Cognitive screening test for rehabilitation using spatiotemporal data extracted from a digital trail making test part-A.

We investigated a newly developed digitized Trail Making Test using an iPad (iTMT) as a brief cognitive function screening test. We found that the iTMT part-A (iTMT-A) can estimate generalized cognitive function in rehabilitation inpatients examined using the Mini-Mental State Examination (MMSE). Forty-two hospitalized participants undergoing rehabilitation (rehab participants), 30 of whom had cerebral infarction/hemorrhage (stroke participants), performed the iTMT five times (first three times: iTMT-A; fourth: paper version of TMT-A; fifth: the inverse version of iTMT-A) and the MMSE once. Each iTMT-A trial's completion time was divided into the move and dwell times. A linear mixed model following post-hoc tests revealed that the completion time of the third and fourth iTMT-A was faster compared to that of the first iTMT-A, suggesting the presence of a learning effect. In the partial least squares (PLS) regression analysis, the coefficient of determination for estimating the MMSE score was increased by using the dwell and move times extracted from the repeated iTMT-A and the availability of TMT-B, even for subjects with low MMSE scores. These findings indicate that the dwell time of iTMT-A may be important for estimating cognitive function. The iTMT-A extracts significant factors temporally and spatially, and by incorporating the learning effect of repeated trials, it may be possible to screen cognitive and physical functions for rehabilitation patients.

The Number of Microbubbles Generated During Cardiopulmonary Bypass Can Be Estimated Using Machine Learning From Suction Flow Rate, Venous Reservoir Level, Perfusion Flow Rate, Hematocrit Level, and Blood Temperature.

GOAL: Microbubbles (MBs) are known to occur within the circuits of cardiopulmonary bypass (CPB) systems, and higher-order dysfunction after cardiac surgery may be caused by MBs as well as atheroma dispersal associated with cannula insertion. As complete MB elimination is not possible, monitoring MB count rates is critical. We propose an online detection system with a neural network-based model to estimate MB count rate using five parameters: suction flow rate, venous reservoir level, perfusion flow rate, hematocrit level, and blood temperature. METHODS: Perfusion experiments were performed using an actual CPB circuit, and MB count rates were measured using the five varying parameters. RESULTS: Bland-Altman analysis indicated a high estimation accuracy (R2 > 0.95, p < 0.001) with no significant systematic error. In clinical practice, although the inclusion of clinical procedures slightly decreased the estimation accuracy, a high coefficient of determination for 30 clinical cases (R2 = 0.8576) was achieved between measured and estimated MB count rates. CONCLUSIONS: Our results highlight the potential of this system to improve patient outcomes and reduce MB-associated complication risk.

Evaluating Classifier Confidence for Surface EMG Pattern Recognition.

Surface electromyogram (EMG) can be employed as an interface signal for various devices and software via pattern recognition. In EMG-based pattern recognition, the classifier should not only be accurate, but also output an appropriate confidence (i.e., probability of correctness) for its prediction. If the confidence accurately reflects the likelihood of true correctness, then it will be useful in various application tasks, such as motion rejection and online adaptation. The aim of this paper is to identify the types of classifiers that provide higher accuracy and better confidence in EMG pattern recognition. We evaluate the performance of various discriminative and generative classifiers on four EMG datasets, both visually and quantitatively. The analysis results show that while a discriminative classifier based on a deep neural network exhibits high accuracy, it outputs a confidence that differs from true probabilities. By contrast, a scale mixture model-based classifier, which is a generative classifier that can account for uncertainty in EMG variance, exhibits superior performance in terms of both accuracy and confidence.

Motor-cognitive functions required for driving in post-stroke individuals identified via machine-learning analysis.

BACKGROUND: People who were previously hospitalised with stroke may have difficulty operating a motor vehicle, and their driving aptitude needs to be evaluated to prevent traffic accidents in today's car-based society. Although the association between motor-cognitive functions and driving aptitude has been extensively studied, motor-cognitive functions required for driving have not been elucidated. METHODS: In this paper, we propose a machine-learning algorithm that introduces sparse regularization to automatically select driving aptitude-related indices from 65 input indices obtained from 10 tests of motor-cognitive function conducted on 55 participants with stroke. Indices related to driving aptitude and their required tests can be identified based on the output probability of the presence or absence of driving aptitude to provide evidence for identifying subjects who must undergo the on-road driving test. We also analyzed the importance of the indices of motor-cognitive function tests in evaluating driving aptitude to further clarify the relationship between motor-cognitive function and driving aptitude. RESULTS: The experimental results showed that the proposed method achieved predictive evaluation of the presence or absence of driving aptitude with high accuracy (area under curve 0.946) and identified a group of indices of motor-cognitive function tests that are strongly related to driving aptitude. CONCLUSIONS: The proposed method is able to effectively and accurately unravel driving-related motor-cognitive functions from a panoply of test results, allowing for autonomous evaluation of driving aptitude in post-stroke individuals. This has the potential to reduce the number of screening tests required and the corresponding clinical workload, further improving personal and public safety and the quality of life of individuals with stroke.

Spatiotemporal patterns of spontaneous movement in neonates are significantly linked to risk of autism spectrum disorders at 18 months old.

Infants make spontaneous movements from the prenatal period. Several studies indicate that an atypical pattern of body motion during infancy could be utilized as an early biomarker of autism spectrum disorders (ASD). However, to date, little is known about whether the body motion pattern in neonates is associated with ASD risk. The present study sought to clarify this point by examining, in a longitudinal design, the link between features of spontaneous movement at about two days after birth and ASD risk evaluated using the Modified Checklist for Autism in Toddlers by their caregivers at 18 months old. The body movement features were quantified by a recently developed markerless system of infant body motion analysis. Logistic regression analysis revealed that ASD risk at 18 months old is associated with the pattern of spontaneous movement at the neonatal stage. Further, logistic regression based on body movement features during sleep shows better performance in classifying high- and low-risk infants than during the awake state. These findings raise the possibility that early signs of ASD risk may emerge at a developmental stage far earlier than previously thought.

Cough sound-based estimation of vital capacity via cough peak flow using artificial neural network analysis.

This study presents a novel approach for estimating vital capacity using cough sounds and proposes a neural network-based model that utilizes the reference vital capacity computed using the lambda-mu-sigma method, a conventional approach, and the cough peak flow computed based on the cough sound pressure level as inputs. Additionally, a simplified cough sound input model is developed, with the cough sound pressure level used directly as the input instead of the computed cough peak flow. A total of 56 samples of cough sounds and vital capacities were collected from 31 young and 25 elderly participants. Model performance was evaluated using squared errors, and statistical tests including the Friedman and Holm tests were conducted to compare the squared errors of the different models. The proposed model achieved a significantly smaller squared error (0.052 L2, p < 0.001) than the other models. Subsequently, the proposed model and the cough sound-based estimation model were used to detect whether a participant's vital capacity was lower than the typical lower limit. The proposed model demonstrated a significantly higher area under the receiver operating characteristic curve (0.831, p < 0.001) than the other models. These results highlight the effectiveness of the proposed model for screening decreased vital capacity.

Prediction of autistic tendencies at 18 months of age via markerless video analysis of spontaneous body movements in 4-month-old infants.

Early intervention is now considered the core treatment strategy for autism spectrum disorders (ASD). Thus, it is of significant clinical importance to establish a screening tool for the early detection of ASD in infants. To achieve this goal, in a longitudinal design, we analyzed spontaneous bodily movements of 4-month-old infants from general population and assessed their ASD-like behaviors at 18 months of age. A total of 26 movement features were calculated from video-recorded bodily movements of infants at 4 months of age. Their risk of ASD was assessed at 18 months of age with the Modified Checklist for Autism in Toddlerhood, a widely used screening questionnaire. Infants at high risk for ASD at 18 months of age exhibited less rhythmic and weaker bodily movement patterns at 4 months of age than low-risk infants. When the observed bodily movement patterns were submitted to a machine learning-based analysis, linear and non-linear classifiers successfully predicted ASD-like behavior at 18 months of age based on the bodily movement patterns at 4 months of age, at the level acceptable for practical use. This study analyzed the relationship between spontaneous bodily movements at 4 months of age and the ASD risk at 18 months of age. Experimental results suggested the utility of the proposed method for the early screening of infants at risk for ASD. We revealed that the signs of ASD risk could be detected as early as 4 months after birth, by focusing on the infant's spontaneous bodily movements.

Neural Network-based Estimation of Microbubbles Generated in Cardiopulmonary Bypass Circuit: A Clinical Application Study.

The cardiopulmonary bypass system used in cardiac surgery can generate microbubbles (MBs) that may cause complications, such as neurocognitive dysfunction, when delivered into the blood vessel. Estimating the number of MBs generated, thus, is necessary to enable the surgeons to deal with it. To this end, we previously proposed a neural network-based model for estimating the number of MBs from four factors measurable from the cardiopulmonary bypass system: suction flow rate, venous reservoir level, blood viscosity, and perfusion flow rate. However, the model has not been adapted to the data collected from actual surgery cases. In this study, the accuracy of MBs estimated by the proposed model was examined in four clinical cases. The results showed that the coefficient of determination between estimated MBs and the measured MBs throughout the surgeries was R2=0.558 (p<0.001). We found that the surgical treatments, such as administration of drugs, fluids and blood transfusions, increased the number of measured MBs. The coefficient of determination increased to R2= 0.8762 (p<0.001) by excluding the duration of these treatments. This result indicates that the model can estimate the number of MBs with high accuracy under the clinical environment.

Beat-to-Beat Estimation of Peripheral Arterial Stiffness From Local PWV for Quantitative Evaluation of Sympathetic Nervous System Activity.

OBJECTIVE: Sympathetic nervous system activity (SNSA) can rapidly modulate arterial stiffness, thus making it an important biomarker for SNSA evaluation. Pulse wave velocity (PWV) is a well-known quantitative indicator of arterial stiffness, but its functional responsivity to SNSA has not been elucidated. This paper reports a method to estimate rapid changes in peripheral arterial stiffness induced by SNSA using local PWV (LPWV) and to further quantify SNSA based on the estimated stiffness. METHODS: LPWV was measured from the artery near the wrist to the artery near the forefinger using a biodegradable piezoelectric sensor and a photoplethysmography sensor in an electrocutaneous stimulus experiment in which pain evokes the SNSA. The relationship between LPWV, simultaneously measured peripheral arterial stiffness index, and self-reported pain intensity was quantified. RESULTS: The stiffness estimated by LPWV alone and the stiffness estimated by LPWV and arterial pressure both approximate the peripheral arterial stiffness index (R2 = 0.9775 and 0.9719). Pain intensity can be quantitatively evaluated in a sigmoidal relationship by either the estimated stiffness based on LPWV alone (r = 0.8594) or the estimated stiffness based on LPWV and arterial pressure (r = 0.9738). CONCLUSION: Our results demonstrated the validity of LPWV in the quantitative evaluation of SNSA and the optionality of blood pressure correction depending on application scenarios. SIGNIFICANCE: This study advances the understanding of sympathetic innervation of peripheral arteries through the sympathetic responsivity of LPWV and contributes a quantitative biomarker for SNSA evaluation.

Increased cerebrovascular reactivity in selected brain regions after extracranial-intracranial bypass improves the speed and accuracy of visual cancellation in patients with severe steno-occlusive disease: a preliminary study.

The effect of the change in cerebrovascular reactivity (CVR) in each brain area on cognitive function after extracranial-intracranial bypass (EC-IC bypass) was examined. Eighteen patients who underwent EC-IC bypass for severe unilateral steno-occlusive disease were included. Single-photon emission CT (SPECT) for evaluating CVR and the visual cancellation (VC) task were performed before and after surgery. The accuracy of VC was expressed by the arithmetic mean of the age-matched correct answer rate and the accurate answer rate, and the averages of the time (time score) and accuracy (accuracy score) of the four VC subtests were used. The speed of VC tended to be slower, whereas accuracy was maintained before surgery. The EC-IC bypass improved CVR mainly in the cerebral hemisphere on the surgical side. On bivariate analysis, when CVR increased post-operatively, accuracy improved on both surgical sides, but the time score was faster on the left and slower on the right surgical side. Stepwise multiple regression analysis showed that the number of the brain regions associated with the time score was 5 and that associated with the accuracy score was 4. In the hemodynamically ischemic brain, processing speed might be adjusted so that accuracy would be maintained based on the speed-accuracy trade-off mechanism that may become engaged separately in the left and right cerebral hemispheres when performing VC. When considering the treatment for hemodynamic ischemia, the relationship between CVR change and the speed-accuracy trade-off in each brain region should be considered.

Predictors of Stroke Outcome Extracted from Multivariate Linear Discriminant Analysis or Neural Network Analysis.

AIM: The prediction of functional outcome is essential in the management of acute ischemic stroke patients. We aimed to explore the various prognostic factors with multivariate linear discriminant analysis or neural network analysis and evaluate the associations between candidate factors, baseline characteristics, and outcome. METHODS: Acute ischemic stroke patients (n=1,916) with premorbid modified Rankin Scale (mRS) scores of 0-2 were analyzed. The prediction models with multivariate linear discriminant analysis (quantification theory type II) and neural network analysis (log-linearized Gaussian mixture network) were used to predict poor functional outcome (mRS 3-6 at 3 months) with various prognostic factors added to age, sex, and initial neurological severity at admission. RESULTS: Both models revealed that several nutritional statuses and serum alkaline phosphatase (ALP) levels at admission improved the predictive ability. Of the 1,484 patients without missing data, 560 patients (37.7%) had poor outcomes. The patients with poor outcomes had higher ALP levels than those without (294.3±259.5 vs. 246.3±92.5 U/l, P＜0.001). Multivariable logistic analyses revealed that higher ALP levels (1-SD increase) were independently associated with poor stroke outcomes after adjusting for several confounding factors, including the neurological severity, malnutrition status, and inflammation (odds ratio 1.21, 95% confidence interval 1.02-1.49). Several nutritional indicators extracted from prediction models were also associated with poor outcome. CONCLUSION: Both the multivariate linear discriminant and neural network analyses identified the same indicators, such as nutritional status and serum ALP levels. These indicators were independently associated with functional stroke outcome.

A Time-Series Scale Mixture Model of EEG with a Hidden Markov Structure for Epileptic Seizure Detection.

In this paper, we propose a time-series stochastic model based on a scale mixture distribution with Markov transitions to detect epileptic seizures in electroencephalography (EEG). In the proposed model, an EEG signal at each time point is assumed to be a random variable following a Gaussian distribution. The covariance matrix of the Gaussian distribution is weighted with a latent scale parameter, which is also a random variable, resulting in the stochastic fluctuations of covariances. By introducing a latent state variable with a Markov chain in the background of this stochastic relationship, time-series changes in the distribution of latent scale parameters can be represented according to the state of epileptic seizures. In an experiment, we evaluated the performance of the proposed model for seizure detection using EEGs with multiple frequency bands decomposed from a clinical dataset. The results demonstrated that the proposed model can detect seizures with high sensitivity and outperformed several baselines.

Pen-point Trajectory Analysis During Trail Making Test Based on a Time Base Generator Model.

The Trail Making test (TMT) is a widely used neuropsychological test to assess the cognitive function of patients. This paper presents the analysis method of pen-point trajectory during the TMT based on a time base generator (TBG). In the proposed method, the movement segments between targets are first extracted from pen-point trajectories, which are measured during performance of the TMT on an iPad. By fitting the extracted trajectories with a TBG-based trajectory generation model, the proposed method can then calculate quantitative indices representing the shape and collapse of the velocity profile. In the experiment, we analyzed TMT data from 25 stroke patients who were classified into three groups according to their scores on the Mini-Mental State Examination (MMSE). The results revealed that most of the measured inter-target trajectories had unimodal bell-shaped velocity profiles, as seen in reaching movements. Furthermore, we found that the degree of collapse in the velocity profile shape increased significantly when the cognitive function decreased.

A Wearable Finger-Tapping Motion Recognition System Using Biodegradable Piezoelectric Film Sensors.

In this paper, we aimed to develop a method for the automatic recognition of individual finger-tapping motion. Biodegradable piezoelectric film sensors were attached to the skin of a forearm near the wrist (16 channels) to measure small movements of the tendons during five-finger tapping. In the proposed method, the segments in which motion occurred were detected by calculating the total activity for all channels. A neural network is trained to classify tapping motion using the extracted data based on the total activity, thereby allowing the accurate classification of flexion/extension of each finger. We collected experimental data from five healthy young adults to verify the motion recognition accuracy of the proposed method. The results revealed that the proposed method can recognize five-finger tapping motions with high accuracy (flexion/extension of each finger: 92.0%; time-series tapping motion: 88.4%).

Pressure-based Detection of Heart and Respiratory Rates from Human Body Surface using a Biodegradable Piezoelectric Sensor.

This study investigates the relationship between respiration and autonomic nervous system (ANS) activity and proposes a parallel detection method that can simultaneously extract the heart rate (HR) and respiration rate (RR) from different pulse waves measured using a novel biodegradable piezoelectric sensor. The synchronous changes in heart rate variability and respiration reveal the interaction between respiration and the cardiovascular system and their interconnection with ANS activity. Following this principle, respiration was extracted from the HR calculated beat-by-beat from pulse waves. Pulse waves were measured using multiple biodegradable piezoelectric sensors each attached to the human body surface. The Valsalva maneuver experiment was conducted on seven healthy young adults, and the extracted respiratory wave was compared with a reference respiratory wave measured simultaneously. The experimental results are consistent with the observations from reference waves, where R2 = 0.9506, p < 0.001 for the extracted RR and the reference RR, thus demonstrating the detection capability under different respiratory statuses.

Cardiorespiratory synchronisation and systolic blood pressure correlation of peripheral arterial stiffness during endoscopic thoracic sympathectomy.

Muscle sympathetic nerve activity (MSNA) is known as an effective measure to evaluate peripheral sympathetic activity; however, it requires invasive measurement with the microneurography method. In contrast, peripheral arterial stiffness affected by MSNA is a measure that allows non-invasive evaluation of mechanical changes of arterial elasticity. This paper aims to clarify the features of peripheral arterial stiffness to determine whether it inherits MSNA features towards non-invasive evaluation of its activity. To this end, we propose a method to estimate peripheral arterial stiffness [Formula: see text] at a high sampling rate. Power spectral analysis of the estimated [Formula: see text] was then performed on data acquired from 15 patients ([Formula: see text] years) who underwent endoscopic thoracic sympathectomy. We examined whether [Formula: see text] exhibited the features of MSNA where its frequency components synchronise with heart and respiration rates and correlates with the low-frequency component of systolic blood pressure. Regression analysis revealed that the local peak frequency in the range of heartbeat frequency highly correlate with the heart rate ([Formula: see text], [Formula: see text]) where the regression slope was approximately 1 and intercept was approximately 0. Frequency analysis then found spectral peaks of [Formula: see text] approximately 0.2 Hz that correspond to the respiratory cycle. Finally, cross power spectral analysis showed a significant magnitude squared coherence between [Formula: see text] and systolic blood pressure in the frequency band from 0.04 to 0.2 Hz. These results indicate that [Formula: see text] inherits the features observed in MSNA that require invasive measurements, and thus [Formula: see text] can be an effective non-invasive substitution for MSNA measure.

Neural network-based modeling of the number of microbubbles generated with four circulation factors in cardiopulmonary bypass.

The need for the estimation of the number of microbubbles (MBs) in cardiopulmonary bypass surgery has been recognized among surgeons to avoid postoperative neurological complications. MBs that exceed the diameter of human capillaries may cause endothelial disruption as well as microvascular obstructions that block posterior capillary blood flow. In this paper, we analyzed the relationship between the number of microbubbles generated and four circulation factors, i.e., intraoperative suction flow rate, venous reservoir level, continuous blood viscosity and perfusion flow rate in cardiopulmonary bypass, and proposed a neural-networked model to estimate the number of microbubbles with the factors. Model parameters were determined in a machine-learning manner using experimental data with bovine blood as the perfusate. The estimation accuracy of the model, assessed by tenfold cross-validation, demonstrated that the number of MBs can be estimated with a determinant coefficient R2 = 0.9328 (p < 0.001). A significant increase in the residual error was found when each of four factors was excluded from the contributory variables. The study demonstrated the importance of four circulation factors in the prediction of the number of MBs and its capacity to eliminate potential postsurgical complication risks.

Measurement of emotional states of zebrafish through integrated analysis of motion and respiration using bioelectric signals.

Fear, anxiety, and preference in fish are generally evaluated by video-based behavioural analyses. We previously proposed a system that can measure bioelectrical signals, called ventilatory signals, using a 126-electrode array placed at the bottom of an aquarium and achieved cameraless real-time analysis of motion and ventilation. In this paper, we propose a method to evaluate the emotional state of fish by combining the motion and ventilatory indices obtained with the proposed system. In the experiments, fear/anxiety and appetitive behaviour were induced using alarm pheromone and ethanol, respectively. We also found that the emotional state of the zebrafish can be expressed on the principal component (PC) space extracted from the defined indices. The three emotional states were discriminated using a model-based machine learning method by feeding the PCs. Based on discrimination performed every 5 s, the F-score between the three emotional states were as follows: 0.84 for the normal state, 0.76 for the fear/anxiety state, and 0.59 for the appetitive behaviour. These results indicate the effectiveness of combining physiological and motional indices to discriminate the emotional states of zebrafish.

Non-Gaussianity Detection of EEG Signals Based on a Multivariate Scale Mixture Model for Diagnosis of Epileptic Seizures.

OBJECTIVE: The detection of epileptic seizures from scalp electroencephalogram (EEG) signals can facilitate early diagnosis and treatment. Previous studies suggested that the Gaussianity of EEG distributions changes depending on the presence or absence of seizures; however, no general EEG signal models can explain such changes in distributions within a unified scheme. METHODS: This article describes the formulation of a stochastic EEG model based on a multivariate scale mixture distribution that can represent changes in non-Gaussianity caused by stochastic fluctuations in EEG. In addition, we propose an EEG analysis method by combining the model with a filter bank and introduce a feature representing the non-Gaussianity latent in each EEG frequency band. RESULTS: We applied the proposed method to multichannel EEG data from twenty patients with focal epilepsy. The results showed a significant increase in the proposed feature during epileptic seizures, particularly in the high-frequency band. The feature calculated in the high-frequency band allowed highly accurate classification of seizure and non-seizure segments [area under the receiver operating characteristic curve (AUC) = 0.881] using only a simple threshold. CONCLUSION: This article proposed a multivariate scale mixture distribution-based stochastic EEG model capable of representing non-Gaussianity associated with epileptic seizures. Experiments using simulated and real EEG data demonstrated the validity of the model and its applicability to epileptic seizure detection. SIGNIFICANCE: The stochastic fluctuations of EEG quantified by the proposed model can help detect epileptic seizures with high accuracy.

Relationships between motor and cognitive functions and subsequent post-stroke mood disorders revealed by machine learning analysis.

Mood disorders (e.g. depression, apathy, and anxiety) are often observed in stroke patients, exhibiting a negative impact on functional recovery associated with various physical disorders and cognitive dysfunction. Consequently, post-stroke symptoms are complex and difficult to understand. In this study, we aimed to clarify the cross-sectional relationship between mood disorders and motor/cognitive functions in stroke patients. An artificial neural network architecture was devised to predict three types of mood disorders from 36 evaluation indices obtained from functional, physical, and cognitive tests on 274 patients. The relationship between mood disorders and motor/cognitive functions were comprehensively analysed by performing input dimensionality reduction for the neural network. The receiver operating characteristic curve from the prediction exhibited a moderate to high area under the curve above 0.85. Moreover, the input dimensionality reduction retrieved the evaluation indices that are more strongly related to mood disorders. The analysis results suggest a stress threshold hypothesis, in which stroke-induced lesions promote stress vulnerability and may trigger mood disorders.