ABSTRACT
STUDY OBJECTIVES: The gold standard for diagnosing obstructive sleep apnea (OSA) is polysomnography (PSG). However, PSG is a time-consuming method with clinical limitations. This study aimed to create a wireless radar framework to screen the likelihood of 2 levels of OSA severity (ie, moderate-to-severe and severe OSA) in accordance with clinical practice standards. METHODS: We conducted a prospective, simultaneous study using a wireless radar system and PSG in a Northern Taiwan sleep center, involving 196 patients. The wireless radar sleep monitor, incorporating hybrid models such as deep neural decision trees, estimated the respiratory disturbance index relative to the total sleep time established by PSG (RDIPSG_TST), by analyzing continuous-wave signals indicative of breathing patterns. Analyses were performed to examine the correlation and agreement between the RDIPSG_TST and apnea-hypopnea index, results obtained through PSG. Cut-off thresholds for RDIPSG_TST were determined using Youden's index, and multiclass classification was performed, after which the results were compared. RESULTS: A strong correlation (ρ = 0.91) and agreement (average difference of 0.59 events/h) between apnea-hypopnea index and RDIPSG_TST were identified. In terms of the agreement between the 2 devices, the average difference between PSG-based apnea-hypopnea index and radar-based RDIPSG_TST was 0.59 events/h, and 187 out of 196 cases (95.41%) fell within the 95% confidence interval of differences. A moderate-to-severe OSA model achieved an accuracy of 90.3% (cut-off threshold for RDIPSG_TST: 19.2 events/h). A severe OSA model achieved an accuracy of 92.4% (cut-off threshold for RDIPSG_TST: 28.86 events/h). The mean accuracy of multiclass classification performance using these cut-off thresholds was 83.7%. CONCLUSIONS: The wireless-radar-based sleep monitoring device, with cut-off thresholds, can provide rapid OSA screening with acceptable accuracy and also alleviate the burden on PSG capacity. However, to independently apply this framework, the function of determining the radar-based total sleep time requires further optimizations and verification in future work. CITATION: Lin S-Y, Tsai C-Y, Majumdar A, et al. Combining a wireless radar sleep monitoring device with deep machine learning techniques to assess obstructive sleep apnea severity. J Clin Sleep Med. 2024;20(8):1267-1277.
Subject(s)
Deep Learning , Polysomnography , Radar , Severity of Illness Index , Sleep Apnea, Obstructive , Humans , Sleep Apnea, Obstructive/diagnosis , Sleep Apnea, Obstructive/physiopathology , Male , Prospective Studies , Polysomnography/instrumentation , Polysomnography/methods , Female , Middle Aged , Radar/instrumentation , Wireless Technology/instrumentation , Taiwan , Adult , AgedABSTRACT
OBJECTIVE: To evaluate the muscle thickness and walking test in people with haemophilia A (PWH) and their correlation to joint health and functional impairments. DESIGN: Cross-sectional study. RESULTS: 29 severe/moderate PWH were enrolled. Muscle thickness of quadriceps and medial gastrocnemius were measured using ultrasound. Joint health and functional capacity were assessed using Haemophilia Joint Health Score (HJHS), Haemophilia Early Arthropathy Detection with Ultrasound (HEAD-US), 6-Minute Walking test (6MWT), Haemophilia Quality of Life Questionnaire for Adults (Haem-A-QoL), and Haemophilia Activities List (HAL). Quadriceps muscle thickness significantly correlated with HJHS knee, HEAD-US knee, and HAL. Calf muscle thickness significantly correlated with the HJHS ankle. After adjusted age and BMI, calf muscle thickness was inversely associated with the HJHS ankle. 6MWT was found to significantly correlate with HJHS total, HEAD-US total, Haem-A-QoL, and HAL. CONCLUSION: Muscle thickness and the distance of 6MWT were linked to assessment of joint health, quality of life and activity participation in PWH. Ultrasound measurement of muscle thickness and walking test appear to be useful tools for the assessment of joint health and functional status in PWH.
ABSTRACT
Introduction: Pulse harmonic analysis is a quantitative and objective methodology within traditional Chinese medicine (TCM) used to evaluate pulse characteristics. However, interpreting pulse wave data is challenging due to its inherent complexity. This study aims to provide a comprehensive review and comparison of existing human pulse wave harmonic analysis methods to elucidate their patterns and characteristics. Methods: A systematic review of clinical research reports published from 1990 to 2021 was conducted, focusing on variations in harmonic characteristics across different medical conditions and physiological states. Keyword searches included terms related to analysis methods (e.g., "Pulse Spectrum," "harmonic analysis," "harmonic index") and measured indicators (e.g., "vascular response," "PPG," "Photoplethysmography," "aortic," "arterial," "blood pressure"). Supplementary research using PubMed's Mesh terms specifically targeted "Pulse wave analysis" within the methods and statistical analysis domain. Articles were filtered based on predefined criteria, including human participants and research related to pulse pressure or vascular volume changes. Conference papers, animal studies, and irrelevant research were excluded, with literature evaluation scales selected based on the retrieved research reports. Results: Initially, 6487 research reports were identified, and after screening, 50 reports were included in the review. The analysis revealed that low-frequency harmonics increase following vigorous activity or sympathetic excitation but decrease during rest or parasympathetic excitation. Cardiovascular patients exhibited elevated first harmonics associated with the liver meridian, while diabetes patients displayed weakened third harmonics related to the spleen meridian. Liver dysfunction was linked to changes in the first harmonic, and cancer patients showed signs of liver and kidney yin deficiency in the first and second harmonics. These findings underscore the potential of harmonic analysis for TCM disease diagnosis and organ assessment. Moreover, individuals with conditions such as liver dysfunction, cancer, and gynecological disorders displayed distinct intensity patterns across harmonics one through ten compared to healthy controls, albeit with some variations. Heterogeneity in these studies mainly stemmed from differences in measurement methods and study populations. Additionally, research suggested that factors like blood circulation and cognitive activity influenced harmonic intensity. Conclusions: In summary, this report consolidates prior research on pulse wave harmonics analysis, revealing unique patterns associated with various physiological conditions. Despite limitations, such as limited sample sizes in previous studies, the observed associations between physiological states and harmonics hold promise for potential clinical applications. This study lays a solid foundation for future applications of arterial wave harmonics analysis, promoting wider adoption of this analytical approach.
ABSTRACT
Common pediatric distal forearm fractures necessitate precise detection. To support prompt treatment planning by clinicians, our study aimed to create a multi-class convolutional neural network (CNN) model for pediatric distal forearm fractures, guided by the AO Foundation/Orthopaedic Trauma Association (AO/ATO) classification system for pediatric fractures. The GRAZPEDWRI-DX dataset (2008-2018) of wrist X-ray images was used. We labeled images into four fracture classes (FRM, FUM, FRE, and FUE with F, fracture; R, radius; U, ulna; M, metaphysis; and E, epiphysis) based on the pediatric AO/ATO classification. We performed multi-class classification by training a YOLOv4-based CNN object detection model with 7006 images from 1809 patients (80% for training and 20% for validation). An 88-image test set from 34 patients was used to evaluate the model performance, which was then compared to the diagnosis performances of two readers-an orthopedist and a radiologist. The overall mean average precision levels on the validation set in four classes of the model were 0.97, 0.92, 0.95, and 0.94, respectively. On the test set, the model's performance included sensitivities of 0.86, 0.71, 0.88, and 0.89; specificities of 0.88, 0.94, 0.97, and 0.98; and area under the curve (AUC) values of 0.87, 0.83, 0.93, and 0.94, respectively. The best performance among the three readers belonged to the radiologist, with a mean AUC of 0.922, followed by our model (0.892) and the orthopedist (0.830). Therefore, using the AO/OTA concept, our multi-class fracture detection model excelled in identifying pediatric distal forearm fractures.
ABSTRACT
IMPORTANCE: Transitioning from the hospital to the community poses significant challenges for stroke survivors and their caregivers. OBJECTIVE: To examine the feasibility and preliminary effects of a dyad-focused strategy training intervention. DESIGN: Single-arm trial with data collection at baseline, postintervention, and 3-mo follow-up. SETTING: Rehabilitation settings in Taiwan. PARTICIPANTS: Sixteen stroke survivor-caregiver dyads. INTERVENTIONS: Dyad-focused strategy training was provided to stroke survivor-caregiver dyads twice a week over 6 wk. The training included shared decision-making, goal setting, performance evaluation, strategy development and implementation, and therapeutic guided discovery. OUTCOMES AND MEASURES: Feasibility indicators were Goal Attainment Scaling, Dyadic Relationship Scale, Participation Measure-3 Domains, 4 Dimensions, Activity Measure for Post-Acute Care, Montreal Cognitive Assessment, Trail Making Test, Stroop Color and Word Test, Preparedness for Caregiving Scale, and Zarit Burden Interview. RESULTS: In total, 15 dyads completed all intervention sessions with full attendance. Both stroke survivors and their caregivers demonstrated high engagement and comprehension and reported moderate to high satisfaction with the intervention. From baseline to postintervention, the effects on goal attainment, frequency and perceived difficulty of community participation, executive function, mobility function, and caregiver preparedness were significant and positive. CONCLUSIONS AND RELEVANCE: Our study supports the feasibility and preliminary efficacy of dyad-focused strategy training for stroke survivor-caregiver dyads transitioning from the hospital to the community in Taiwan. Our preliminary evidence indicates that dyads who receive strategy training exhibit advancement toward their goals and experience considerable enhancements in their individual outcomes. Plain-Language Summary: This study addresses the scarcity of interventions catering to both stroke survivors and their caregivers. By demonstrating the feasibility of our dyad-focused intervention, the research offers preliminary evidence that supports the potential advantages of involving both stroke survivors and their caregivers in the intervention process.
Subject(s)
Stroke Rehabilitation , Stroke , Humans , Caregivers/psychology , Feasibility Studies , Stroke/psychology , Survivors/psychologyABSTRACT
PURPOSE: Strategy training is a rehabilitation intervention that aims to enhance problem-solving skills with respect to daily activity-related challenges and has achieved favorable results in Western countries. This study explored the perspectives of individuals with acquired brain injury (ABI) in Taiwan who received strategy training. MATERIALS AND METHODS: Semi-structured interviews with community-dwelling adults with ABI were conducted, and reflective memos made by research team members were recorded. Interviews and memos were analyzed through thematic analysis. RESULTS: This study included 55 participants. The analysis of the participants' interview responses and memos yielded nine themes under three categories: 1) expectations regarding strategy training, 2) perceived benefits of strategy training, and 3) barriers affecting the process and outcomes of strategy training. CONCLUSIONS: All the participants endorsed strategy training through different gains. Most participants' expectations before the intervention were uncertain. Including family members into the strategy training is of key importance for a successfulness of their goals. The participants' experiences about strategy training were affected by various barriers (i.e., health and medical problems, the physical environment, and natural events). Clinicians and researchers should consider these expectations, benefits, and barriers when studying and implementing strategy training in non-Western contexts.IMPLICATIONS FOR REHABILITATIONStrategy training provides clients the opportunity to actively engage in their own goal setting and decision making.Strategy training increases the client's confidence in their ability to participate in the community, communicate, and perform daily living and physical activities.Therapists should consider the health conditions and physical environment of clients when helping them set goals and before facilitating their engagement in the community.Taiwanese family members play a crucial role in supporting acquired brain injury survivors in strategy training.
Subject(s)
Activities of Daily Living , Brain Injuries , Adult , Humans , Taiwan , Independent Living , Brain Injuries/rehabilitation , Family , Qualitative ResearchABSTRACT
An altered brain-gut axis is suspected to be one of the pathomechanisms in fibromyalgia (FM). This cross-sectional study investigated the associations among altered microbiota, psychological distress, and brain functional connectivity (FC) in FM. We recruited 25 FM patients and 25 healthy people in the present study. Psychological distress was measured using standardized questionnaires. Microbiota analysis was performed on the participants' stools. Functional magnetic resonance imaging data were acquired, and seed-based resting-state FC (rs-FC) analysis was conducted with the salience network nodes as seeds. Linear regression and mediation analyses evaluated microbiota, symptoms, and rs-FCs associations. We found altered microbiota diversity in FM, of which Phascolarctobacterium and Lachnoclostridium taxa increased the most and Faecalibacterium taxon decreased the most compared to controls. The Phascolarctobacterium abundance significantly predicted Beck depression inventory (BDI-II) scores in FM (ß = 6.83; P = .033). Rs-FCs from salience network nodes were reduced in FM, of which rs-FCs from the right lateral rostral prefrontal cortex (RPFC) to the lateral occipital cortex, superior division right (RPFC-sLOC) could be predicted by BDI-II scores in patients (ß = -.0064; P = .0054). In addition, the BDI-II score was a mediator in the association between Phascolarctobacterium abundance and rs-FCs of RPFC-sLOC (ab = -.06; 95% CI: -.16 to -9.10-3). In conclusion, microbial dysbiosis might be associated with altered neural networks mediated by psychological distress in FM, emphasizing the critical role of the brain-gut axis in FM's non-pain symptoms and supporting further analysis of mechanism-targeted therapies to reduce FM symptoms. PERSPECTIVE: Our study suggests microbial dysbiosis might be associated with psychological distress and the altered salience network, supporting the role of brain-gut axis dysfunction in fibromyalgia pathomechanisms. Further targeting therapies for microbial dysbiosis should be investigated to manage fibromyalgia patients in the future.
Subject(s)
Fibromyalgia , Psychological Distress , Humans , Fibromyalgia/diagnostic imaging , Fibromyalgia/complications , Brain-Gut Axis , Cross-Sectional Studies , Dysbiosis , Magnetic Resonance Imaging , BrainABSTRACT
OBJECTIVE: To identify meaningful changes in patients in 3 functional domains (basic mobility [BM], daily activity [DA], and applied cognition [AC]) after discharge from inpatient stroke rehabilitation and to identify the predictors of 1-year functional improvement. DESIGN: A longitudinal, multicenter, prospective cohort study. SETTING: The acute care wards of 3 hospitals in the Greater Taipei area of Taiwan. PARTICIPANTS: Five hundred patients with stroke in acute care wards (mean age=60±12.2 years, 62% men, N=500). INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURE(S): The Mandarin version of the Activity Measure for Post-Acute Care (AM-PAC) short forms were assessed at discharge and 3-, 6-, and 12-month follow-up. The minimal detectable change (MDC) was used to categorize changes in the scores as improved and unimproved at the 4 time points. RESULTS: The mean scores of the AM-PAC BM and DA subscales substantially increased over the first 3 months after discharge (86% of participants exhibited improvement) and slightly increased during the subsequent 9 months (5â¼26% of participants exhibited improvement). However, the mean score of the AC subscale decreased within the first 3 months and increased over the subsequent 9 months (22-23% of participants exhibited improvement). The BM, AC scores at discharge were the dominant predictors of subsequent functional improvement (P<.05). Patients with a higher functional stage at discharge were more likely to experience significant improvement. CONCLUSION: This study established the capacity of the AM-PAC to predict functional improvement in 3 domains during the early, middle, and late stages of recovery. The findings can assist clinicians in identifying patients at risk of unfavorable long-term functional recovery and providing such patients with tailored interventions during the early stage of rehabilitation.
Subject(s)
Stroke Rehabilitation , Stroke , Male , Humans , Middle Aged , Aged , Female , Inpatients , Longitudinal Studies , Prospective StudiesABSTRACT
OBJECTIVES: To prospectively investigate associations of frailty and other predictor variables with functional recovery and health outcomes in middle-aged and older patients with trauma. DESIGN: Single-center prospective cohort study. SETTING: Emergency department of Wan Fang Hospital in Taiwan. PARTICIPANTS: Trauma patients aged 45 and older. MEASUREMENTS: Frailty was assessed with the Clinical Frailty Scale (CFS). Injury mechanisms, pre-existing diseases, and fracture locations were recorded at baseline. The primary outcome was functional recovery assessed using the Barthel Index (BI). Secondary outcomes were new care needs, unscheduled return visits, and falls 3 months postinjury. RESULTS: A total of 588 participants were included in the final analysis. For every one-point increase in the CFS, the multivariable-adjusted odds ratio (OR, 95% confidence interval [CI]) of failure to retain the preinjury BI was 1.34 (1.16-1.55); associations were consistent across levels of age and injury severities. Significant joint associations of frailty and age with poor functional recovery were observed. CFS was also associated with new care needs (OR for every one-point increase, 1.36, 95% CI, 1.17-1.58), unscheduled return visits (OR 1.26, 95% CI, 1.04-1.51), and falls (OR 1.23, 95% CI, 1.01-1.51). Other variables associated with failure to retain preinjury BI included road traffic accident and presence of hip fracture. CONCLUSION: Frailty was significantly associated with poor functional and health outcomes regardless of injury severity in middle-aged and older patients with trauma. Injury mechanisms and fracture locations were also significant predictors of functional recovery postinjury.
Subject(s)
Fractures, Bone , Frailty , Aged , Humans , Middle Aged , Frailty/epidemiology , Prospective Studies , Geriatric Assessment , Taiwan/epidemiologyABSTRACT
A new approach to treating vascular blockages has been developed to overcome the limitations of current thrombolytic therapies. This approach involves biosafety and multimodal plasma-derived theranostic platelet vesicle incorporating iron oxide constructed nano-propellers platformed technology that possesses fluorescent and magnetic features and manifold thrombus targeting modes. The platform is capable of being guided and visualized remotely to specifically target thrombi, and it can be activated using near-infrared phototherapy along with an actuated magnet for magnetotherapy. In a murine model of thrombus lesion, this proposed multimodal approach showed an approximately 80 % reduction in thrombus residues. Moreover, the new strategy not only improves thrombolysis but also boosts the rate of lysis, making it a promising candidate for time-sensitive thrombolytic therapy.
ABSTRACT
BACKGROUND: Mobile phones, laptops, and computers have become an indispensable part of our lives in recent years. Workers may have an incorrect posture when using a computer for a prolonged period of time. Using these products with an incorrect posture can lead to neck pain. However, there are limited data on postures in real-life situations. METHODS: In this study, we used a common camera to record images of subjects carrying out three different tasks (a typing task, a gaming task, and a video-watching task) on a computer. Different artificial intelligence (AI)-based pose estimation approaches were applied to analyze the head's yaw, pitch, and roll and coordinate information of the eyes, nose, neck, and shoulders in the images. We used machine learning models such as random forest, XGBoost, logistic regression, and ensemble learning to build a model to predict whether a subject had neck pain by analyzing their posture when using the computer. RESULTS: After feature selection and adjustment of the predictive models, nested cross-validation was applied to evaluate the models and fine-tune the hyperparameters. Finally, the ensemble learning approach was utilized to construct a model via bagging, which achieved a performance with 87% accuracy, 92% precision, 80.3% recall, 95.5% specificity, and an AUROC of 0.878. CONCLUSIONS: We developed a predictive model for the identification of non-specific neck pain using 2D video images without the need for costly devices, advanced environment settings, or extra sensors. This method could provide an effective way for clinically evaluating poor posture during real-world computer usage scenarios.
ABSTRACT
BACKGROUND: Air pollution may alter body water distribution, it may also be linked to low-arousal-threshold obstructive sleep apnoea (low-ArTH OSA). Here, we explored the mediation effects of air pollution on body water distribution and low-ArTH OSA manifestations. METHODS: In this retrospective study, we obtained sleep centre data from healthy participants and patients with low-ArTH OSA (N=1924) in northern Taiwan. Air pollutant exposure at different time intervals (1, 3, 6 and 12 months) was estimated using the nearest station estimation method, and government air-quality data were also obtained. Regression models were used to assess the associations of estimated exposure, sleep disorder indices and body water distribution with the risk of low-ArTH OSA. Mediation analysis was performed to explore the relationships between air pollution, body water distribution and sleep disorder indices. RESULTS: First, exposure to particulate matter (PM) with a diameter of ≤10 µm (PM10) for 1 and 3 months and exposure to PM with a diameter of ≤2.5 µm (PM2.5) for 3 months were significantly associated with the Apnoea-Hypopnoea Index (AHI), Oxygen Desaturation Index (ODI), Arousal Index (ArI) and intracellular-to-extracellular water ratio (I-E water ratio). Significant associations were observed between the risk of low-ArTH OSA and 1- month exposure to PM10 (OR 1.42, 95% CI 1.09 to 1.84), PM2.5 (OR 1.33, 95% CI 1.02 to 1.74) and ozone (OR 1.27, 95% CI 1.01 to 1.6). I-E water ratio alternation caused by 1-month exposure to PM10 and 3-month exposure to PM2.5 and PM10 had partial mediation effects on AHI and ODI. CONCLUSION: Air pollution can directly increase sleep disorder indices (AHI, ODI and ArI) and alter body water distribution, thus mediating the risk of low-ArTH OSA.
Subject(s)
Air Pollutants , Sleep Apnea, Obstructive , Humans , Air Pollutants/adverse effects , Air Pollutants/analysis , Retrospective Studies , Body Water/chemistry , Sleep Apnea, Obstructive/epidemiology , Particulate Matter/adverse effects , Particulate Matter/analysis , Oxygen , Arousal , WaterABSTRACT
Objective: Obstructive sleep apnea is a global health concern, and several tools have been developed to screen its severity. However, most tools focus on respiratory events instead of sleep arousal, which can also affect sleep efficiency. This study employed easy-to-measure parameters-namely heart rate variability, oxygen saturation, and body profiles-to predict arousal occurrence. Methods: Body profiles and polysomnography recordings were collected from 659 patients. Continuous heart rate variability and oximetry measurements were performed and then labeled based on the presence of sleep arousal. The dataset, comprising five body profiles, mean heart rate, six heart rate variability, and five oximetry variables, was then split into 80% training/validation and 20% testing datasets. Eight machine learning approaches were employed. The model with the highest accuracy, area under the receiver operating characteristic curve, and area under the precision recall curve values in the training/validation dataset was applied to the testing dataset and to determine feature importance. Results: InceptionTime, which exhibited superior performance in predicting sleep arousal in the training dataset, was used to classify the testing dataset and explore feature importance. In the testing dataset, InceptionTime achieved an accuracy of 76.21%, an area under the receiver operating characteristic curve of 84.33%, and an area under the precision recall curve of 86.28%. The standard deviations of time intervals between successive normal heartbeats and the square roots of the means of the squares of successive differences between normal heartbeats were predominant predictors of arousal occurrence. Conclusions: The established models can be considered for screening sleep arousal occurrence or integrated in wearable devices for home-based sleep examination.
ABSTRACT
Background: Exposure to air pollution may be a risk factor for obstructive sleep apnea (OSA) because air pollution may alter body water distribution and aggravate OSA manifestations. Objectives: This study aimed to investigate the mediating effects of air pollution on the exacerbation of OSA severity through body water distribution. Methods: This retrospective study analyzed body composition and polysomnographic data collected from a sleep center in Northern Taiwan. Air pollution exposure was estimated using an adjusted nearest method, registered residential addresses, and data from the databases of government air quality motioning stations. Next, regression models were employed to determine the associations between estimated air pollution exposure levels (exposure for 1, 3, 6, and 12 months), OSA manifestations (sleep-disordered breathing indices and respiratory event duration), and body fluid parameters (total body water and body water distribution). The association between air pollution and OSA risk was determined. Results: Significant associations between OSA manifestations and short-term (1 month) exposure to PM2.5 and PM10 were identified. Similarly, significant associations were identified among total body water and body water distribution (intracellular-to-extracellular body water distribution), short-term (1 month) exposure to PM2.5 and PM10, and medium-term (3 months) exposure to PM10. Body water distribution might be a mediator that aggravates OSA manifestations, and short-term exposure to PM2.5 and PM10 may be a risk factor for OSA. Conclusion: Because exposure to PM2.5 and PM10 may be a risk factor for OSA that exacerbates OSA manifestations and exposure to particulate pollutants may affect OSA manifestations or alter body water distribution to affect OSA manifestations, mitigating exposure to particulate pollutants may improve OSA manifestations and reduce the risk of OSA. Furthermore, this study elucidated the potential mechanisms underlying the relationship between air pollution, body fluid parameters, and OSA severity.
Subject(s)
Air Pollutants , Air Pollution , Environmental Pollutants , Sleep Apnea, Obstructive , Humans , Air Pollutants/adverse effects , Air Pollutants/analysis , Air Pollution/analysis , Environmental Exposure/adverse effects , Particulate Matter/adverse effects , Particulate Matter/analysis , Retrospective Studies , Sleep Apnea, Obstructive/epidemiology , Body WaterABSTRACT
Introduction: Age-specific risk factors may delay posttraumatic functional recovery; complex interactions exist between these factors. In this study, we investigated the prediction ability of machine learning models for posttraumatic (6 months) functional recovery in middle-aged and older patients on the basis of their preexisting health conditions. Methods: Data obtained from injured patients aged ≥45 years were divided into training-validation (n = 368) and test (n = 159) data sets. The input features were the sociodemographic characteristics and baseline health conditions of the patients. The output feature was functional status 6 months after injury; this was assessed using the Barthel Index (BI). On the basis of their BI scores, the patients were categorized into functionally independent (BI >60) and functionally dependent (BI ≤60) groups. The permutation feature importance method was used for feature selection. Six algorithms were validated through cross-validation with hyperparameter optimization. The algorithms exhibiting satisfactory performance were subjected to bagging to construct stacking, voting, and dynamic ensemble selection models. The best model was evaluated on the test data set. Partial dependence (PD) and individual conditional expectation (ICE) plots were created. Results: In total, nineteen of twenty-seven features were selected. Logistic regression, linear discrimination analysis, and Gaussian Naive Bayes algorithms exhibited satisfactory performances and were, therefore, used to construct ensemble models. The k-Nearest Oracle Elimination model outperformed the other models when evaluated on the training-validation data set (sensitivity: 0.732, 95% CI: 0.702-0.761; specificity: 0.813, 95% CI: 0.805-0.822); it exhibited compatible performance on the test data set (sensitivity: 0.779, 95% CI: 0.559-0.950; specificity: 0.859, 95% CI: 0.799-0.912). The PD and ICE plots showed consistent patterns with practical tendencies. Conclusion: Preexisting health conditions can predict long-term functional outcomes in injured middle-aged and older patients, thus predicting prognosis and facilitating clinical decision-making.
Subject(s)
Algorithms , Machine Learning , Middle Aged , Humans , Aged , Bayes Theorem , Risk Factors , PrognosisABSTRACT
OBJECTIVE: This study proposed a moving average (MA) approach to dynamically process heart rate variability (HRV) and developed aberrant driving behavior (ADB) prediction models by using long short-term memory (LSTM) networks. BACKGROUND: Fatigue-associated ADBs have traffic safety implications. Numerous models to predict such acts based on physiological responses have been developed but are still in embryonic stages. METHOD: This study recorded the data of 20 commercial bus drivers during their routine tasks on four consecutive days and subsequently asked them to complete questionnaires, including subjective sleep quality, driver behavior questionnaire and the Karolinska Sleepiness Scale. Driving behaviors and corresponding HRV were determined using a navigational mobile application and a wristwatch. The dynamic-weighted MA (DWMA) and exponential-weighted MA were used to process HRV in 5-min intervals. The data were independently separated for training and testing. Models were trained with 10-fold cross-validation strategy, their accuracies were evaluated, and Shapley additive explanation (SHAP) values were used to determine feature importance. RESULTS: Significant increases in the standard deviation of NN intervals (SDNN), root mean square of successive heartbeat interval differences (RMSSD), and normalized spectrum of high frequency (nHF) were observed in the pre-event stage. The DWMA-based model exhibited the highest accuracy for both driver types (urban: 84.41%; highway: 80.56%). The SDNN, RMSSD, and nHF demonstrated relatively high SHAP values. CONCLUSION: HRV metrics can serve as indicators of mental fatigue. DWMA-based LSTM could predict the occurrence of the level of fatigue associated with ADBs. APPLICATION: The established models can be used in realistic driving scenarios.
ABSTRACT
OBJECTIVE: Abnormal functional connectivity (FC) and structure in the brain are found in patients with fibromyalgia (FM). This study investigated FC and structural alterations of the visual cortical system, the emerging contributor to pain processing, in patients with FM. METHODS: Thirty pain-free participants and 26 patients with FM were enrolled. Clinical characteristics were evaluated using standardized scales. Structural and resting-state functional magnetic resonance imaging were conducted. Seed-based FC analyses, voxel-based morphometry, and surface-based morphometry were performed. The FC and cortical structure of the visual system were compared between the 2 groups. The correlation between functional and structural changes in the visual cortical system with clinical presentation in the FM group was analyzed. RESULTS: The patients with FM showed increased FCs within visual networks, of which the FC between the visual medial network and the right lingual gyrus (LG) was positively correlated with the Fibromyalgia Impact Questionnaire (FIQ) score. However, the FM group showed decreased FCs from the visual occipital network (VON) to several regions, of which the FCs from the VON to the bilateral frontal orbital cortices were negatively correlated with the FIQ and Pittsburgh Sleep Quality Index scores. Cortical thickness of the lateral occipital cortex, LG, and pericalcarine in FM tended to increase. CONCLUSION: Altered FCs and structure in the visual cortical system might be involved in the pathomechanisms and clinical presentation in FM. These findings could potentially support further studies that seek to find diagnostic methods and mechanism-based therapies in patients with FM.
Subject(s)
Fibromyalgia , Humans , Fibromyalgia/pathology , Magnetic Resonance Imaging/methods , Brain , PainABSTRACT
Obstructive sleep apnea (OSA) with a low arousal threshold (low-ArTH) phenotype can cause minor respiratory events that exacerbate sleep fragmentation. Although anthropometric features may affect the risk of low-ArTH OSA, the associations and underlying mechanisms require further investigation. This study investigated the relationships of body fat and water distribution with polysomnography parameters by using data from a sleep center database. The derived data were classified as those for low-ArTH in accordance with criteria that considered oximetry and the frequency and type fraction of respiratory events and analyzed using mean comparison and regression approaches. The low-ArTH group members (n = 1850) were significantly older and had a higher visceral fat level, body fat percentage, trunk-to-limb fat ratio, and extracellular-to-intracellular (E-I) water ratio compared with the non-OSA group members (n = 368). Significant associations of body fat percentage (odds ratio [OR]: 1.58, 95% confident interval [CI]: 1.08 to 2.3, p < 0.05), trunk-to-limb fat ratio (OR: 1.22, 95% CI: 1.04 to 1.43, p < 0.05), and E-I water ratio (OR: 1.32, 95% CI: 1.08 to 1.62, p < 0.01) with the risk of low-ArTH OSA were noted after adjustments for sex, age, and body mass index. These observations suggest that increased truncal adiposity and extracellular water are associated with a higher risk of low-ArTH OSA.
ABSTRACT
Obstructive sleep apnea (OSA) is a risk factor for neurodegenerative diseases. This study determined whether continuous positive airway pressure (CPAP), which can alleviate OSA symptoms, can reduce neurochemical biomarker levels. Thirty patients with OSA and normal cognitive function were recruited and divided into the control (n = 10) and CPAP (n = 20) groups. Next, we examined their in-lab sleep data (polysomnography and CPAP titration), sleep-related questionnaire outcomes, and neurochemical biomarker levels at baseline and the 3-month follow-up. The paired t-test and Wilcoxon signed-rank test were used to examine changes. Analysis of covariance (ANCOVA) was performed to increase the robustness of outcomes. The Epworth Sleepiness Scale and Pittsburgh Sleep Quality Index scores were significantly decreased in the CPAP group. The mean levels of total tau (T-Tau), amyloid-beta-42 (Aß42), and the product of the two (Aß42 × T-Tau) increased considerably in the control group (ΔT-Tau: 2.31 pg/mL; ΔAß42: 0.58 pg/mL; ΔAß42 × T-Tau: 48.73 pg2/mL2), whereas the mean levels of T-Tau and the product of T-Tau and Aß42 decreased considerably in the CPAP group (ΔT-Tau: -2.22 pg/mL; ΔAß42 × T-Tau: -44.35 pg2/mL2). The results of ANCOVA with adjustment for age, sex, body mass index, baseline measurements, and apnea-hypopnea index demonstrated significant differences in neurochemical biomarker levels between the CPAP and control groups. The findings indicate that CPAP may reduce neurochemical biomarker levels by alleviating OSA symptoms.
ABSTRACT
Acute epiglottitis (AE) is a life-threatening condition and needs to be recognized timely. Diagnosis of AE with a lateral neck radiograph yields poor reliability and sensitivity. Convolutional neural networks (CNN) are powerful tools to assist the analysis of medical images. This study aimed to develop an artificial intelligence model using CNN-based transfer learning to identify AE in lateral neck radiographs. All cases in this study are from two hospitals, a medical center, and a local teaching hospital in Taiwan. In this retrospective study, we collected 251 lateral neck radiographs of patients with AE and 936 individuals without AE. Neck radiographs obtained from patients without and with AE were used as the input for model transfer learning in a pre-trained CNN including Inception V3, Densenet201, Resnet101, VGG19, and Inception V2 to select the optimal model. We used five-fold cross-validation to estimate the performance of the selected model. The confusion matrix of the final model was analyzed. We found that Inception V3 yielded the best results as the optimal model among all pre-train models. Based on the average value of the fivefold cross-validation, the confusion metrics were obtained: accuracy = 0.92, precision = 0.94, recall = 0.90, and area under the curve (AUC) = 0.96. Using the Inception V3-based model can provide an excellent performance to identify AE based on radiographic images. We suggest using the CNN-based model which can offer a non-invasive, accurate, and fast diagnostic method for AE in the future.