Relationship of Rurality and Confidence in Video Skills With Persistent Nonuse of Video Appointments for Health Care.

INTRODUCTION: Limited research has examined how technology and digital literacy may affect patients' use of video visits. This study explored the relationship of demographic factors and patient-reported confidence in digital literacy skills to access to video visits among patients who never used them during the COVID-19 pandemic. METHODS: Using existing survey data, the current study examined data from respondents who did not engage in video appointments but instead attended face-to-face appointments between April and December 2020 for nonemergent health concerns. A multivariable logistic regression model was used to investigate whether demographic and social determinants of health factors, context of care (primary care or psychiatry/psychology), and digital literacy confidence were associated with video visit engagement. Collinearity was assessed using the variance inflation factor. RESULTS: This study found that living in rural areas and having a self-reported lack of confidence in logging video appointments using the Mayo Clinic patient portal were associated with persistent nonuse of video appointments in a cohort of patients who did not use video visits at this institution during the early part of the COVID-19 pandemic. DISCUSSION: The research findings reported herein reveal that individuals living in rural areas and those who lack confidence in logging into patient portals to access video visits tend to persistently avoid using video appointments. More investment is needed at the federal and corporate levels to improve digital connectivity. Digital navigators and community involvement can promote digital adoption. CONCLUSION: To encourage digital competency in rural communities, it is important to implement support strategies through community stakeholders and other resources.

Injury severity analysis of e-bike riders in China based on the in-vehicle recording video crash data: a random parameter ordered logit model.

This study investigates the impacts of various factors on e-bike riders' injury severity in crashes with motor vehicles, based on the in-vehicle recording video crash data in China. Variables from human factors, vehicle characteristics, road conditions, and environmental attributes are extracted from the video, especially for drivers and riders' illegal and avoidance behaviour before the crash, and sun shade canopy use. Results of mixed logit models reveal that drivers' speeding, running red lights, slow-down and swerve behaviour, light trucks, heavy trucks, and buses have significantly varied impacts on riders' injury. Moreover, both drivers and riders' illegal behaviour leads to an increased injury, while their avoidance behaviour before crashes can protect riders. In addition, types of visual obstacles, accidents occurring at night, large vehicles' involvement, and the application of sunshade canopies by riders increased the probability of severe injury, while helmet use can protect riders in accidents with motor vehicles.

Effects of Video-Based Exercises and Conventional Physiotherapy on Upper Extremity Functionality, Selective Motor Control, and Proprioception in Unilateral Cerebral Palsy: A Randomized Controlled Trial.

Objective: The present study was conducted to evaluate the effects of video-based exercises added to conventional physiotherapy (CPT) on upper extremity functionality, selective motor control, and proprioception in individuals with unilateral cerebral palsy (UCP). Materials and Methods: Thirty patients with UCP were randomized into two groups: the intervention group (15 individuals with a mean age of 9.2 ± 3.8 years) and the control group (15 individuals with a mean age of 8.3 ± 4.1 years). The intervention group received 8 weeks of video-based exercises, and the control group received 8 weeks of conventional physiotherapy. Upper extremity functional abilities, upper extremity selective motor control, proprioception, and entertainment levels were evaluated before and after the intervention for all groups. Results: While a significant change was observed in the mean scores of the ABILHAND-Kids, Selective Control of the Upper Extremity Scale right-left scores, shoulder flexion, shoulder abduction, and elbow flexion proprioception angles after the video-based exercises in the intervention group (P < 0.05), a significant change was observed only in the 60-degree flexion angle in the control group (P = 0.001). In the comparison between the groups, there were significant differences in post-intervention value only in shoulder flexion and abduction angles, whereas there was no difference between the groups in elbow flexion angles (P > 0.05). Conclusion: Incorporating video-based exercises into the upper extremity rehabilitation processes of individuals with UCP is beneficial in terms of upper extremity functionality, selective motor control, and proprioception.

Outcomes of Video-assisted Thoracic Surgery-guided Early Evacuation of Traumatic Hemothorax: A Randomized Pilot Study at Level I Trauma Center.

Introduction: Traumatic hemothorax is accounted for about 20% of traumatic chest injuries. Although majority can be managed with the timely placement of intercostal tube (ICT) drainage, the remaining pose a challenge owing to high complication rates associated with retained hemothorax. Although various treatment modalities including intrapleural instillation of fibrinolytics, radioimage guided drainage, VATS guided evacuation and thoractomy do exist to address the retained hemothorax, but indications along with timing to employ a specific treatment option is still unclear and ambiguous. Methods: Patient with residual hemothorax (>200 mL) on ultrasonography after 48 h of indwelling ICT was randomized into either early video-assisted thoracic surgery (VATS) or conventional approach cohort. Early VATS cohort was subjected to video-assisted thoracoscopic evacuation of undrained blood along with normal saline irrigation and ICT placement. The conventional cohort underwent intrapleural thrombolytic instillation for 3 consecutive days. The outcome measures were the duration of indwelling ICT, removal rate of tube thoracostomy, length of hospital stay, duration of intensive care unit (ICU) monitoring, need for mechanical ventilation, incidence of pulmonary and pleural complications, and requirement of additional intervention to address undrained hemothorax and mortality rate. Results: The early VATS cohort had shorter length of hospital stay (7.50 ± 0.85 vs. 9.50 ± 3.03, P = 0.060), reduced duration of indwelling ICT (6.70 ± 1.25 vs. 8.30 ± 2.91, P = 0.127) with higher rate of tube thoracostomy removal (70% vs. 30%, P = 0.003) and lesser need of additional interventions (0% vs. 30%, P = 0.105). Thoracotomy (3 patients) and image-guided drainage (4 patients) were additional interventions to address retained hemothorax in the conventional cohort. However, similar length of ventilator assistance (0.7 ± 0.48 vs. 0.60 ± 1.08, P = 0.791) and prolonged ICU monitoring (1.30 ± 1.06 vs. 0.90 ± 1.45, P = 0.490) was observed in early VATS cohort. Both the cohorts had no mortality. Conclusion: VATS-guided early evacuation of traumatic hemothorax is associated with shorter length of hospital stay along with abbreviated indwelling ICT duration, reduced incidence of complications, lesser readmissions, and improved rate of tube thoracostomy removal. However, the duration of ventilator requirement, ICU stay, and mortality remain unchanged.

An experimental study to assess the effectiveness of video-assisted childbirth education programme on knowledge regarding childbirth among primigravida mothers attending antenatal out patient department of a selected hospital in Bhopal.

Context: Childbirth is a normal physiological process in a woman's life, yet it is a life-changing experience for the primigravida mothers. Boosting confidence regarding pregnancy and childbirth by enhancing knowledge is an important component of care. Aims: The aim was to assess the effectiveness of the video-assisted childbirth education programme on the knowledge of the primigravida mothers regarding childbirth. Settings and Design: A quantitative research approach with a quasi-experimental non-equivalent control group design was adopted. Non-probability convenient sampling technique was used to select the 60 primigravida mothers who completed 34 weeks of gestation and are attending the Gynaec Outpatient Department (OPD) of a tertiary care hospital in Bhopal. Methods and Material: The self-structured knowledge questionnaire was developed. On the first day of contact with the sample, pretest was taken from both groups. A tailor-made video was administered to the experimental group on the same day. Posttest was taken on the seventh day from both groups. Statistical Analysis Used: Repeated measures analysis of variance (ANOVA) used to determine the effectiveness of video-assisted childbirth education programme revealed a significant interaction between the intervention and time, F (1,58) =29.398, P < .001. Results: The mean difference in knowledge score over time was greater in the experimental group compared to the control group and the difference was statistically significant. Conclusions: The video-assisted childbirth education programme was effective in enhancing the knowledge of primigravida mothers regarding childbirth.

Clinical and demographic features of patients undergoing video-capsule endoscopy management: A descriptive study.

BACKGROUND: Video-capsule endoscopy (VCE) is an efficient tool that has proven to be highly useful in approaching several gastrointestinal diseases. VCE was implemented in Colombia in 2003, however current characterization of patients undergoing VCE in Colombia is limited, and mainly comes from two investigations conducted before the SARS-CoV-2 pandemic period. AIM: To describe the characteristics of patients undergoing VCEs and establish the main indications, findings, technical limitations, and other outstanding features. METHODS: A descriptive study was carried out using data from reports of VCE (PillCam SB3 system) use in a Gastroenterology Unit in Bogotá, Colombia between September 2019 and January 2023. Demographic and clinical variables such as indication for the VCE, gastric and small bowel transit times (GTT, SBTT), endoscopic preparation quality, and limitations were described [n (%), median (IQR)]. RESULTS: A total of 133 VCE reports were analyzed. Most were in men with a median age of 70 years. The majority had good preparation (96.2%), and there were technical limitations in 15.8% of cases. The main indications were unexplained anemia (91%) or occult bleeding (23.3%). The median GTT and SBTT were 14 and 30 minutes, respectively. The frequencies of bleeding stigma (3.79%) and active bleeding (9.09%) were low, and the most frequent abnormal findings were red spots (28.3%), erosions (17.6%), and vascular ectasias (12.5%). CONCLUSION: VCE showed high-level safety. The main indication was unexplained anemia. Active bleeding was the most frequent finding. Combined with artificial intelligence, VCE can improve diagnostic precision and targeted therapeutic interventions.

Mediastinal parathyroid cyst: A case report and review of the literature.

Mediastinal parathyroid cysts (MPCs) are extremely rare, benign lesions arising from the parathyroid glands and residing within the thoracic cavity. This study aims to advance understanding of MPC, emphasizing accurate diagnosis and management approaches for this rare condition. A 46-year-old woman presented with dysphagia for one week. Blood tests revealed elevated parathyroid hormone (PTH) (112.8 pg/mL) and normal serum calcium (9.54 mg/dL). Ultrasonography identified a large, well-defined cystic nodule measuring 46 × 30 × 25 mm, extending retro-sternally in the right upper third of the chest. A subsequent high-resolution computed tomography scan of the chest revealed a large space-occupying lesion (47 × 43 × 31 mm) in the superior mediastinum, near the esophagus, suggesting an esophageal duplication cyst or, less likely, a bronchogenic cyst. Video-assisted thoracoscopic surgery (VATS) was performed, and the entire cyst was excised, confirmed histologically as a mediastinal parathyroid cyst. Mediastinal involvement of PCs poses diagnostic challenges due to their rarity and diverse clinical presentations. Surgical excision is necessary for symptomatic cases, with VATS emerging as a favorable approach.

Adherence and Retention Rates to Home-Based Video Exercise Programs in Older Adults-Systematic Review and Meta-Analysis.

Introduction: This systematic review and meta-analysis aimed to investigate adherence and retention rates to home-based video exercise programs and identify key factors associated with these rates in older adults to understand the effectiveness of home-based video exercise interventions. Methods: We searched PubMed, Web of Science, and Scopus for articles addressing adherence to and retention of home-based video exercise programs. The study was conducted following PRISMA recommendations. Results: A total of 26 articles, including 1,292 participants older than 65, were included in the final qualitative and quantitative syntheses. The weighted mean of the retention rate was 91.1, and of the attendance rate was 85.0, with low I2 = 3.5, not significant p = 0.409 heterogeneity. The generalized regression models showed a positive effect of session duration on the attendance rate (%), where the possible change from <20 min to >60 min duration could decrease the attendance rate (%) B = -24.390 (p <0.001). The delivery method had a significant effect, where the absence of live contact with the coach in web-based or DVD-delivered interventions could decrease the attendance rate (%) compared to the online sessions B = -11.482 (p = 0.010). The lockdown during the COVID-19 pandemic had a positive effect on both the attendance rate (%) B = 10.321 (p = 0.019) and retention rate (%) B = 9.577 (p = 0.032). Conclusions: This systematic review and meta-analysis indicate that supervised home-based video exercise programs lasting less than 60 min might be a suitable and sustainable exercise mode to keep older adults active, especially in times resembling feelings of confinement.

Trunk Instability in the Pitch, Yaw, and Roll Planes during Clinical Balance Tests: Axis Differences and Correlations to vHIT Asymmetries Following Acute Unilateral Vestibular Loss.

BACKGROUND: Clinical dynamic posturography concentrates on the pitch and roll but not on the yaw plane instability measures. This emphasis may not represent the axis instability observed in clinical stance and gait tasks for patients with balance deficits in comparison to healthy control (HC) subjects, nor the expected instability based on correlations with vestibulo-ocular reflex (VOR) deficits. To examine the axis stability changes with vestibular loss, we measured trunk sway in all three directions (pitch, roll, and yaw) during the stance and gait tasks of patients with acute unilateral vestibular neuritis (aUVN) and compared the results with those of HC. Concurrent changes in VORs were also examined and correlated with trunk balance deficits. METHODS: The results of 11 patients (mean age of 61 years) recorded within 6 days of aUVN onset were compared within those of 8 age-matched healthy controls (HCs). All subjects performed a two-legged stance task-standing with eyes closed on foam (s2ecf), a semi-gait task-walking eight tandem steps (tan8), and four gait tasks-walking 3 m with head rotating laterally, pitching, or eyes closed (w3hr, w3hp, w3ec), and walking over four barriers 24 cm high, spaced 1 m apart (barr). The tasks' peak-to-peak yaw, pitch and roll angles, and angular velocities were measured with a gyroscope system (SwayStarTM) mounted at L1-3 and combined into three, axis-specific, balance control indexes (BCI), using angles (a) for the tandem gait and barriers task, and angular velocities (v) for all other tasks, as follows: axis BCI = (2 × 2ecf)v + 1.5 × (w3hr + w3hp + w3ec)v + (tan8 + 12 × barr)a. RESULTS: Yaw and pitch BCIs were significantly (p ≤ 0.004) greater (88 and 30%, respectively) than roll BCIs for aUVN patients. For HCs, only yaw but not pitch BCIs were greater (p = 0.002) than those of roll (72%). The order of BCI aUVN vs. HC differences was pitch, yaw, and roll at 55, 44, and 31%, respectively (p ≤ 0.002). This difference with respect to roll corresponded to the known greater yaw plane than roll plane asymmetry (40 vs. 22%) following aUVN based on VOR responses. However, the lower pitch plane asymmetry (3.5%) in VOR responses did not correspond with the pitch plane instability observed in the balance control tests. The increases in pitch plane instability in UVL subjects were, however, highly correlated with those of roll and yaw. CONCLUSIONS: These results indicate that greater yaw than pitch and roll trunk motion during clinical balance tasks is common for aUVN patients and HCs. However, aUVN leads to a larger increase in pitch than yaw plane instability and a smaller increase in roll plane instability. This difference with respect to roll corresponds to the known greater yaw plane than roll plane asymmetry (40 vs. 22%) following aUVN observed in VOR responses. However, the lower pitch plane asymmetry (3.5%) in VOR responses does not correspond with the enhanced movements in the pitch plane, observed in balance control tasks. Whether asymmetries in vestibular-evoked myogenic potentials (Vemps) are better correlated with the deficits in pitch plane balance control remains to be investigated. The current results provide a strong rationale for the clinical testing of directional specific balance responses, especially yaw and pitch, and the linking of balance results for yaw and roll to VOR asymmetries.

Assessing Autistic Traits, Hikikomori Tendencies, Pathological Videogaming, and Eating Disorders in University Students: Are Pathological Videogaming and Eating Disorders Gender-Specific Manifestations of the Autism Spectrum?

In the previous literature, specific attention has been paid to investigate autism spectrum symptoms and traits in university students. In this framework, we aimed to evaluate the presence and correlates of autistic traits, hikikomori tendencies, altered eating behaviors, and pathological videogaming in a sample of Italian university students enrolled in bachelor's degree courses. A total of 1192 students were recruited via an online survey and assessed with the Hikikomori Questionnaire-25, the Adult Autism Subthreshold Spectrum Questionnaire, the Eating Attitude test-26, and the Assessment of Internet and Computer Game Addiction. Our results highlighted significant differences in the prevalence of autistic traits, social withdrawal tendencies, altered eating habits, and pathological videogame use in university students based on gender, age, parents' level of instruction, and field of study. A significant effect of the presence of autistic traits and gender on the scores obtained with the other questionnaires was reported. Our results not only support the role of autistic traits as a vulnerability factor for the development of a set of psychopathological conditions but also suggest that gender could modulate this vulnerability, supporting the hypothesis of gender-specific phenotypes in the autism spectrum.

Video Gaming in Older People: What Are the Implications for Cognitive Functions?

Mild cognitive impairment impacts a sizable segment of the older population, and often evolves into dementia within a few years. At this stage, subjects may benefit from non-pharmacological therapies that can delay or stop the progression of the mild cognitive impairment into dementia and are crucial for improvement in the subject's quality of life, while also being easily accessible and safe for use. Many research studies have shown that a variety of exercises, including cognitive training, have the potential to enhance or optimize cognitive function and general well-being. Recently, many authors have suggested video games as a promising approach for cognitive training and neurorehabilitation in older people, thanks to their increasing motivation and training effects through immersion in stimulating environments. Under this premise, our narrative review's objective is to discuss and summarize the body of existing material on the role of video games in improving cognitive performance, daily life activities, and depression symptoms in older individuals with different levels of cognitive decline. From the papers reviewed, it emerged that older subjects trained with video games showed a significant improvement in cognitive functions, sleep quality, and psychiatric symptoms, positioning video games as an intriguing and useful tool.

Wearable Near-Eye Tracking Technologies for Health: A Review.

With the rapid advancement of computer vision, machine learning, and consumer electronics, eye tracking has emerged as a topic of increasing interest in recent years. It plays a key role across diverse domains including human-computer interaction, virtual reality, and clinical and healthcare applications. Near-eye tracking (NET) has recently been developed to possess encouraging features such as wearability, affordability, and interactivity. These features have drawn considerable attention in the health domain, as NET provides accessible solutions for long-term and continuous health monitoring and a comfortable and interactive user interface. Herein, this work offers an inaugural concise review of NET for health, encompassing approximately 70 related articles published over the past two decades and supplemented by an in-depth examination of 30 literatures from the preceding five years. This paper provides a concise analysis of health-related NET technologies from aspects of technical specifications, data processing workflows, and the practical advantages and limitations. In addition, the specific applications of NET are introduced and compared, revealing that NET is fairly influencing our lives and providing significant convenience in daily routines. Lastly, we summarize the current outcomes of NET and highlight the limitations.

Analysis of YouTube-Based Therapeutic Content for Children with Cerebral Palsy.

BACKGROUND/OBJECTIVES: Cerebral palsy (CP) causes movement and posture challenges due to central nervous system damage, requiring lifelong management. During the COVID-19 pandemic, there was limited access to facility-based treatments, which increased the demand for home-based therapies and digital resources. We analyzed the qualitative and quantitative aspects of YouTube videos focusing on CP therapy for children. METHODS: A total of 95 videos were evaluated for content quality using the modified DISCERN (mDISCERN) tool and Global Quality Scale (GQS). The therapeutic program efficacy was assessed via the International Consensus on Therapeutic Exercise and Training (i-CONTENT) tool, Consensus on Therapeutic Exercise Training (CONTENT) scale, and Consensus on Exercise Reporting Template (CERT), and popularity was measured by the video power index (VPI). RESULTS: YouTube-based therapeutic videos for children with CP generally exhibit reliability in video content and effectiveness in therapeutic programming, and no correlations were found between video popularity and quality. However, the qualitative analysis reveals insufficient mention of uncertainty in the treatment principles within the video content as well as a lack of detailed treatment descriptions encompassing aspects such as intensity, frequency, timing, setting, outcome measurement during and post-treatment, and safety considerations within therapeutic programs. In particular, this tendency was consistent regardless of the uploader's expertise level and the classification of the neuromotor therapy type in contrast to that of the exercise type. CONCLUSIONS: YouTube-based content for CP children still has significant limitations in how substantive viewers, such as caregivers, can acquire tailored information and apply practical information to their exercise and treatment programs.

Evaluation of Problematic Video Game Use in Adolescents with ADHD and without ADHD: New Evidence and Recommendations.

Video game addiction among adolescents, particularly those with ADHD, is a significant concern. To gather more insights into video game usage patterns in this population, we investigated levels of potentially problematic use, passion, motivations, and emotional/behavioral symptoms in adolescents with and without ADHD. Our cross-sectional, multicenter study involved 2513 subjects (Age M = 15.07; SD = 2.82) from 24 schools in Galicia (Spain), including 158 (6.3%) diagnosed with ADHD. We employed an ad hoc questionnaire covering sociodemographic data and ADHD diagnosis, the Questionnaire of Experiences Associated with Video Games (CERV), the scale of passion for video games, reasons for playing video games Questionaire (VMQ), and emotional/behavioral problems assessment (SDQ). Results indicated heightened vulnerability in adolescents with ADHD, manifesting in longer usage durations and higher problematic use scores. Interestingly, ADHD medication did not affect internet gaming disorder development. Motivations for gaming differed between groups, with the ADHD cohort showing distinctions in cognitive development, coping, and violent reward. Additionally, the ADHD group exhibited greater emotional/behavioral symptoms, hyperactivity, and reduced prosocial behavior.

The Impact of Color Cues on the Learning Performance in Video Lectures.

This study explores the learning effects of color cues in video lectures and their underlying mechanisms. With the rapid growth of online education, lifelong learning, and blended learning, video lectures have become integral to teaching and learning. Color, a crucial element in visual design, directs attention, organizes content, and integrates information. Evaluating 78 college students, we assessed learning performance by comparing video content with no-color, single-color, and multi-color cues using eye-tracking technology and cognitive load scales. Results indicate that students viewing videos with color cues demonstrated better retention and transfer test performance, while absence or excess of color cues increased cognitive load. These findings have practical implications for video producers and provide a theoretical foundation for enhancing learners' viewing experience and overall effectiveness. This study not only offers an in-depth analysis of color cue utilization in video lectures, highlighting their positive impact on learning outcomes but also introduces fresh perspectives for educational technology and cognitive psychology research. Future investigations should consider color cue effects in diverse cultural contexts and subject areas, exploring varied strategies to optimize the learning experience.

Exploring Chinese Secondary School Students' Acceptance of Live Video-Streamed Teaching Platforms in EFL Class: An Application of the Technology Acceptance Model.

Live video-streamed teaching platforms are widely used in language teaching. However, how students perceive these platforms has scarcely been investigated. By adopting the Technology Acceptance Model (TAM), this study investigated Chinese secondary school students' perceptions of the platforms (i.e., Tencent meeting, Tencent classroom and Dingtalk) being adopted in English as a foreign language (EFL) class. Gender and age differences were also investigated. Data were collected from 602 students; the results showed the following: (1) The acceptance level of all the participants was high for the five variables in TAM, i.e., perceived ease of use (PEU), perceived usefulness (PU), attitude (ATT), computer self-efficacy (CSE) and behavioral intention to use (BI), but with significant individual differences. There existed no gender differences, while age differences existed between junior high school students and those from senior high school. (2) The five variables were correlated with each other significantly. In addition, CSE, PEU, PU and ATT can predict BI in parallel. (3) The relationship between CSE and BI was mediated by PEU, PU and ATT. Also, PU had the strongest mediating effect, with PEU and ATT exerting slightly lower effects. The theoretical and practical implications are discussed at the end.

Customized Vestibular Rehabilitation for Vestibular Schwannoma Excision via Translabyrinthine Approach: A Single-Center Experience.

Objectives: To evaluate the effectiveness of intensive customized vestibular rehabilitation after vestibular schwannoma (VS) excision. Methods: 52 patients who underwent VS removal via a translabyrinthine approach from 2020 to 2022 were involved in this study. Bedside examination, video head impulse test (vHIT), functional head impulse test (fHIT), and the dizziness handicap inventory (DHI) were performed before and after the rehabilitation, which consisted of 10 sessions of specifically designed vestibular, visual, and physical integrated training. Results: After rehabilitation, the vHIT showed overall unchanged values on the affected and healthy side. In contrast, the scores of fHIT, which explores the higher connection of the vestibular system with visual and cerebellar pathways, improved on both the pathological and healthy sides after training (p-value 0.004 and 0.000, respectively). The effectiveness of the rehabilitation was reinforced by the DHI scores, which were considerably lower after training. Conclusions: To our knowledge, this is the first study to explore fHIT outcomes after removal of VS, estimating the impact of rehabilitation on the overall compensation process. The outcomes support the role of extensive postsurgical rehabilitation in the compensatory process, even just a few days after surgery.

Deep Visual Computing of Behavioral Characteristics in Complex Scenarios and Embedded Object Recognition Applications.

By leveraging artificial intelligence and big data to analyze and assess classroom conditions, we can significantly enhance teaching quality. Nevertheless, numerous existing studies primarily concentrate on evaluating classroom conditions for student groups, often neglecting the need for personalized instructional support for individual students. To address this gap and provide a more focused analysis of individual students in the classroom environment, we implemented an embedded application design using face recognition technology and target detection algorithms. The Insightface face recognition algorithm was employed to identify students by constructing a classroom face dataset and training it; simultaneously, classroom behavioral data were collected and trained, utilizing the YOLOv5 algorithm to detect students' body regions and correlate them with their facial regions to identify students accurately. Subsequently, these modeling algorithms were deployed onto an embedded device, the Atlas 200 DK, for application development, enabling the recording of both overall classroom conditions and individual student behaviors. Test results show that the detection precision for various types of behaviors is above 0.67. The average false detection rate for face recognition is 41.5%. The developed embedded application can reliably detect student behavior in a classroom setting, identify students, and capture image sequences of body regions associated with negative behavior for better management. These data empower teachers to gain a deeper understanding of their students, which is crucial for enhancing teaching quality and addressing the individual needs of students.

Multi-Directional Long-Term Recurrent Convolutional Network for Road Situation Recognition.

Understanding road conditions is essential for implementing effective road safety measures and driving solutions. Road situations encompass the day-to-day conditions of roads, including the presence of vehicles and pedestrians. Surveillance cameras strategically placed along streets have been instrumental in monitoring road situations and providing valuable information on pedestrians, moving vehicles, and objects within road environments. However, these video data and information are stored in large volumes, making analysis tedious and time-consuming. Deep learning models are increasingly utilized to monitor vehicles and identify and evaluate road and driving comfort situations. However, the current neural network model requires the recognition of situations using time-series video data. In this paper, we introduced a multi-directional detection model for road situations to uphold high accuracy. Deep learning methods often integrate long short-term memory (LSTM) into long-term recurrent network architectures. This approach effectively combines recurrent neural networks to capture temporal dependencies and convolutional neural networks (CNNs) to extract features from extensive video data. In our proposed method, we form a multi-directional long-term recurrent convolutional network approach with two groups equipped with CNN and two layers of LSTM. Additionally, we compare road situation recognition using convolutional neural networks, long short-term networks, and long-term recurrent convolutional networks. The paper presents a method for detecting and recognizing multi-directional road contexts using a modified LRCN. After balancing the dataset through data augmentation, the number of video files increased, resulting in our model achieving 91% accuracy, a significant improvement from the original dataset.

Fall Detection Method for Infrared Videos Based on Spatial-Temporal Graph Convolutional Network.

The timely detection of falls and alerting medical aid is critical for health monitoring in elderly individuals living alone. This paper mainly focuses on issues such as poor adaptability, privacy infringement, and low recognition accuracy associated with traditional visual sensor-based fall detection. We propose an infrared video-based fall detection method utilizing spatial-temporal graph convolutional networks (ST-GCNs) to address these challenges. Our method used fine-tuned AlphaPose to extract 2D human skeleton sequences from infrared videos. Subsequently, the skeleton data was represented in Cartesian and polar coordinates and processed through a two-stream ST-GCN to recognize fall behaviors promptly. To enhance the network's recognition capability for fall actions, we improved the adjacency matrix of graph convolutional units and introduced multi-scale temporal graph convolution units. To facilitate practical deployment, we optimized time window and network depth of the ST-GCN, striking a balance between model accuracy and speed. The experimental results on a proprietary infrared human action recognition dataset demonstrated that our proposed algorithm accurately identifies fall behaviors with the highest accuracy of 96%. Moreover, our algorithm performed robustly, identifying falls in both near-infrared and thermal-infrared videos.