Development of an effective clustering algorithm for older fallers.

Falls are common and often lead to serious physical and psychological consequences for older persons. The occurrence of falls are usually attributed to the interaction between multiple risk factors. The clinical evaluation of falls risks is time-consuming as a result, hence limiting its availability. The purpose of this study was, therefore, to develop a clustering-based algorithm to determine falls risk. Data from the Malaysian Elders Longitudinal Research (MELoR), comprising 1411 subjects aged ≥55 years, were utilized. The proposed algorithm was developed through the stages of: data pre-processing, feature identification and extraction with either t-Distributed Stochastic Neighbour Embedding (t-SNE) or principal component analysis (PCA)), clustering (K-means clustering, Hierarchical clustering, and Fuzzy C-means clustering) and characteristics interpretation with statistical analysis. A total of 1279 subjects and 9 variables were selected for clustering after the data pre-possessing stage. Using feature extraction with the t-SNE and the K-means clustering algorithm, subjects were clustered into low, intermediate A, intermediate B and high fall risk groups which corresponded with fall occurrence of 13%, 19%, 21% and 31% respectively. Slower gait, poorer balance, weaker muscle strength, presence of cardiovascular disorder, poorer cognitive performance, and advancing age were the key variables identified. The proposed fall risk clustering algorithm grouped the subjects according to features. Such a tool could serve as a case identification or clinical decision support tool for clinical practice to enhance access to falls prevention efforts.

Brain Functional Network and Amino Acid Metabolism Association in Females with Subclinical Depression.

This study aimed to investigate the association between complex brain functional networks and the metabolites in urine in subclinical depression. Electroencephalography (EEG) signals were recorded from 78 female college students, including 40 with subclinical depression (ScD) and 38 healthy controls (HC). The phase delay index was utilized to construct functional connectivity networks and quantify the topological properties of brain networks using graph theory. Meanwhile, the urine of all participants was collected for non-targeted LC-MS metabolic analysis to screen differential metabolites. The global efficiency was significantly increased in the α-2, ß-1, and ß-2 bands, while the characteristic path length of ß-1 and ß-2 and the clustering coefficient of ß-2 were decreased in the ScD group. The severity of depression was negatively correlated with the level of cortisone (p = 0.016, r = -0.40). The metabolic pathways, including phenylalanine metabolism, phenylalanine tyrosine tryptophan biosynthesis, and nitrogen metabolism, were disturbed in the ScD group. The three metabolic pathways were negatively correlated (p = 0.014, r = -0.493) with the global efficiency of the brain network of the ß-2 band, whereas they were positively correlated (p = 0.014, r = 0.493) with the characteristic path length of the ß-2 band. They were mainly associated with low levels of L-phenylalanine, and the highest correlation sparsity was 0.11. The disturbance of phenylalanine metabolism and the phenylalanine, tryptophan, tyrosine biosynthesis pathways cause depressive symptoms and changes in functional brain networks. The decrease in the L-phenylalanine level may be related to the randomization trend of the ß-1 frequency brain functional network.

Temporal and Spatial Dynamics of EEG Features in Female College Students with Subclinical Depression.

Synchronization of the dynamic processes in structural networks connect the brain across a wide range of temporal and spatial scales, creating a dynamic and complex functional network. Microstate and omega complexity are two reference-free electroencephalography (EEG) measures that can represent the temporal and spatial complexities of EEG data. Few studies have focused on potential brain spatiotemporal dynamics in the early stages of depression to use as an early screening feature for depression. Thus, this study aimed to explore large-scale brain network dynamics of individuals both with and without subclinical depression, from the perspective of temporal and spatial dimensions and to input them as features into a machine learning framework for the automatic diagnosis of early-stage depression. To achieve this, spatio-temporal dynamics of rest-state EEG signals in female college students (n = 40) with and without (n = 38) subclinical depression were analyzed using EEG microstate and omega complexity analysis. Then, based on differential features of EEGs between the two groups, a support vector machine was utilized to compare performances of spatio-temporal features and single features in the classification of early depression. Microstate results showed that the occurrence rate of microstate class B was significantly higher in the group with subclinical depression when compared with the group without. Moreover, the duration and contribution of microstate class C in the subclinical group were both significantly lower than in the group without subclinical depression. Omega complexity results showed that the global omega complexity of ß-2 and γ band was significantly lower for the subclinical depression group compared with the other group (p < 0.05). In addition, the anterior and posterior regional omega complexities were lower for the subclinical depression group compared to the comparison group in α-1, ß-2 and γ bands. It was found that AUC of 81% for the differential indicators of EEG microstates and omega complexity was deemed better than a single index for predicting subclinical depression. Thus, since temporal and spatial complexity of EEG signals were manifestly altered in female college students with subclinical depression, it is possible that this characteristic could be adopted as an early auxiliary diagnostic indicator of depression.

Positive and Negative Emotion Classification Based on Multi-channel.

The EEG features of different emotions were extracted based on multi-channel and forehead channels in this study. The EEG signals of 26 subjects were collected by the emotional video evoked method. The results show that the energy ratio and differential entropy of the frequency band can be used to classify positive and negative emotions effectively, and the best effect can be achieved by using an SVM classifier. When only the forehead and forehead signals are used, the highest classification accuracy can reach 66%. When the data of all channels are used, the highest accuracy of the model can reach 82%. After channel selection, the best model of this study can be obtained. The accuracy is more than 86%.

Robust PPG motion artifact detection using a 1-D convolution neural network.

BACKGROUND AND OBJECTIVES: Continuous monitoring of physiological parameters such as photoplethysmography (PPG) has attracted increased interest due to advances in wearable sensors. However, PPG recordings are susceptible to various artifacts, and thus reducing the reliability of PPG-driven parameters, such as oxygen saturation, heart rate, blood pressure and respiration. This paper proposes a one-dimensional convolution neural network (1-D-CNN) to classify five-second PPG segments into clean or artifact-affected segments, avoiding data-dependent pulse segmentation techniques and heavy manual feature engineering. METHODS: Continuous raw PPG waveforms were blindly allocated into segments with an equal length (5s) without leveraging any pulse location information and were normalized with Z-score normalization methods. A 1-D-CNN was designed to automatically learn the intrinsic features of the PPG waveform, and perform the required classification. Several training hyperparameters (initial learning rate and gradient threshold) were varied to investigate the effect of these parameters on the performance of the network. Subsequently, this proposed network was trained and validated with 30 subjects, and then tested with eight subjects, with our local dataset. Moreover, two independent datasets downloaded from the PhysioNet MIMIC II database were used to evaluate the robustness of the proposed network. RESULTS: A 13 layer 1-D-CNN model was designed. Within our local study dataset evaluation, the proposed network achieved a testing accuracy of 94.9%. The classification accuracy of two independent datasets also achieved satisfactory accuracy of 93.8% and 86.7% respectively. Our model achieved a comparable performance with most reported works, with the potential to show good generalization as the proposed network was evaluated with multiple cohorts (overall accuracy of 94.5%). CONCLUSION: This paper demonstrated the feasibility and effectiveness of applying blind signal processing and deep learning techniques to PPG motion artifact detection, whereby manual feature thresholding was avoided and yet a high generalization ability was achieved.

Adaptive template matching of photoplethysmogram pulses to detect motion artefact.

OBJECTIVE: The photoplethysmography (PPG) signal, commonly used in the healthcare settings, is easily affected by movement artefact leading to errors in the extracted heart rate and SpO2 estimates. This study aims to develop an online artefact detection system based on adaptive (dynamic) template matching, suitable for continuous PPG monitoring during daily living activities or in the intensive care units (ICUs). APPROACH: Several master templates are initially generated by applying principal component analysis to data obtained from the PhysioNet MIMIC II database. The master template is then updated with each incoming clean PPG pulse. The correlation coefficient is used to classify the PPG pulse into either good or bad quality categories. The performance of our algorithm was evaluated using data obtained from two different sources: (i) our own data collected from 19 healthy subjects using the wearable Sotera Visi Mobile system (Sotera Wireless Inc.) as they performed various movement types; and (ii) ICU data provided by the PhysioNet MIMIC II database. The developed algorithm was evaluated against a manually annotated 'gold standard' (GS). MAIN RESULTS: Our algorithm achieved an overall accuracy of 91.5% ± 2.9%, with a sensitivity of 94.1% ± 2.7% and a specificity of 89.7% ± 5.1%, when tested on our own data. When applying the algorithm to data from the PhysioNet MIMIC II database, it achieved an accuracy of 98.0%, with a sensitivity and specificity of 99.0% and 96.1%, respectively. SIGNIFICANCE: The proposed method is simple and robust against individual variations in the PPG characteristics, thus making it suitable for a diverse range of datasets. Integration of the proposed artefact detection technique into remote monitoring devices could enhance reliability of the PPG-derived physiological parameters.

An Overview on Image Registration Techniques for Cardiac Diagnosis and Treatment.

Image registration has been used for a wide variety of tasks within cardiovascular imaging. This study aims to provide an overview of the existing image registration methods to assist researchers and impart valuable resource for studying the existing methods or developing new methods and evaluation strategies for cardiac image registration. For the cardiac diagnosis and treatment strategy, image registration and fusion can provide complementary information to the physician by using the integrated image from these two modalities. This review also contains a description of various imaging techniques to provide an appreciation of the problems associated with implementing image registration, particularly for cardiac pathology intervention and treatments.

Knee cartilage segmentation and thickness computation from ultrasound images.

Quantitative thickness computation of knee cartilage in ultrasound images requires segmentation of a monotonous hypoechoic band between the soft tissue-cartilage interface and the cartilage-bone interface. Speckle noise and intensity bias captured in the ultrasound images often complicates the segmentation task. This paper presents knee cartilage segmentation using locally statistical level set method (LSLSM) and thickness computation using normal distance. Comparison on several level set methods in the attempt of segmenting the knee cartilage shows that LSLSM yields a more satisfactory result. When LSLSM was applied to 80 datasets, the qualitative segmentation assessment indicates a substantial agreement with Cohen's κ coefficient of 0.73. The quantitative validation metrics of Dice similarity coefficient and Hausdorff distance have average values of 0.91 ± 0.01 and 6.21 ± 0.59 pixels, respectively. These satisfactory segmentation results are making the true thickness between two interfaces of the cartilage possible to be computed based on the segmented images. The measured cartilage thickness ranged from 1.35 to 2.42 mm with an average value of 1.97 ± 0.11 mm, reflecting the robustness of the segmentation algorithm to various cartilage thickness. These results indicate a potential application of the methods described for assessment of cartilage degeneration where changes in the cartilage thickness can be quantified over time by comparing the true thickness at a certain time interval.

Standing beat-to-beat blood pressure variability is reduced among fallers in the Malaysian Elders Longitudinal Study.

The aim of this study was to determine the relationship between falls and beat-to-beat blood pressure (BP) variability.Continuous noninvasive BP measurement is as accurate as invasive techniques. We evaluated beat-to-beat supine and standing BP variability (BPV) using time and frequency domain analysis from noninvasive continuous BP recordings.A total of 1218 older adults were selected. Continuous BP recordings obtained were analyzed to determine standard deviation (SD) and root mean square of real variability (RMSRV) for time domain BPV and fast-Fourier transform low frequency (LF), high frequency (HF), total power spectral density (PSD), and LF:HF ratio for frequency domain BPV.Comparisons were performed between 256 (21%) individuals with at least 1 fall in the past 12 months and nonfallers. Fallers were significantly older (P = .007), more likely to be female (P = .006), and required a longer time to complete the Timed-Up and Go test (TUG) and frailty walk test (P ≤ .001). Standing systolic BPV (SBPV) was significantly lower in fallers compared to nonfallers (SBPV-SD, P = .016; SBPV-RMSRV, P = .033; SBPV-LF, P = .003; SBPV-total PSD, P = .012). Nonfallers had significantly higher supine to standing ratio (SSR) for SBPV-SD, SBPV-RMSRV, and SBPV-total PSD (P = .017, P = .013, and P = .009). In multivariate analyses, standing BPV remained significantly lower in fallers compared to nonfallers after adjustment for age, sex, diabetes, frailty walk, and supine systolic BP. The reduction in frequency-domain SSR among fallers was attenuated by supine systolic BP, TUG, and frailty walk.In conclusion, reduced beat-to-beat BPV while standing is independently associated with increased risk of falls. Changes between supine and standing BPV are confounded by supine BP and walking speed.

Multimodality registration of two-dimensional echocardiography and cardiac CT for mitral valve diagnosis and surgical planning.

A registration method to fuse two-dimensional (2-D) echocardiography images with cardiac computed tomography (CT) volume is presented. The method consists of two major procedures: temporal and spatial registrations. In temporal registration, the echocardiography frames at similar cardiac phases as the CT volume were interpolated based on electrocardiogram signal information, and the noise of the echocardiography image was reduced using the speckle reducing anisotropic diffusion technique. For spatial registration, an intensity-based normalized mutual information method was applied with a pattern search optimization algorithm to produce an interpolated cardiac CT image. The proposed registration framework does not require optical tracking information. Dice coefficient and Hausdorff distance for the left atrium assessments were [Formula: see text] and [Formula: see text], respectively; for left ventricle, they were [Formula: see text] and [Formula: see text], respectively. There was no significant difference in the mitral valve annulus diameter measurement between the manually and automatically registered CT images. The transformation parameters showed small deviations ([Formula: see text] deviation in translation and [Formula: see text] for rotation) between manual and automatic registrations. The proposed method aids the physician in diagnosing mitral valve disease as well as provides surgical guidance during the treatment procedure.

2D to 3D fusion of echocardiography and cardiac CT for TAVR and TAVI image guidance.

This study proposed a registration framework to fuse 2D echocardiography images of the aortic valve with preoperative cardiac CT volume. The registration facilitates the fusion of CT and echocardiography to aid the diagnosis of aortic valve diseases and provide surgical guidance during transcatheter aortic valve replacement and implantation. The image registration framework consists of two major steps: temporal synchronization and spatial registration. Temporal synchronization allows time stamping of echocardiography time series data to identify frames that are at similar cardiac phase as the CT volume. Spatial registration is an intensity-based normalized mutual information method applied with pattern search optimization algorithm to produce an interpolated cardiac CT image that matches the echocardiography image. Our proposed registration method has been applied on the short-axis "Mercedes Benz" sign view of the aortic valve and long-axis parasternal view of echocardiography images from ten patients. The accuracy of our fully automated registration method was 0.81 ± 0.08 and 1.30 ± 0.13 mm in terms of Dice coefficient and Hausdorff distance for short-axis aortic valve view registration, whereas for long-axis parasternal view registration it was 0.79 ± 0.02 and 1.19 ± 0.11 mm, respectively. This accuracy is comparable to gold standard manual registration by expert. There was no significant difference in aortic annulus diameter measurement between the automatically and manually registered CT images. Without the use of optical tracking, we have shown the applicability of this technique for effective fusion of echocardiography with preoperative CT volume to potentially facilitate catheter-based surgery.

Automated cervical precancerous cells screening system based on Fourier transform infrared spectroscopy features.

Fourier transform infrared (FTIR) spectroscopy technique can detect the abnormality of a cervical cell that occurs before the morphological change could be observed under the light microscope as employed in conventional techniques. This paper presents developed features extraction for an automated screening system for cervical precancerous cell based on the FTIR spectroscopy as a second opinion to pathologists. The automated system generally consists of the developed features extraction and classification stages. Signal processing techniques are used in the features extraction stage. Then, discriminant analysis and principal component analysis are employed to select dominant features for the classification process. The datasets of the cervical precancerous cells obtained from the feature selection process are classified using a hybrid multilayered perceptron network. The proposed system achieved 92% accuracy.

Evaluation of Two New Indices of Blood Pressure Variability Using Postural Change in Older Fallers.

To evaluate the utility of blood pressure variability (BPV) calculated using previously published and newly introduced indices using the variables falls and age as comparators.While postural hypotension has long been considered a risk factor for falls, there is currently no documented evidence on the relationship between BPV and falls.A case-controlled study involving 25 fallers and 25 nonfallers was conducted. Systolic (SBPV) and diastolic blood pressure variability (DBPV) were assessed using 5 indices: standard deviation (SD), standard deviation of most stable continuous 120 beats (staSD), average real variability (ARV), root mean square of real variability (RMSRV), and standard deviation of real variability (SDRV). Continuous beat-to-beat blood pressure was recorded during 10 minutes' supine rest and 3 minutes' standing.Standing SBPV was significantly higher than supine SBPV using 4 indices in both groups. The standing-to-supine-BPV ratio (SSR) was then computed for each subject (staSD, ARV, RMSRV, and SDRV). Standing-to-supine ratio for SBPV was significantly higher among fallers compared to nonfallers using RMSRV and SDRV (P = 0.034 and P = 0.025). Using linear discriminant analysis (LDA), 3 indices (ARV, RMSRV, and SDRV) of SSR SBPV provided accuracies of 61.6%, 61.2%, and 60.0% for the prediction of falls which is comparable with timed-up and go (TUG), 64.4%.This study suggests that SSR SBPV using RMSRV and SDRV is a potential predictor for falls among older patients, and deserves further evaluation in larger prospective studies.

Classification of Implantable Rotary Blood Pump States With Class Noise.

A medical case study related to implantable rotary blood pumps is examined. Five classifiers and two ensemble classifiers are applied to process the signals collected from the pumps for the identification of the aortic valve nonopening pump state. In addition to the noise-free datasets, up to 40% class noise has been added to the signals to evaluate the classification performance when mislabeling is present in the classifier training set. In order to ensure a reliable diagnostic model for the identification of the pump states, classifications performed with and without class noise are evaluated. The multilayer perceptron emerged as the best performing classifier for pump state detection due to its high accuracy as well as robustness against class noise.

Improved Measurement of Blood Pressure by Extraction of Characteristic Features from the Cuff Oscillometric Waveform.

We present a novel approach to improve the estimation of systolic (SBP) and diastolic blood pressure (DBP) from oscillometric waveform data using variable characteristic ratios between SBP and DBP with mean arterial pressure (MAP). This was verified in 25 healthy subjects, aged 28 ± 5 years. The multiple linear regression (MLR) and support vector regression (SVR) models were used to examine the relationship between the SBP and the DBP ratio with ten features extracted from the oscillometric waveform envelope (OWE). An automatic algorithm based on relative changes in the cuff pressure and neighbouring oscillometric pulses was proposed to remove outlier points caused by movement artifacts. Substantial reduction in the mean and standard deviation of the blood pressure estimation errors were obtained upon artifact removal. Using the sequential forward floating selection (SFFS) approach, we were able to achieve a significant reduction in the mean and standard deviation of differences between the estimated SBP values and the reference scoring (MLR: mean ± SD = -0.3 ± 5.8 mmHg; SVR and -0.6 ± 5.4 mmHg) with only two features, i.e., Ratio2 and Area3, as compared to the conventional maximum amplitude algorithm (MAA) method (mean ± SD = -1.6 ± 8.6 mmHg). Comparing the performance of both MLR and SVR models, our results showed that the MLR model was able to achieve comparable performance to that of the SVR model despite its simplicity.

Multiple LREK active contours for knee meniscus ultrasound image segmentation.

Quantification of knee meniscus degeneration and displacement in an ultrasound image requires simultaneous segmentation of femoral condyle, meniscus, and tibial plateau in order to determine the area and the position of the meniscus. In this paper, we present an active contour for image segmentation that uses scalable local regional information on expandable kernel (LREK). It includes using a strategy to adapt the size of a local window in order to avoid being confined locally in a homogeneous region during the segmentation process. We also provide a multiple active contours framework called multiple LREK (MLREK) to deal with multiple object segmentation without merging and overlapping between the neighboring contours in the shared boundaries of separate regions. We compare its performance to other existing active contour models and show an improvement offered by our model. We then investigate the choice of various parameters in the proposed framework in response to the segmentation outcome. Dice coefficient and Hausdorff distance measures over a set of real knee meniscus ultrasound images indicate a potential application of MLREK for assessment of knee meniscus degeneration and displacement.

Regional assessment of LV wall in infarcted heart using tagged MRI and cardiac modelling.

A segmental two-parameter empirical deformable model is proposed for evaluating regional motion abnormality of the left ventricle. Short-axis tagged MRI scans were acquired from 10 healthy subjects and 10 postinfarct patients. Two motion parameters, contraction and rotation, were quantified for each cardiac segment by fitting the proposed model using a non-rigid registration algorithm. The accuracy in motion estimation was compared to a global model approach. Motion parameters extracted from patients were correlated to infarct transmurality assessed with delayed-contrast-enhanced MRI. The proposed segmental model allows markedly improved accuracy in regional motion analysis as compared to the global model for both subject groups (1.22-1.40 mm versus 2.31-2.55 mm error). By end-systole, all healthy segments experienced radial displacement by ~25-35% of the epicardial radius, whereas the 3 short-axis planes rotated differently (basal: 3.3°; mid: -1° and apical: -4.6°) to create a twisting motion. While systolic contraction showed clear correspondence to infarct transmurality, rotation was nonspecific to either infarct location or transmurality but could indicate the presence of functional abnormality. Regional contraction and rotation derived using this model could potentially aid in the assessment of severity of regional dysfunction of infarcted myocardium.

Intelligent screening systems for cervical cancer.

Advent of medical image digitalization leads to image processing and computer-aided diagnosis systems in numerous clinical applications. These technologies could be used to automatically diagnose patient or serve as second opinion to pathologists. This paper briefly reviews cervical screening techniques, advantages, and disadvantages. The digital data of the screening techniques are used as data for the computer screening system as replaced in the expert analysis. Four stages of the computer system are enhancement, features extraction, feature selection, and classification reviewed in detail. The computer system based on cytology data and electromagnetic spectra data achieved better accuracy than other data.

Robust aortic valve non-opening detection for different cardiac conditions.

In recent years, extensive studies have been conducted in the area of pumping state detection for implantable rotary blood pumps. However, limited studies have focused on automatically identifying the aortic valve non-opening (ANO) state despite its importance in the development of control algorithms aiming for myocardial recovery. In the present study, we investigated the performance of 14 ANO indices derived from the pump speed waveform using four different types of classifiers, including linear discriminant analysis, logistic regression, back propagation neural network, and k-nearest neighbors (KNN). Experimental measurements from four greyhounds, which take into consideration the variations in cardiac contractility, systemic vascular resistance, and total blood volume were used. By having only two indices, (i) the root mean square value, and (ii) the standard deviation, we were able to achieve an accuracy of 92.8% with the KNN classifier. Further increase of the number of indices to five for the KNN classifier increases the overall accuracy to 94.6%.

Investigation of CPD and HMDS sample preparation techniques for cervical cells in developing computer-aided screening system based on FE-SEM/EDX.

This paper investigated the effects of critical-point drying (CPD) and hexamethyldisilazane (HMDS) sample preparation techniques for cervical cells on field emission scanning electron microscopy and energy dispersive X-ray (FE-SEM/EDX). We investigated the visualization of cervical cell image and elemental distribution on the cervical cell for two techniques of sample preparation. Using FE-SEM/EDX, the cervical cell images are captured and the cell element compositions are extracted for both sample preparation techniques. Cervical cell image quality, elemental composition, and processing time are considered for comparison of performances. Qualitatively, FE-SEM image based on HMDS preparation technique has better image quality than CPD technique in terms of degree of spread cell on the specimen and morphologic signs of cell deteriorations (i.e., existence of plate and pellet drying artifacts and membrane blebs). Quantitatively, with mapping and line scanning EDX analysis, carbon and oxygen element compositions in HMDS technique were higher than the CPD technique in terms of weight percentages. The HMDS technique has shorter processing time than the CPD technique. The results indicate that FE-SEM imaging, elemental composition, and processing time for sample preparation with the HMDS technique were better than CPD technique for cervical cell preparation technique for developing computer-aided screening system.