PAR-Net: An Enhanced Dual-Stream CNN-ESN Architecture for Human Physical Activity Recognition.

Physical exercise affects many facets of life, including mental health, social interaction, physical fitness, and illness prevention, among many others. Therefore, several AI-driven techniques have been developed in the literature to recognize human physical activities. However, these techniques fail to adequately learn the temporal and spatial features of the data patterns. Additionally, these techniques are unable to fully comprehend complex activity patterns over different periods, emphasizing the need for enhanced architectures to further increase accuracy by learning spatiotemporal dependencies in the data individually. Therefore, in this work, we develop an attention-enhanced dual-stream network (PAR-Net) for physical activity recognition with the ability to extract both spatial and temporal features simultaneously. The PAR-Net integrates convolutional neural networks (CNNs) and echo state networks (ESNs), followed by a self-attention mechanism for optimal feature selection. The dual-stream feature extraction mechanism enables the PAR-Net to learn spatiotemporal dependencies from actual data. Furthermore, the incorporation of a self-attention mechanism makes a substantial contribution by facilitating targeted attention on significant features, hence enhancing the identification of nuanced activity patterns. The PAR-Net was evaluated on two benchmark physical activity recognition datasets and achieved higher performance by surpassing the baselines comparatively. Additionally, a thorough ablation study was conducted to determine the best optimal model for human physical activity recognition.

Gait Pattern in Charcot-Marie-Tooth Disease Type 1A According to Disease Severity.

The aim of this study was to evaluate the characteristics of gait patterns in Charcot-Marie-Tooth disease type 1A (CMT1A) patients according to disease severity. Twenty-two CMT1A patients were enrolled and classified into two groups, according to the disease severity. The healthy control group consisted of 22 subjects with no gait impairment. Full barefoot three-dimensional gait analysis with temporospatial, kinematic, and kinetic data was performed among the mild and moderate CMT1A group and the control group. Minimal hip abduction, maximal hip extension generation, peak knee flexion moment at stance, ankle dorsiflexion at initial contact, maximal ankle plantarflexion at push-off and maximal ankle rotation moment at stance in the CMT1A group showed a significant difference compared to the control group (p < 0.05). In the moderate group, there were greater maximal hip flexion angles in swing, and smaller dorsiflexion angles at initial contact compared to the control group and mild group. CMT patients had typical gait characteristics and their gait patterns were different according to severity. The analysis of gait patterns in patients with CMT1A will help to understand gait function and provide important information for the treatment of patients with CMT in the future.

Visual Intelligence in Precision Agriculture: Exploring Plant Disease Detection via Efficient Vision Transformers.

In order for a country's economy to grow, agricultural development is essential. Plant diseases, however, severely hamper crop growth rate and quality. In the absence of domain experts and with low contrast information, accurate identification of these diseases is very challenging and time-consuming. This leads to an agricultural management system in need of a method for automatically detecting disease at an early stage. As a consequence of dimensionality reduction, CNN-based models use pooling layers, which results in the loss of vital information, including the precise location of the most prominent features. In response to these challenges, we propose a fine-tuned technique, GreenViT, for detecting plant infections and diseases based on Vision Transformers (ViTs). Similar to word embedding, we divide the input image into smaller blocks or patches and feed these to the ViT sequentially. Our approach leverages the strengths of ViTs in order to overcome the problems associated with CNN-based models. Experiments on widely used benchmark datasets were conducted to evaluate the proposed GreenViT performance. Based on the obtained experimental outcomes, the proposed technique outperforms state-of-the-art (SOTA) CNN models for detecting plant diseases.

Cervical Fusion-Requiring Mineral and Bone Disorder Disguised as Osteoporosis in a Chronic Kidney Disease Patient on Hemodialysis: A Case Report.

Bony deformities and fragility fractures in end-stage renal disease (ESRD) patients on long-term hemodialysis can be caused by either osteoporosis or chronic kidney disease-mineral and bone disorder (CKD-MBD). Correct identification of the underlying mechanism is critical since the treatment methods differ, and one treatment approach could negatively affect the other. Cervical kyphosis, severe enough to require immediate surgical treatment, can be caused by uncontrolled CKD-MBD, albeit in limited cases. This report presents the case of a 61-year-old female with an 11-year history of hemodialysis treatment and severe cervical kyphosis with myelopathy, which required 2-stage spinal surgeries. Our report calls for a careful diagnostic approach in ESRD patients with skeletal disorders, the points to consider before calcium replacement, and early detection of fragility fractures in them. Moreover, early mobilization and weight-bearing after the surgical procedure may lead to better neurological and functional improvements.

Human Activity Recognition via Hybrid Deep Learning Based Model.

In recent years, Human Activity Recognition (HAR) has become one of the most important research topics in the domains of health and human-machine interaction. Many Artificial intelligence-based models are developed for activity recognition; however, these algorithms fail to extract spatial and temporal features due to which they show poor performance on real-world long-term HAR. Furthermore, in literature, a limited number of datasets are publicly available for physical activities recognition that contains less number of activities. Considering these limitations, we develop a hybrid model by incorporating Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) for activity recognition where CNN is used for spatial features extraction and LSTM network is utilized for learning temporal information. Additionally, a new challenging dataset is generated that is collected from 20 participants using the Kinect V2 sensor and contains 12 different classes of human physical activities. An extensive ablation study is performed over different traditional machine learning and deep learning models to obtain the optimum solution for HAR. The accuracy of 90.89% is achieved via the CNN-LSTM technique, which shows that the proposed model is suitable for HAR applications.

Adaptive Node Clustering for Underwater Sensor Networks.

Monitoring of an underwater environment and communication is essential for many applications, such as sea habitat monitoring, offshore investigation and mineral exploration, but due to underwater current, low bandwidth, high water pressure, propagation delay and error probability, underwater communication is challenging. In this paper, we proposed a sensor node clustering technique for UWSNs named as adaptive node clustering technique (ANC-UWSNs). It uses a dragonfly optimization (DFO) algorithm for selecting ideal measure of clusters needed for routing. The DFO algorithm is inspired by the swarming behavior of dragons. The proposed methodology correlates with other algorithms, for example the ant colony optimizer (ACO), comprehensive learning particle swarm optimizer (CLPSO), gray wolf optimizer (GWO) and moth flame optimizer (MFO). Grid size, transmission range and nodes density are used in a performance matrix, which varies during simulation. Results show that DFO outperform the other algorithms. It produces a higher optimized number of clusters as compared to other algorithms and hence optimizes overall routing and increases the life span of a network.

Machine Learning-Based Cardiovascular Disease Prediction Model: A Cohort Study on the Korean National Health Insurance Service Health Screening Database.

BACKGROUND: This study proposes a cardiovascular diseases (CVD) prediction model using machine learning (ML) algorithms based on the National Health Insurance Service-Health Screening datasets. METHODS: We extracted 4699 patients aged over 45 as the CVD group, diagnosed according to the international classification of diseases system (I20-I25). In addition, 4699 random subjects without CVD diagnosis were enrolled as a non-CVD group. Both groups were matched by age and gender. Various ML algorithms were applied to perform CVD prediction; then, the performances of all the prediction models were compared. RESULTS: The extreme gradient boosting, gradient boosting, and random forest algorithms exhibited the best average prediction accuracy (area under receiver operating characteristic curve (AUROC): 0.812, 0.812, and 0.811, respectively) among all algorithms validated in this study. Based on AUROC, the ML algorithms improved the CVD prediction performance, compared to previously proposed prediction models. Preexisting CVD history was the most important factor contributing to the accuracy of the prediction model, followed by total cholesterol, low-density lipoprotein cholesterol, waist-height ratio, and body mass index. CONCLUSIONS: Our results indicate that the proposed health screening dataset-based CVD prediction model using ML algorithms is readily applicable, produces validated results and outperforms the previous CVD prediction models.

Reassessment of Inclusion Criteria in the 2013 the American College of Cardiology and the American Heart Association Cholesterol Guidelines for Cardiovascular Disease Prevention.

BACKGROUND AND PURPOSE: The American College of Cardiology and the American Heart Association (ACC-AHA) have released new guidelines and expanded indications for statin treatment. We aimed to reveal the clinical efficacy of each indication in the guidelines using a large-scale national cohort. METHODS: We used National Health Screening Cohort data to determine the proportions of participants for whom statin therapy would be recommended using the different guidelines. We assessed the cumulative incidence rates of major adverse cardiovascular events (MACE) using the Cox proportional-hazards model. RESULTS: Under the 2013 ACC-AHA guidelines, 111,600 participants were additionally eligible to receive statins, compared with 50,023 participants according to the Third Adult Treatment Panel (ATP-III). Most of the additional statin-eligible participants in the ACC-AHA guidelines were indicated by their 10-year cardiovascular disease risk. The increase in statineligible participants in the ACC-AHA guidelines mainly involved elderly patients aged 60-75 years. Among participants not requiring statin, participants who were eligible for a statin under the ACC-AHA guidelines had a significantly higher hazard ratio of MACE when compared with those eligible under the ATP-III guidelines. Among the not-recommended groups, patients with diabetes and low-density lipoprotein <70 mg/dL constituted the group with the highest risk of MACE. CONCLUSIONS: The 2013 ACC-AHA guidelines increase the number of statin-eligible participants, especially among the elderly. These guidelines provide a stronger recommendation for statins to high-risk groups, but it remains necessary to consider the characteristics of the population in the risk equation. In addition, the aggressive use of statin in diabetes patients and further studies of older subjects are needed.

Recurrent Cerebral Infarctions and Multi-Vessel Thrombosis in a Male Patient with Refractory Idiopathic Thrombocytopenic Purpura: a Case Report.

Idiopathic thrombocytopenic purpura (ITP) mostly presents with bleeding tendencies, and thrombotic events are very uncommon. Our case report presents a male patient with ITP refractory to standardized therapies who continuously showed thrombocytopenia and hematuria. With no evidence of autoimmune diseases or other secondary causes of ITP, he developed recurrent cerebral infarctions and deep venous thrombosis. Our report calls for attention to possible thrombotic events, as well as more common bleeding tendencies in patients with ITP and outlines rehabilitation treatment specially designed for ITP patients with rare thrombotic complications.

DLNR-SIQA: Deep Learning-Based No-Reference Stitched Image Quality Assessment.

Due to recent advancements in virtual reality (VR) and augmented reality (AR), the demand for high quality immersive contents is a primary concern for production companies and consumers. Similarly, the topical record-breaking performance of deep learning in various domains of artificial intelligence has extended the attention of researchers to contribute to different fields of computer vision. To ensure the quality of immersive media contents using these advanced deep learning technologies, several learning based Stitched Image Quality Assessment methods have been proposed with reasonable performances. However, these methods are unable to localize, segment, and extract the stitching errors in panoramic images. Further, these methods used computationally complex procedures for quality assessment of panoramic images. With these motivations, in this paper, we propose a novel three-fold Deep Learning based No-Reference Stitched Image Quality Assessment (DLNR-SIQA) approach to evaluate the quality of immersive contents. In the first fold, we fined-tuned the state-of-the-art Mask R-CNN (Regional Convolutional Neural Network) on manually annotated various stitching error-based cropped images from the two publicly available datasets. In the second fold, we segment and localize various stitching errors present in the immersive contents. Finally, based on the distorted regions present in the immersive contents, we measured the overall quality of the stitched images. Unlike existing methods that only measure the quality of the images using deep features, our proposed method can efficiently segment and localize stitching errors and estimate the image quality by investigating segmented regions. We also carried out extensive qualitative and quantitative comparison with full reference image quality assessment (FR-IQA) and no reference image quality assessment (NR-IQA) on two publicly available datasets, where the proposed system outperformed the existing state-of-the-art techniques.

Automatic 360° Mono-Stereo Panorama Generation Using a Cost-Effective Multi-Camera System.

In recent years, 360° videos have gained the attention of researchers due to their versatility and applications in real-world problems. Also, easy access to different visual sensor kits and easily deployable image acquisition devices have played a vital role in the growth of interest in this area by the research community. Recently, several 360° panorama generation systems have demonstrated reasonable quality generated panoramas. However, these systems are equipped with expensive image sensor networks where multiple cameras are mounted in a circular rig with specific overlapping gaps. In this paper, we propose an economical 360° panorama generation system that generates both mono and stereo panoramas. For mono panorama generation, we present a drone-mounted image acquisition sensor kit that consists of six cameras placed in a circular fashion with optimal overlapping gap. The hardware of our proposed image acquisition system is configured in such way that no user input is required to stitch multiple images. For stereo panorama generation, we propose a lightweight, cost-effective visual sensor kit that uses only three cameras to cover 360° of the surroundings. We also developed stitching software that generates both mono and stereo panoramas using a single image stitching pipeline where the panorama generated by our proposed system is automatically straightened without visible seams. Furthermore, we compared our proposed system with existing mono and stereo contents generation systems in both qualitative and quantitative perspectives, and the comparative measurements obtained verified the effectiveness of our system compared to existing mono and stereo generation systems.

Pathogenesis of cerebral microbleeds: In vivo imaging of amyloid and subcortical ischemic small vessel disease in 226 individuals with cognitive impairment.

OBJECTIVE: Cerebral microbleeds (CMBs) are a neuroimaging marker of small vessel disease (SVD) with relevance for understanding disease mechanisms in cerebrovascular disease, cognitive impairment, and normal aging. It is hypothesized that lobar CMBs are due to cerebral amyloid angiopathy (CAA) and deep CMBs are due to subcortical ischemic SVD. We tested this hypothesis using structural magnetic resonance imaging (MRI) markers of subcortical SVD and in vivo imaging of amyloid in patients with cognitive impairment. METHODS: We included 226 patients: 89 with Alzheimer disease-related cognitive impairment (ADCI) and 137 with subcortical vascular cognitive impairment (SVCI). All subjects underwent amyloid imaging with [(11) C] Pittsburgh compound B (PiB) positron emission tomography, and MRI to detect CMBs and markers of subcortical SVD, including the volume of white matter hyperintensities (WMH) and the number of lacunes. RESULTS: Parietal and occipital lobar CMBs counts were higher in PiB(+) ADCI with moderate WMH than PiB(+) ADCI with minimal WMH, whereas PiB(-) patients with SVCI (ie, "pure" SVCI) showed both lobar and deep CMBs. In multivariate analyses of the whole cohort, WMH volume and lacuna counts were positively associated with both lobar and deep CMBs, whereas amyloid burden (PiB) was only associated with lobar CMBs. There was an interaction between lacuna burden and PiB retention on lobar (but not deep) CMBs (p<0.001). INTERPRETATION: Our findings suggest that although deep CMBs are mainly linked to subcortical SVD, both subcortical SVD and amyloid-related pathologies (eg, CAA) contribute to the pathogenesis of lobar CMBs, at least in subjects with mixed lobar and deep CMBs. Furthermore, subcortical SVD and amyloid-related pathologies interact to increase the risk of lobar CMBs.

Cortical asymmetries in normal, mild cognitive impairment, and Alzheimer's disease.

There are functional and structural neocortical hemispheric asymmetries in people with normal cognition. These asymmetries may be altered in patients with Alzheimer's disease (AD) because there is a loss of neuronal connectivity in the heteromodal cortex. The purpose of this study is to test the hypothesis that individuals with amnestic mild cognitive impairment (aMCI), mild AD, and moderate to severe AD have progressive reductions in thickness asymmetries of the heteromodal neocortex. Right-handed elderly volunteers including normal cognition (NC), aMCI, and AD underwent 3-D volume imaging for cortical thickness. Although the cortical asymmetry pattern observed in normal cognition brains was generally maintained in aMCI and AD, there was a progressive decrease in the degree of asymmetry, especially in the inferior parietal lobule. A reduction of neocortical asymmetries may be a characteristic sign that occurs in patients with AD. Future studies are needed to evaluate whether this loss is specific to AD and if measurements of asymmetry can be used as diagnostic markers and for monitoring disease progression.