lung segmentation, Montgomery Chest X-ray, Covid-Radiopaedia, Wound, Medetec, Brain MRI, and Covid-19 lung CT dataset. Additionally, we have provided the performance for various ablations. The experimental results show the superiority of the proposed network. The proposed D3MSU-Net and ablation models are available at www.github.com/shirshabose/D3MSUNET.

Our motivation is to enable non-biomechanical engineering specialists to use sophisticated biomechanical models in the clinic to predict tumour resection-induced brain shift, and subsequently know the location of the residual tumour and its boundary. To achieve this goal, we developed a framework for automatically generating and solving patient-specific biomechanical models of the brain. This framework automatically determines patient-specific brain geometry from MRI data, generates patient-specific computational grid, assigns material properties, defines boundary conditions, applies external loads to the anatomical structures, and solves differential equations of nonlinear elasticity using Meshless Total Lagrangian Explicit Dynamics (MTLED) algorithm. We demonstrated the effectiveness and appropriateness of our framework on real clinical cases of tumour resection-induced brain shift.

Numerous serological detection kits are being rapidly developed and approved for screening and diagnosing suspected coronavirus disease 2019 (COVID-19) cases. However, cross-reactivity between pre-existing antibodies against other coronaviruses and the captured antigens in these kits can affect detection accuracy, emphasizing the necessity for identifying highly specific antigen fragments for antibody detection. Thus, we performed a conservation and specificity analysis of the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) nucleocapsid (N) protein. We also integrated various B-cell epitope prediction methods to obtain possible dominant epitope regions for the N protein, analyzed the differences in serological antibody levels for different epitopes using ELISA, and identified N protein epitopes for IgG and IgM with high-specificity. The SARS-CoV-2 N protein showed low mutation rates and shared the highest amino acid similarity with SARS-CoV; however, it differed substantially from other coronaviruses. Tests targeting the SARS-CoV-2 N protein produce strong positive results in patients recovering from SARS-CoV. The N18-39 and N183-197 epitopes for IgG and IgM detection, respectively, can effectively overcome cross-reactivity, and even exhibit good specificity between SARS-CoV-2 and SARS-CoV. The antibody levels detected with these were consistent with those detected using the complete N protein. These findings provide a basis for serological diagnosis and determining the kinetics of SARS-CoV-2 antibody detection in patients.

COVID-19 is a fast-spreading pandemic, and early detection is crucial for stopping the spread of infection. Lung images are used in the detection of coronavirus infection. Chest X-ray (CXR) and computed tomography (CT) images are available for the detection of COVID-19. Deep learning methods have been proven efficient and better performing in many computer vision and medical imaging applications. In the rise of the COVID pandemic, researchers are using deep learning methods to detect coronavirus infection in lung images. In this paper, the currently available deep learning methods that are used to detect coronavirus infection in lung images are surveyed. The available methodologies, public datasets, datasets that are used by each method and evaluation metrics are summarized in this paper to help future researchers. The evaluation metrics that are used by the methods are comprehensively compared.

BACKGROUND: Recently, the value of natural products has been extensively considered because these resources can potentially be applied to prevent and treat coronavirus pneumonia 2019 (COVID-19). However, the discovery of nature drugs is problematic because of their complex composition and active mechanisms. METHODS: This comprehensive study was performed on flavonoids, which are compounds with anti-inflammatory and antiviral effects, to show drug discovery and active mechanism from natural products in the treatment of COVID-19 via a systems pharmacological model. First, a chemical library of 255 potential flavonoids was constructed. Second, the pharmacodynamic basis and mechanism of action between flavonoids and COVID-19 were explored by constructing a compound-target and target-disease network, targets protein-protein interaction (PPI), MCODE analysis, gene ontology (GO), and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment. RESULTS: In total, 105 active flavonoid components were identified, of which 6 were major candidate compounds (quercetin, epigallocatechin-3-gallate (EGCG), luteolin, fisetin, wogonin, and licochalcone A). 152 associated targets were yielded based on network construction, and 7 family proteins (PTGS, GSK3β, ABC, NOS, EGFR, and IL) were included as central hub targets. Moreover, 528 GO items and 178 KEGG pathways were selected through enrichment of target functions. Lastly, molecular docking demonstrated good stability of the combination of selected flavonoids with 3CL Pro and ACEⅡ. CONCLUSION: Natural flavonoids could enable resistance against COVID-19 by regulating inflammatory, antiviral, and immune responses, and repairing tissue injury. This study has scientific significance for the selective utilization of natural products, medicinal value enhancement of flavonoids, and drug screening for the treatment of COVID-19 induced by SARS-COV-2.

Supervised learning-based medical image segmentation solutions usually require sufficient labeled training data. Insufficient available labeled training data often leads to the limitations of model performances, such as over-fitting, low accuracy, and poor generalization ability. However, this dilemma may worsen in the field of medical image analysis. Medical image annotation is usually labor-intensive and professional work. In this work, we propose a novel shape and boundary-aware deep learning model for medical image segmentation based on semi-supervised learning. The model makes good use of labeled data and also enables unlabeled data to be well applied by using task consistency loss. Firstly, we adopt V-Net for Pixel-wise Segmentation Map (PSM) prediction and Signed Distance Map (SDM) regression. In addition, we multiply multi-scale features, extracted by Pyramid Pooling Module (PPM) from input X, with 2 - |SDM| to enhance the features around the boundary of the segmented target, and then feed them into the Feature Fusion Module (FFM) for fine segmentation. Besides boundary loss, the high-level semantics implied in SDM facilitate the accurate segmentation of boundary regions. Finally, we get the ultimate result by fusing coarse and boundary-enhanced features. Last but not least, to mine unlabeled training data, we impose consistency constraints on the three core outputs of the model, namely PSM1, SDM, and PSM3. Through extensive experiments over three representative but challenging medical image datasets (LA2018, BraTS2019, and ISIC2018) and comparisons with the existing representative methods, we validate the practicability and superiority of our model.

High-throughput technologies produce gene expression time-series data that need fast and specialized algorithms to be processed. While current methods already deal with different aspects, such as the non-stationarity of the process and the temporal correlation, they often fail to take into account the pairing among replicates. We propose PairGP, a non-stationary Gaussian process method to compare gene expression time-series across several conditions that can account for paired longitudinal study designs and can identify groups of conditions that have different gene expression dynamics. We demonstrate the method on both simulated data and previously unpublished RNA sequencing (RNA-seq) time-series with five conditions. The results show the advantage of modeling the pairing effect to better identify groups of conditions with different dynamics. The pairing effect model displays good capabilities of selecting the most probable grouping of conditions even in the presence of a high number of conditions. The developed method is of general application and can be applied to any gene expression time series dataset. The model can identify common replicate effects among the samples coming from the same biological replicates and model those as separate components. Learning the pairing effect as a separate component, not only allows us to exclude it from the model to get better estimates of the condition effects, but also to improve the precision of the model selection process. The pairing effect that was accounted before as noise, is now identified as a separate component, resulting in more accurate and explanatory models of the data.

The global pandemic caused by a single-stranded RNA (ssRNA) virus known as the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is still at its peak, with new cases being reported daily. Although the vaccines have been administered on a massive scale, the frequent mutations in the viral gene and resilience of the future strains could be more problematic. Therefore, new compounds are always needed to be available for therapeutic approaches. We carried out the present study to discover potential drug compounds against the SARS-CoV-2 main protease (Mpro). A total of 16,000 drug-like small molecules from the ChemBridge database were virtually screened to obtain the top hits. As a result, 1032 hits were selected based on their docking scores. Next, these structures were prepared for molecular docking, and each small molecule was docked into the active site of the Mpro. Only compounds with solid interactions with the active site residues and the highest docking score were subjected to molecular dynamics (MD) simulation. The post-simulation analyses were carried out using the in-built GROMACS tools to gauge the stability, flexibility, and compactness. Principal component analysis (PCA) and hydrogen bonding were also calculated to observe trends and affinity of the drugs towards the target. Among the five top compounds, C1, C3, and C6 revealed strong interaction with the target's active site and remained highly stable throughout the simulation. We believe the predicted compounds in this study could be potential inhibitors in the natural system and can be utilized in designing therapeutic strategies against the SARS-CoV-2.

Cancer is the second deadliest disease globally that can affect any human body organ. Early detection of cancer can increase the chances of survival in humans. Morphometric appearances of histopathology images make it difficult to segment nuclei effectively. We proposed a model to segment overlapped nuclei from H&E stained images. U-Net model achieved state-of-the-art performance in many medical image segmentation tasks; however, we modified the U-Net to learn a distinct set of consistent features. In this paper, we proposed the DenseRes-Unet model by integrating dense blocks in the last layers of the encoder block of U-Net, focused on relevant features from previous layers of the model. Moreover, we take advantage of residual connections with Atrous blocks instead of conventional skip connections, which helps to reduce the semantic gap between encoder and decoder paths. The distance map and binary threshold techniques intensify the nuclei interior and contour information in the images, respectively. The distance map is used to detect the center point of nuclei; moreover, it differentiates among nuclei interior boundary and core area. The distance map lacks a contour problem, which is resolved by using a binary threshold. Binary threshold helps to enhance the pixels around nuclei. Afterward, we fed images into the proposed DenseRes-Unet model, a deep, fully convolutional network to segment nuclei in the images. We have evaluated our model on four publicly available datasets for Nuclei segmentation to validate the model's performance. Our proposed model achieves 89.77% accuracy 90.36% F1-score, and 78.61% Aggregated Jaccard Index (AJI) on Multi organ Nucleus Segmentation (MoNuSeg).

Asymptomatic transmission of the coronavirus disease and the infected individual prediction has become very important in the COVID-19 outbreak study. The asymptomatic and symptomatic transmission studies are still ongoing to assess their impacts on disease monitoring and burden. However, there has been limited research on how asymptomatic and symptomatic transmissions together can affect the coronavirus disease outbreak. A mathematical model is therefore needed to be developed in order to assess the effect of these transmissions on the coronavirus disease dynamics. This paper develops a mathematical model concerning asymptomatic and symptomatic disease transmission processes in the COVID-19 outbreak. The model sensitivity has been analysed in terms of the variance of each parameter, and the local stability at two equilibrium points have been discussed in terms of the basic reproduction number (R0). It is found that the disease-free equilibrium gets stable for R0 < 1 whereas the endemic equilibrium becomes stable for R0 > 1 and unstable otherwise. The proportion of the effect of asymptomatic and symptomatic transmission rates on R0 is calculated to be approximately between 1 and 3. The results demonstrate that asymptomatic transmission has a significant impact compared to symptomatic transmission in the disease outbreak. Outcomes of this study will contribute to setting an effective control strategy for the COVID-19 outbreak.

Deep learning-based computer-aided diagnosis techniques have demonstrated encouraging performance in endoscopic lesion identification and detection, and have reduced the rate of missed and false detections of disease during endoscopy. However, the interpretability of the model-based results has not been adequately addressed by existing methods. This phenomenon is directly manifested by a significant bias in the representation of feature localization. Good recognition models experience severe feature localization errors, particularly for lesions with subtle morphological features, and such unsatisfactory performance hinders the clinical deployment of models. To effectively alleviate this problem, we proposed a solution to optimize the localization bias in feature representations of cancer-related recognition models that is difficult to accurately label and identify in clinical practice. Optimization was performed in the training phase of the model through the proposed data augmentation method and auxiliary loss function based on clinical priors. The data augmentation method, called partial jigsaw, can "break" the spatial structure of lesion-independent image blocks and enrich the data feature space to decouple the interference of background features on the space and focus on fine-grained lesion features. The annotation-based auxiliary loss function used class activation maps for sample distribution correction and led the model to present localization representation converging on the gold standard annotation of visualization maps. The results show that with the improvement of our method, the precision of model recognition reached an average of 92.79%, an F1-score of 92.61%, and accuracy of 95.56% based on a dataset constructed from 23 hospitals. In addition, we quantified the evaluation representation of visualization feature maps. The improved model yielded significant offset correction results for visualized feature maps compared with the baseline model. The average visualization-weighted positive coverage improved from 51.85% to 83.76%. The proposed approach did not change the deployment capability and inference speed of the original model and can be incorporated into any state-of-the-art neural network. It also shows the potential to provide more accurate localization inference results and assist in clinical examinations during endoscopies.

Central aortic blood pressure (CABP) is a better predictor for cardiovascular events than brachial blood pressure. However, direct CABP measurement is invasive. The objective of this paper is to develop an ultrasound-based method using individualized Windkessel (WK) models for non-invasive estimation of CABP. Three WK models (with two-, three- and four-element WK, named, WK2, WK3 and WK4, respectively) were created and the model parameters were individualized based on aortic flow velocity and diameter waveforms measured by ultrasound (US). Experimental data were acquired in 42 subjects aged 21-67 years. The CABP estimated by WK models was compared with the reference CABP obtained using a commercial system. The results showed that the overall performance of the WK3 and WK4 models was similar, outperforming the WK2 model. The estimated CABP based on WK3/WK4 model showed good agreement with the reference CABP: the absolute errors of systolic blood pressure (SBP), 2.4 ± 2.1/2.4 ± 2.0 mmHg; diastolic blood pressure (DBP), 1.4 ± 1.1/1.7 ± 1.5 mmHg; mean blood pressure (MBP), 1.3 ± 0.8/1.3 ± 0.8 mmHg; pulse pressure (PP), 3.0 ± 2.3/3.2 ± 2.6 mmHg; the root mean square error (RMSE) of the waveforms, 2.5 ± 1.0/2.6 ± 1.1 mmHg. Therefore, the proposed method can provide a non-invasive CABP estimation during routine cardiac US examination.

Identifying motor and mental imagery electroencephalography (EEG) signals is imperative to realizing automated, robust brain-computer interface (BCI) systems. In the present study, we proposed a pretrained convolutional neural network (CNN)-based new automated framework feasible for robust BCI systems with small and ample samples of motor and mental imagery EEG training data. The framework is explored by investigating the implications of different limiting factors, such as learning rates and optimizers, processed versus unprocessed scalograms, and features derived from untuned pretrained models in small, medium, and large pretrained CNN models. The experiments were performed on three public datasets obtained from BCI Competition III. The datasets were denoised with multiscale principal component analysis, and time-frequency scalograms were obtained by employing a continuous wavelet transform. The scalograms were fed into several variants of ten pretrained models for feature extraction and identification of different EEG tasks. The experimental results showed that ShuffleNet yielded the maximum average classification accuracy of 99.52% using an RMSProp optimizer with a learning rate of 0.000 1. It was observed that low learning rates converge to more optimal performances compared to high learning rates. Moreover, noisy scalograms and features extracted from untuned networks resulted in slightly lower performance than denoised scalograms and tuned networks, respectively. The overall results suggest that pretrained models are robust when identifying EEG signals because of their ability to preserve the time-frequency structure of EEG signals and promising classification outcomes.

Diffusion magnetic resonance imaging (dMRI) is being extensively used to study the neural architecture of the brain. High angular resolution diffusion imaging (HARDI), a variant of diffusion MRI, measures the diffusion of water molecules along the angular gradient directions in the q-space. It provides better estimates of fiber orientations compared to the traditionally used diffusion tensor imaging (DTI). However, HARDI requires acquisition of relatively large number of samples leading to longer scanning times. Several approaches based on compressive sensing (CS) have been proposed to accelerate HARDI acquisition, leveraging on the sparse representation of the HARDI signal in a pre-specified sparsifying basis. In this paper, we propose to carry out reconstruction of compressively sensed HARDI data using an adaptively learned transform. The transform is learned (i) from the compressive measurements on-the-fly, thereby, eliminating the overhead of choosing fixed sparsifying transforms, and (ii) on overlapping patches of the data, thereby, capturing local image structure effectively. Experiments are conducted on multiple real HARDI data for varying sampling ratios and sampling schemes. The performance of the proposed "TL-HARDI" method is compared with the state-of-the-art methods on various known image quality metrics as well as on dMRI feature maps derived from the reconstructed images. The proposed method is observed to yield better reconstruction than the existing state-of-the-art methods in both quantitative and qualitative comparisons.

OBJECTIVE: To investigate the ability of our convolutional neural network (CNN) to predict axillary lymph node metastasis using primary breast cancer ultrasound (US) images. METHODS: In this IRB-approved study, 338 US images (two orthogonal images) from 169 patients from 1/2014-12/2016 were used. Suspicious lymph nodes were seen on US and patients subsequently underwent core-biopsy. 64 patients had metastatic lymph nodes. A custom CNN was utilized on 248 US images from 124 patients in the training dataset and tested on 90 US images from 45 patients. The CNN was implemented entirely of 3 × 3 convolutional kernels and linear layers. The 9 convolutional kernels consisted of 6 residual layers, totaling 12 convolutional layers. Feature maps were down-sampled using strided convolutions. Dropout with a 0.5 keep probability and L2 normalization was utilized. Training was implemented by using the Adam optimizer and a final SoftMax score threshold of 0.5 from the average of raw logits from each pixel was used for two class classification (metastasis or not). RESULTS: Our CNN achieved an AUC of 0.72 (SD ± 0.08) in predicting axillary lymph node metastasis from US images in the testing dataset. The model had an accuracy of 72.6% (SD ± 8.4) with a sensitivity and specificity of 65.5% (SD ± 28.6) and 78.9% (SD ± 15.1) respectively. Our algorithm is available to be shared for research use. (https://github.com/stmutasa/MetUS). CONCLUSION: It's feasible to predict axillary lymph node metastasis from US images using a deep learning technique. This can potentially aid nodal staging in patients with breast cancer.

INTRODUCTION: Computational modeling can enhance the understanding of cell mechanics. To achieve this, finite element models of endothelial cells were proposed with shapes mimicking their natural state inside the endothelium within the cardiovascular system. Implementing the recently proposed bendo-tensegrity concept, these models consider flexural (buckling) as well as tensional/compressional behavior of microtubules and also incorporate the waviness of intermediate filaments. MATERIALS AND METHODS: Four different models were created (flat and domed hexagons, both regular and elongated in the direction of blood flow) and loaded by biaxial deformation, blood pressure, and shear load from blood flow - natural physiological conditions of the arterial endothelium - aiming to investigate the "in situ" mechanical response of the cell. RESULTS: The impact of individual components of loads on the nucleus deformation (more specifically on the first principal strain) potentially influencing mechanotransduction was investigated and the role of the cytoskeleton and its constituents in the mechanical response of the endothelial cell was assessed. The results show (i) the impact of pulsating blood pressure on cyclic deformations of the nucleus, which increase substantially with decreasing axial pre-stretch of the cell, (ii) the importance of relatively low shear stresses in the cell response and nucleus deformation. CONCLUSION: Not only the pulsatile blood pressure but also the wall shear stress may induce significant deformation of the nucleus and thus trigger remodelation processes in endothelial cells.

The user does not have any idea about the credibility of outcomes from deep neural networks (DNN) when uncertainty quantification (UQ) is not employed. However, current Deep UQ classification models capture mostly epistemic uncertainty. Therefore, this paper aims to propose an aleatory-aware Deep UQ method for classification problems. First, we train DNNs through transfer learning and collect numeric output posteriors for all training samples instead of logical outputs. Then we determine the probability of happening a certain class from K-nearest output posteriors of the same DNN in training samples. We name this probability as opacity score, as the paper focuses on the detection of opacity on X-ray images. This score reflects the level of aleatory on the sample. When the NN is certain on the classification of the sample, the probability of happening a class becomes much higher than the probabilities of others. Probabilities for different classes become close to each other for a highly uncertain classification outcome. To capture the epistemic uncertainty, we train multiple DNNs with different random initializations, model selection, and augmentations to observe the effect of these training parameters on prediction and uncertainty. To reduce execution time, we first obtain features from the pre-trained NN. Then we apply features to the ensemble of fully connected layers to get the distribution of opacity score during the test. We also train several ResNet and DenseNet DNNs to observe the effect of model selection on prediction and uncertainty. The paper also demonstrates a patient referral framework based on the proposed uncertainty quantification. The scripts of the proposed method are available at the following link: https://github.com/dipuk0506/Aleatory-aware-UQ.

At present, hundreds of millions of Chinese people face increasingly serious health risks, and health checks have undoubtedly played a significant role in finding health risks. However, the current health check in China mainly judges the quality of physical functions by a single index value without dynamic analysis of the changing trends of the index, which may lead to unreasonable diagnostic conclusions. In this paper, the data characteristics of physical indicators are systematically analyzed, and grey system models dedicated to data with the characteristics are applied to simulate and predict the changing trends of body indicators. On this basis, possible pathological changes in body organs were identified. Specifically, this paper analyses the state of human kidney functions by grey prediction models. The results showed that even when the renal function index (serum creatinine) is within the normal range, the human renal function might be abnormal. The grey model analysis of the change trends of serum creatinine can predict the potential health hazards of renal functions.

Continuous ambulatory cardiac monitoring plays a critical role in early detection of abnormality in at-risk patients, thereby increasing the chance of early intervention. In this study, we present an automated ECG classification approach for distinguishing between healthy heartbeats and pathological rhythms. The proposed lightweight solution uses quantized one-dimensional deep convolutional neural networks and is ideal for real-time continuous monitoring of cardiac rhythm, capable of providing one output prediction per second. Raw ECG data is used as the input to the classifier, eliminating the need for complex data preprocessing on low-powered wearable devices. In contrast to many compute-intensive approaches, the data analysis can be carried out locally on edge devices, providing privacy and portability. The proposed lightweight solution is accurate (sensitivity of 98.5% and specificity of 99.8%), and implemented on a smartphone, it is energy-efficient and fast, requiring 5.85 mJ and 7.65 ms per prediction, respectively.

Sleep is imperative for a healthy life as it rejuvenates memory, cognitive performance, cell repair and eliminates waste from the muscles. Sleep-related disorders such as insomnia, narcolepsy, sleep-disordered breathing (SDB), periodic leg movement (PLM), and bruxism lead to hormonal imbalance, slower reaction time, memory problems, depression, and headaches. This adversity of sleep disorder gained the attention of many sleep researchers. To examine the reasons for sleep disorders, it is imperative to monitor and analyze the sleep of the affected patients. The conventional method of monitoring sleep and identifying the sleep disorders using polysomnographic (PSG) recording is a complicated and cumbersome task in which multiple physiological signals with multiple modalities are recorded for a long (overnight) duration. The PSG recordings are carried out in sophisticated sleep laboratories and cannot be considered suitable for real-time sleep monitoring. Thus, a simple and patient-convenient system is highly desirable to monitor and analyze the quality of sleep. We proposed an automatic detection of sleep disorders using single modal electrooculogram (EOG) and electromyogram (EMG) signals. We have used a new maximally flat multiplier-less biorthogonal filter bank for obtaining discrete wavelet transform of the signals. We computed Hjorth parameters (HOP) such as activity, mobility, and complexity from the wavelet sub-bands. Highly discriminative HOP features are fed to different machine learning classifiers to develop the model. Our results show that the developed system can classify insomnia, narcolepsy, NFLE, PLM, and REM behaviour disorder (RBD) against normal healthy subjects with an accuracy of 99.7%, 97.6%, 97.5%, 97.5%, and 98.3%, respectively using combined features from EOG and EMG signal. The proposed model has yielded an accuracy of 94.3% in classifying six classes using an ensemble bagged trees classifier (EBTC) with a 10-fold cross-validation technique. Hence, EOG and EMG-based proposed methods can be deployed in a portable home-based environment to identify the type of sleep disorders automatically.

BACKGROUND: Superficial femoral arteries (SFAs) treated with self-expanding stents are widely affected by in-stent restenosis (ISR), especially in case of long lesions and multiple overlapping devices. The altered hemodynamics provoked by the stent is considered as a promoting factor of ISR. In this context, this work aims to analyze the impact of stent design and stent overlapping on patient-specific SFA hemodynamics. METHODS: Through a morphing technique, single or multiple stents were virtually implanted within two patient-specific, post-operative SFA models reconstructed from computed tomography. The stented domains were used to perform computational fluid dynamics simulations, quantifying wall shear stress (WSS) based descriptors including time-averaged WSS (TAWSS), oscillatory shear index (OSI), transverse WSS (transWSS), and WSS ratio (WSSRATIO). Four stent designs (three laser-cut - EverFlex, Zilver and S.M.A.R.T. - and one prototype braided stent), and three typical clinical scenarios accounting for different order of stent implantation and overlapping length were compared. RESULTS: The main hemodynamic differences were found between the two types of stent designs (i.e. laser-cut vs. braided stents). The braided stent presented lower median transWSS and higher median WSSRATIO than the laser-cut stents (p < 0.0001). The laser-cut stents presented comparable WSS-based descriptor values, except for the Zilver, exhibiting a median TAWSS ∼30% higher than the other stents. Stent overlapping provoked an abrupt alteration of the WSS-based descriptors. The overlapping length, rather than the order of stent implantation, highly and negatively impacted the hemodynamics. CONCLUSION: The proposed computational workflow compared different SFA stent designs and stent overlapping configurations, highlighting those providing the most favorable hemodynamic conditions.

BACKGROUND: The larynx of the human respiratory tract plays a vital role in breathing and voice production. Both can be influenced by functional and/or morphological changes of the larynx, e.g., immobility of one or both vocal folds (VF). The immobile VF can become stationary in different positions such as the median, paramedian, intermediate or lateral position. The impact of unilateral vocal fold immobility (UVFI) on inhalation is the focus of this study. METHODS: Transient numerical simulations of the inhalation process in patient-specific airways are performed. Five configurations are considered: paramedian and intermediate VF positions on the left and right, and healthy. Large eddy simulations are used to describe the complex laryngeal turbulent flow. Airway resistance, power loss, and spectral entropy are calculated to quantify the work of inspiration and evaluate flow regimes. RESULTS: The laryngeal jet intensity and flow disturbance increase with the severity of immobility. In comparison to the healthy configuration, UVFI with right/left intermediate and right/left paramedian VF position increases the airway resistance over the oropharynx to the trachea by 69%/58% and 310%/285%, respectively. When the entire respiratory system is considered, an increase of up to 48% is estimated. Spectral entropy increases of up to 2.5 times indicate higher turbulence levels due to UVFI. CONCLUSIONS: Surgery of immobile VF aims to improve glottis closure. However, this can have a negative impact on breathing efficiency. To that end, this study provides initial insights into the conflicting objectives of open versus closed VFs.

The challenge of identifying modules in a gene interaction network is important for a better understanding of the overall network architecture. In this work, we develop a novel similarity measure called Scaling-and-Shifting Normalized Mean Residue Similarity (SNMRS), based on the existing NMRS technique [1]. SNMRS yields correlation values in the range of 0 to +1 corresponding to negative and positive dependency. To study the performance of our measure, internal validation of extracted clusters resulting from different methods is carried out. Based on the performance, we choose hierarchical clustering and apply the same using the corresponding dissimilarity (distance) values of SNMRS scores, and utilize a dynamic tree cut method for extracting dense modules. The modules are validated using a literature search, KEGG pathway analysis, and gene-ontology analyses on the genes that make up the modules. Moreover, our measure can handle absolute, shifting, scaling, and shifting-and-scaling correlations and provides better performance than several other measures in terms of cluster-validity indices. Also, SNMRS based module detection method results in interesting biologically relevant patterns from gene microarray and RNA-seq dataset. A set of crucial genes having high relevance with the ESCC are also identified.

Gastric cancer is the second leading cause of cancer-related deaths worldwide. Early diagnosis significantly increases the chances of survival; therefore, improved assisted exploration and screening techniques are necessary. Previously, we made use of an augmented multi-spectral endoscope by inserting an optical probe into the instrumentation channel. However, the limited field of view and the lack of markings left by optical biopsies on the tissue complicate the navigation and revisit of the suspect areas probed in-vivo. In this contribution two innovative tools are introduced to significantly increase the traceability and monitoring of patients in clinical practice: (i) video mosaicing to build a more comprehensive and panoramic view of large gastric areas; (ii) optical biopsy targeting and registration with the endoscopic images. The proposed optical flow-based mosaicing technique selects images that minimize texture discontinuities and is robust despite the lack of texture and illumination variations. The optical biopsy targeting is based on automatic tracking of a free-marker probe in the endoscopic view using deep learning to dynamically estimate its pose during exploration. The accuracy of pose estimation is sufficient to ensure a precise overlapping of the standard white-light color image and the hyperspectral probe image, assuming that the small target area of the organ is almost flat. This allows the mapping of all spatio-temporally tracked biopsy sites onto the panoramic mosaic. Experimental validations are carried out from videos acquired on patients in hospital. The proposed technique is purely software-based and therefore easily integrable into clinical practice. It is also generic and compatible to any imaging modality that connects to a fiberscope.

Widely used traditional supervised deep learning methods require a large number of training samples but often fail to generalize on unseen datasets. Therefore, a more general application of any trained model is quite limited for medical imaging for clinical practice. Using separately trained models for each unique lesion category or a unique patient population will require sufficiently large curated datasets, which is not practical to use in a real-world clinical set-up. Few-shot learning approaches can not only minimize the need for an enormous number of reliable ground truth labels that are labour-intensive and expensive, but can also be used to model on a dataset coming from a new population. To this end, we propose to exploit an optimization-based implicit model agnostic meta-learning (iMAML) algorithm under few-shot settings for medical image segmentation. Our approach can leverage the learned weights from diverse but small training samples to perform analysis on unseen datasets with high accuracy. We show that, unlike classical few-shot learning approaches, our method improves generalization capability. To our knowledge, this is the first work that exploits iMAML for medical image segmentation and explores the strength of the model on scenarios such as meta-training on unique and mixed instances of lesion datasets. Our quantitative results on publicly available skin and polyp datasets show that the proposed method outperforms the naive supervised baseline model and two recent few-shot segmentation approaches by large margins. In addition, our iMAML approach shows an improvement of 2%-4% in dice score compared to its counterpart MAML for most experiments.