Search Results (189)

Recent advances in exoskeleton-assisted rehabilitation have highlighted the significance of lower limb movement intention recognition through deep learning. However, discrete motion phase classification and continuous real-time joint kinematics estimation are typically handled as independent tasks, leading to temporal misalignment or delayed assistance during dynamic movements. To address this issue, this study presents iTransformer-DTL, a dual-task learning framework with an improved Transformer designed to identify end-to-end locomotion modes and predict joint trajectories during sit-to-stand transitions. Employing a learnable query mechanism and a non-autoregressive decoding approach, the proposed iTransformer-DTL can produce the complete output sequence at once, without relying on any previously generated elements. The proposed framework has been tested with a dataset of lower limb movements involving seven healthy individuals and seven stroke patients. The experimental results indicate that the proposed framework achieves satisfactory performance in dual tasks. An average angle prediction Mean Absolute Error (MAE) of 3.84° and a classification accuracy of 99.42% were obtained in the healthy group, while 4.62° MAE and 99.01% accuracy were achieved in the stroke group. These results suggest that iTransformer-DTL could support adaptable rehabilitation exoskeleton controllers, enhancing human–robot interactions. Full article

Sepsis is a life-threatening condition that requires early recognition and intervention to improve patient outcomes. Optimizing hemodynamic management is crucial, and clinicians must utilize all available tools to guide therapy effectively. Echocardiography is a rapid, non-invasive, and repeatable method that has emerged as a valuable tool in the management of septic patients. Studying its role can provide insights into both therapeutic guidance and prognostic assessment. The primary aim of this review is to highlight the importance of echocardiography in the hemodynamic management of patients with sepsis. The secondary objective is to assess its prognostic value, as echocardiography can inform both the immediate management of critically ill patients and their overall prognosis. A narrative review of the literature published in the last 15 years was conducted using PubMed, and references were managed with Mendeley. Articles focusing on adult and pediatric patients, as well as relevant animal studies, which evaluated echocardiographic assessment of cardiac function, fluid responsiveness, or hemodynamic management were included. Multiple studies demonstrate that echocardiography is a reliable, non-invasive, and easily repeatable tool for assessing fluid responsiveness in septic patients. It allows for dynamic monitoring of stroke volume, VTI, and other hemodynamic parameters, supporting tailored fluid and vasoactive therapy. Additionally, echocardiography provides prognostic insights, with right ventricular dysfunction emerging as a strong predictor of increased mortality. Other parameters, including global longitudinal strain and left ventricular diastolic function, further contribute to risk stratification. Echocardiography is an indispensable tool in the management of sepsis, offering both real-time guidance for hemodynamic optimization and valuable prognostic information. Its routine use can enhance personalized care and improve clinical outcomes in critically ill septic patients. Full article

Timely identification and accurate delineation of ultra-early ischemic stroke lesions in non-contrast computed tomography (CT) scans of the human brain are of paramount importance for prompt medical intervention and improved patient outcomes. In this study, we propose a deep learning-driven methodology specifically designed for segmenting ultra-early ischemic regions, with a particular emphasis on both the ischemic core and the surrounding penumbra during the initial stages of stroke progression. We introduce a lightweight preprocessing model based on convolutional filtering techniques, which enhances image clarity while preserving the structural integrity of medical scans, a critical factor when detecting subtle signs of ultra-early ischemic strokes. Unlike conventional preprocessing methods that directly modify the image and may introduce artifacts or distortions, our approach ensures the absence of neural network-induced artifacts, which is especially crucial for accurate diagnosis and segmentation of ultra-early ischemic lesions. The model employs predefined differentiable filters with trainable parameters, allowing for artifact-free and precision-enhanced image refinement tailored to the challenges of ultra-early stroke detection. In addition, we incorporated into the combined preprocessing pipeline a newly proposed trainable linear combination of pretrained image filters, a concept first introduced in this study. For model training and evaluation, we utilize a publicly available dataset of acute ischemic stroke cases, focusing on the subset relevant to ultra-early stroke manifestations, which contains annotated non-contrast CT brain scans from 112 patients. The proposed model demonstrates high segmentation accuracy for ultra-early ischemic regions, surpassing existing methodologies across key performance metrics. The results have been rigorously validated on test subsets from the dataset, confirming the effectiveness of our approach in supporting the early-stage diagnosis and treatment planning for ultra-early ischemic strokes. Full article

Ischemic stroke is now widely recognized as a disease with a strong inflammatory profile. Cerebral vascular damage is both preceded and followed by a chain of molecular events involving immune cells and inflammatory markers, irrespective of the etiology of the ischemic injury. Over time, an increasingly comprehensive understanding of these markers has led to a better insight into the mechanisms behind the vascular event and recovery following ischemic stroke. However, to date, there are still no available circulating or tissue biomarkers for early diagnosis or prognostic stratification, making ischemic stroke diagnosis contingent on clinical and instrumental investigations. However, neurological and internal medicine research is progressing in identifying markers that could potentially take on this role. This manuscript, therefore, aims to review the most recent and innovative results of medical advances, summarising the current state of the art and future perspectives. If ischaemic stroke is an inflammatory disease, it is also true that it is not just a singular condition, but a group of entities with their own neuroinflammatory features. Thus, given that, in ischemic cerebral vascular damage, “time is brain,” tracking increasingly accurate markers in the diagnosis of ischemic stroke is a valuable tool that will potentially enable earlier recognition of this disease and, hopefully, make it less disabling and more widely treated. Full article

Spontaneous intracerebral hemorrhage (ICH) is a devastating form of stroke, disproportionately affecting older adults and is associated with high rates of mortality, functional dependence, and long-term cognitive decline. Aging profoundly alters the structure and function of the cerebral vasculature, predisposing the brain to both covert hemorrhage and the development of cerebral microbleeds (CMBs), small, often subclinical lesions that share common pathophysiological mechanisms with ICH. These mechanisms include endothelial dysfunction, impaired cerebral autoregulation, blood–brain barrier breakdown, vascular senescence, and chronic inflammation. Systemic factors such as age-related insulin-like growth factor 1 (IGF-1) deficiency further exacerbate microvascular vulnerability. CMBs and ICH represent distinct yet interconnected manifestations along a continuum of hemorrhagic small vessel disease, with growing recognition of their contribution to vascular cognitive impairment and dementia (VCID). Despite their increasing burden, older adults remain underrepresented in clinical trials, and few therapeutic approaches specifically target aging-related mechanisms. This review synthesizes current knowledge on the cellular, molecular, and systemic drivers of ICH and CMBs in aging, highlights diagnostic and therapeutic challenges, and outlines opportunities for age-sensitive prevention and individualized care strategies. Full article

Background: Stroke is a leading cause of seizures and epilepsy, both of which are linked to increased mortality, disability, and hospital readmissions. Early recognition and management of seizures in acute stroke are crucial for improving outcomes. Electroencephalogram (EEG) is not routinely used for post-stroke seizure monitoring and is typically initiated only after clinical suspicion arises, making bedside recognition essential. This scoping review aimed to map the existing literature on clinical methods used for identifying and observing early post-stroke seizures (EPSSs) at the bedside. Methods: We included literature involving adults with acute ischaemic stroke or primary intracerebral haemorrhage who were diagnosed or suspected of having inpatient EPSS. Searches were conducted in Medline, CINAHL, Embase, and the Cochrane Library for English-language publications up to April 2023. Eligible sources included primary research, case reports, systematic reviews, clinical guidelines, consensus statements, and expert opinion. Reference lists of included articles were also reviewed. Data were charted and synthesised to assess the scope, type, and gaps in the evidence. Results: Thirty papers met inclusion criteria: 17 research studies, six expert opinions, four case reports, and three clinical guidelines. Empirical evidence on clinical methods for seizure recognition and monitoring in acute stroke was limited. No studies evaluated the effectiveness of different approaches, and existing recommendations lacked detail and consensus. Conclusions: Accurate EPSS diagnosis is vital due to its impact on outcomes. This review highlights inconsistency in monitoring methods and a clear need for targeted research into effective clinical identification strategies in acute stroke care. Full article

Background/Objectives: Blood vessel tortuosity can complicate endovascular procedures such as endovascular thrombectomy in acute ischemic stroke. This study aimed to assess the morphometric characteristics of carotid arteries and investigate the association between the tortuosity of the carotid arteries and the technical aspects of endovascular thrombectomy, patient demographics and clinical characteristics, and treatment outcome. Methods: This retrospective study included 84 patients with ischemic stroke treated by endovascular thrombectomy at the newly established thrombectomy-capable stroke center. The following data were collected from prethrombectomy computed tomography angiography: aortic arch type, type of carotid artery tortuosity, and tortuosity index (TI). The technical aspects of the procedure, as well as patient demographics, were collected from the radiological information system. Results: Time from arterial puncture to the first pass was significantly shorter in patients with a nontortuous carotid artery compared to a tortuous one (p = 0.006). There were no significant differences in the number of passes, total duration of the procedure, and the difference in National Institutes of Health Stroke Scale (NIHSS) score before and after the procedure regarding the form of tortuosity. Patients with hypertension had significantly higher tortuosity index values compared to those without hypertension (p = 0.008), and patients with a nontortuous carotid tree were significantly younger compared to those with all forms of tortuosity (p = 0.003). Conclusions: The majority of patients had tortuous carotid arteries, which were associated with older age and hypertension. A high index of tortuosity was associated with a longer time from arterial puncture to the first pass, but not to the treatment outcome. Preprocedural recognition of carotid artery tortuosity may aid in endovascular thrombectomy procedural planning. Full article

Background/Objectives: Stroke is a major cause of disability worldwide, with ischemic stroke being the most common type. This study investigated face perception in patients with lacunar strokes, specifically examining the ability to distinguish and recognize familiar and unfamiliar faces. Methods: We tested 52 patients with lacunar stroke (mean age = 65.97 ± 9.96) and 28 age-matched healthy controls (HC) (mean age = 66.24 ± 10.15). The participants received three face perception tasks: Name that Celebrity, Identity Sorting Task, and Face & Object Solitaire, and were also given the MMSE and mRS clinical assessments. Results: For the Name that Celebrity task, the stroke group had a lower efficiency score than the control group (i.e., they needed 2–3 extra slides of cues to recognize famous persons). For the Face Identity Sorting task, both groups were more accurate when sorting familiar faces; however, the stroke group performed significantly worse than the healthy group when sorting unfamiliar faces. For the Face/Object Solitaire task, the control group performed better than the stroke group on the face solitaire, but there were no differences in the object solitaire condition. Conclusions: Our findings suggest that despite having a normal mean MMSE score (HC: 28.22, Stroke: 27.96), patients with lacunar stroke had difficulties recognizing famous faces and discriminating among unfamiliar faces. This may reveal an overlooked deficit in face perception, highlighting the importance of future interventions that specifically focus on face recognition skills to enhance patients’ daily social interactions and the overall effectiveness of post-stroke rehabilitation programs. Full article

This study presents a low-cost wearable system for monitoring and classifying paddle strokes in open-water environments. Building upon our previous work in controlled aquatic and dryland settings, the proposed system consists of ESP32-based embedded nodes equipped with MPU6050 accelerometer–gyroscope sensors. These nodes communicate via the ESP-NOW protocol in a master–slave architecture. With minimal hardware modifications, the system implements gesture classification using Dynamic Time Warping (DTW) to distinguish between left and right paddle strokes. The collected data, including stroke type, count, and motion similarity, are transmitted in real time to a local interface for visualization. Field experiments were conducted on a calm lake using a paddleboard, where users performed a series of alternating strokes. In addition to gesture recognition, the study includes empirical testing of ESP-NOW communication range in the open lake environment. The results demonstrate reliable wireless communication over distances exceeding 100 m with minimal packet loss, confirming the suitability of ESP-NOW for low-latency data transfer in open-water conditions. The system achieved over 80% accuracy in stroke classification and sustained more than 3 h of operational battery life. This approach demonstrates the feasibility of real-time, wearable-based motion tracking for water sports in natural environments, with potential applications in kayaking, rowing, and aquatic training systems. Full article

Digital ink Chinese characters (DICCs) written by international students often contain various errors and irregularities, making the recognition of these characters a highly challenging pattern recognition problem. This paper designs a graph convolutional network with agent attention (GCNAA) for recognizing DICCs written by international students. Each sampling point is treated as a vertex in a graph, with connections between adjacent sampling points within the same stroke serving as edges to create a Chinese character graph structure. The GCNAA is used to process the data of the Chinese character graph structure, implemented by stacking Block modules. In each Block module, the graph agent attention module not only models the global context between graph nodes but also reduces computational complexity, shortens training time, and accelerates inference speed. The graph convolution block module models the local adjacency structure of the graph by aggregating local geometric information from neighboring nodes, while graph pooling is employed to learn multi-resolution features. Finally, the Softmax function is used to generate prediction results. Experiments conducted on public datasets such as CASIA-OLWHDB1.0-1.2, SCUT-COUCH2009 GB1&GB2, and HIT-OR3C-ONLINE demonstrate that the GCNAA performs well even on large-category datasets, showing strong generalization ability and robustness. The recognition accuracy for DICCs written by international students reaches 98.7%. Accurate and efficient handwritten Chinese character recognition technology can provide a solid technical foundation for computer-assisted Chinese character writing for international students, thereby promoting the development of international Chinese character education. Full article

Background/Objectives: Tick-borne diseases (TBDs) are increasingly recognized as causes of both systemic and neurologic illness. While their impact on vascular health is established, their role in cerebrovascular disease remains underexplored. This review aims to synthesize clinical evidence linking TBDs with cerebrovascular events, focusing on mechanisms of injury, pathogen-specific associations, and treatment outcomes. Methods: A narrative review was conducted using Boolean keyword searches across PubMed, Scopus, EMBASE, and Web of Science. Relevant literature on ischemic and hemorrhagic stroke, cerebral vasculitis, and stroke mimics associated with TBDs was examined. The review included case reports, observational studies, and mechanistic research. Pathogen-specific data and disease characteristics were extracted and summarized. Results: Several tick-borne pathogens were associated with cerebrovascular complications. Borrelia burgdorferi was most commonly implicated and typically presented with large-vessel vasculitis. Rickettsia, Ehrlichia, and Anaplasma species caused endothelial injury through immune-mediated inflammation. Powassan virus and Crimean–Congo hemorrhagic fever virus exhibited central nervous system involvement and hemorrhagic potential. Babesia species contributed to vascular injury through thrombocytopenia and embolic complications. Neuroimaging frequently demonstrated multifocal stenoses and vessel wall inflammation. Antimicrobial treatment, particularly with doxycycline or ceftriaxone, was often effective, especially when administered early. Supportive care for stroke symptoms varied by presentation and underlying pathogen. Conclusions: Cerebrovascular disease caused by tick-borne pathogens is an underrecognized but potentially reversible condition. Despite diverse etiologies, most pathogens share a final common pathway of endothelial dysfunction. Early recognition and targeted antimicrobial therapy, combined with supportive stroke care, are essential to improving patient outcomes. Full article

Introduction: Hypertrophic olivary degeneration (HOD) is a rare form of trans-synaptic degeneration involving the Guillain–Mollaret triangle, characterized by enlargement of the inferior olivary nucleus—unlike the atrophy typical of most neurodegenerative processes. It is usually associated with stroke, surgical injury, or demyelination, but rarely follows hemorrhage from a cavernous malformation (CM). This report presents a case of HOD secondary to a mesencephalic CM hemorrhage, with emphasis on imaging findings and diagnostic considerations. Case Description: A 55-year-old woman presented with acute-onset, right-sided facial, torso, and limb hypoesthesia, along with gait instability. Neurological examination revealed sensory impairment in the right maxillary (V2) and mandibular (V3) trigeminal territories, as well as diminished pain and temperature sensation throughout the right hemibody. MRI revealed a hemorrhage in the posterior mesencephalon near the left red nucleus, leading to the diagnosis of a CM with an associated venous angioma. She was managed conservatively and improved clinically. Six months later, MRI showed hypertrophy and T2/FLAIR hyperintensity of the left inferior olive, consistent with developing HOD. At 1.5 years follow-up, olivary enlargement had progressed—now consistent with stage 2 HOD—and a bilateral palatal tremor was observed, more pronounced on the right side. DTI revealed asymmetric volume loss in the left brainstem fiber pathways at the level of the medulla oblongata, confirming trans-synaptic degeneration. Conclusions: This case highlights HOD as a rare but important complication of mesencephalic CM hemorrhage. Recognition of its characteristic imaging features—olivary hypertrophy with persistent T2/FLAIR hyperintensity—is essential for accurate diagnosis. DTI supports the trans-synaptic mechanism, helping distinguish HOD from other pathologies and preventing unnecessary investigations. Full article

Wi-Fi sensing leverages variations in CSI (channel state information) to infer human activities in a contactless and low-cost manner, with growing applications in smart homes, healthcare, and security. While deep learning has advanced macro-motion sensing tasks, micro-motion sensing such as keypad stroke recognition remains underexplored due to subtle inter-class CSI variations and significant intra-class variance. These challenges make it difficult for existing deep learning models typically relying on fully connected MLPs to accurately recognize keypad inputs. To address the issue, we propose a novel approach that combines a discriminative feature extractor with a Kolmogorov–Arnold Network (KAN)-based classifier. The combined model is trained to reduce intra-class variability by clustering features around class-specific centers. The KAN classifier learns nonlinear spline functions to efficiently delineate the complex decision boundaries between different keypad inputs with fewer parameters. To validate our method, we collect a CSI dataset with low-cost Wi-Fi devices (ESP8266 and Raspberry Pi 4) in a real-world keypad sensing environment. Experimental results verify the effectiveness and practicality of our method for keypad input sensing applications in that it outperforms existing approaches in sensing accuracy while requiring fewer parameters. Full article

Chinese Optical Character Recognition (OCR) technology is essential for digital transformation in Chinese regions, enabling automated document processing across various applications. However, Chinese OCR systems struggle with visually similar characters, where subtle stroke differences lead to systematic recognition errors that limit practical deployment accuracy. This study develops FLIP (Feedback Learning-based Intelligent Plugin), a lightweight post-processing plugin designed to improve Chinese OCR accuracy across different systems without external dependencies. The plugin operates through three core components as follows: UTF-8 encoding-based output parsing that converts OCR results into mathematical representations, error correction using information entropy and weighted similarity measures to identify and fix character-level errors, and adaptive feedback learning that optimizes parameters through user interactions. The approach functions entirely through mathematical calculations at the character encoding level, ensuring universal compatibility with existing OCR systems while effectively handling complex Chinese character similarities. The plugin’s modular design enables seamless integration without requiring modifications to existing OCR algorithms, while its feedback mechanism adapts to domain-specific terminology and user preferences. Experimental evaluation on 10,000 Chinese document images using four state-of-the-art OCR models demonstrates consistent improvements across all tested systems, with precision gains ranging from 1.17% to 10.37% and overall Chinese character recognition accuracy exceeding 98%. The best performing model achieved 99.42% precision, with ablation studies confirming that feedback learning contributes additional improvements from 0.45% to 4.66% across different OCR architectures. Full article

Background/Objectives: Successful discourse relies not only on linguistic but also on prosodic information. Difficulty recognizing emotion conveyed through prosody (receptive affective aprosodia) following right hemisphere stroke (RHS) significantly disrupts communication participation and personal relationships. Growing evidence suggests that damage to white matter in addition to gray matter structures impairs affective prosody recognition. The current study investigates lesion–symptom associations in receptive affective aprosodia during RHS recovery by assessing whether disruptions in distinct white matter structures impact different underlying affective prosody recognition skills. Methods: Twenty-eight adults with RHS underwent neuroimaging and behavioral testing at acute, subacute, and chronic timepoints. Fifty-seven healthy matched controls completed the same behavioral testing, which comprised tasks targeting affective prosody recognition and underlying perceptual, cognitive, and linguistic skills. Linear mixed-effects models and multivariable linear regression were used to assess behavioral performance recovery and lesion–symptom associations. Results: Controls outperformed RHS participants on behavioral tasks earlier in recovery, and RHS participants’ affective prosody recognition significantly improved from acute to chronic testing. Affective prosody and emotional facial expression recognition were affected by external capsule and inferior fronto-occipital fasciculus lesions while sagittal stratum lesions impacted prosodic feature recognition. Accessing semantic representations of emotions implicated the superior longitudinal fasciculus. Conclusions: These findings replicate previously observed associations between right white matter tracts and affective prosody recognition and further identify lesion–symptom associations of underlying prosodic recognition skills throughout recovery. Investigation into prosody’s behavioral components and how they are affected by injury can help further intervention development and planning. Full article