Search

Different biological labs conduct similar experiments on same diseases. It is highly desirable to have a better model based on more experimental results than that on a single result. In this paper, we propose a method for integrating microarray data from multiple sources for building classification models. To test the method, we use three real world microarray data sets from different labs with different experimental devices and environments. Although microarray data is well known for its inconsistencies across labs, we demonstrate that it is possible to build consistent models using data sets from multiple labs. We report our method, experimental results and observations in the paper. Refereed/Peer-reviewed

We describe a contact-less method for measurement of respiration rate during sleep using a 5.8GHz radio-frequency bio-motion sensor. The sensor operates by sensing phase shifts in reflected radio waves from the torso caused by respiratory movements and other bodily movements such as twitches, positional changes etc. These non-respiratory motion artefacts can obscure reliable estimation of breathinig rates if conventional spectral analysis is used. This paper reports on the accuracy of the respiration rate estimates obtained via algorithmic approaches using Lomb-periodogram based analysis (which can deal with missing or corrupted data), as compared to conventional spectral analysis. Gold-standard respiration rates are derived by expert scoring of respiration rates measured through polysomnography (PSG) from sensors (Respiratory Inductance Plethysmography (RIP) belts) in contact with the subject in an accredited sleep laboratory. Specifically, respiration rates for 15-minute segments chosen from 10 subjects free of Sleep-Disorded Breathing (AHI〈5) were selected for analysis in this paper. Comparison to the expert annotation indicates strong agreement, with the Lomb-periodogram respiration rates with the average error between the measurements being less than 0.4 breaths/min and a standard deviation of 0.3 breaths/minute. Moreover, we showed that the proposed algorithm could track respiration rate over the complete night's recordings for those 10 subjects. We conclude that the non-contact biomotion sensor may provide a promising approach to continuous respiration rate monitoring of reasonable accuracy.

Critical Arrhythmic ECG such as Ventricular Tachycardia (VT) and Ventricular Fibrillation (VF) are both distinguishable by its waveform characteristics. A VF waveform is often described as disorganized and has an irregular rhythm while a VT waveform exhibits abnormal signatures and presents a regular rhythm pattern. This paper presents a fast cross-correlation algorithm using multiple waveform templates for automatic detection of life threatening arrhythmias such as VT and VF from the Normal Sinus Rhythm (NSR) waveforms. A sliding-window template cross-correlation technique is applied to an ECG signal to generate an array of correlation coefficients. Then a correlation coefficient curve is used to detect high coefficient values for a type of template that will quantify the similarity between an examined ECG signal and a template. The method presented in this paper is able to detect all three different types of ECG signals from a total 21 testing signal set with a satisfied correct rate.

The aim of this paper is to introduce a novel non-Gaussian statistical model-based approach for medical image fusion based on the Meridian distribution. The paper also includes a new approach to estimate the parameters of generalized Cauchy distribution. The input images are first decomposed using the Dual-Tree Complex Wavelet Transform (DT-CWT) with the subband coefficients modelled as Meridian random variables. Then, the convolution of Meridian distributions is applied as a probabilistic prior to model the fused coefficients, and the weights used to combine the source images are optimised via Maximum Likelihood (ML) estimation. The superior performance of the proposed method is demonstrated using medical images.

In this paper we present a new approach for automated cell detection in single frames of 2D microscopic phase contrast images of cancer cells which is based on learning cellular texture features. The main challenge addressed in this paper is to deal with clusters of cells where each cell has a rather complex appearance composed of sub-regions with different texture features. Our approach works on two different levels of abstraction. First, we apply statistical learning to learn 6 different types of different local cellular texture features, classify each pixel according to them and we obtain an image partition composed of 6 different pixel categories. Based on this partitioned image we decide in a second step if pre-selected seeds belong to the same cell or not. Experimental results show the high accuracy of the proposed method and especially average precision above 95%.

Independent Component Analysis (ICA) is a very common instantiation of the Blind Source Separation (BSS) problem. In the context of biomedical signal analysis, ICA is generally applied to multi-channel recordings of physiological phenomena in order to de-noise and extract meaningful information underlying the recordings. This paper assesses the Spatio-Temporal ICA (ST-ICA) framework, which uses both spatial and temporal information derived from multi-channel time-series to extract underlying sources. In contrast, the standard implementation of the ICA algorithm generally uses only limited spatial information to inform the separation process. One of the major steps in the implementation of any ICA algorithm is the selection of relevant components from the many ICA usually returns. With ST-ICA there is a rich data-set of components exhibiting spatial as well as temporal/spectral information that could be used to identify the underlying process subspaces extracted by the ST-ICA algorithm. This paper highlights the methodology for performing ST-ICA and assesses the possible ways in which process subspace identification may take place.

Our goal is to enhance the quality of life of patients with hemiplegia by means of an active motion support system that assists the impaired motion such as to make it as close as possible to the motion of an able bodied person. We have developed the Robot Suit HAL (Hybrid Assistive Limb) to actively support and enhance the human motor functions. The purpose of the research presented in this paper is to propose the required control method to support voluntarily motion using a trigger based on patient's bioelectrical signal. Clinical trials were conducted in order to investigate the effectiveness of the proposed control method. The first stage of the trials, described in this paper, involved the participation of one hemiplegic patient who is not able to bend his right knee. As a result, the motion support provided by the HAL moved the paralyzed knee joint according to his intention and improved the range of the subject's knee flexion. The first evaluation of the control method with one subject showed promising results for future trials to explore the effectiveness for a wide range of types of hemiplegia.

This paper examines the feasibility of accurate state classification of autonomic nervous activity (ANA) based on the power spectral pattern of the heart rate fluctuations (HRFs). Some attempts have been made to utilize artificial neural networks (ANNs) to classify HRFs for clinical diagnoses such as ischemic cardiomyopathy, arrhythmia or sleep apnea. To establish the firm bases for making such clinical diagnoses, it may be important to examine the classification accuracy for the data in physiologically well defined conditions by e.g. application of autonomic blocking agents. In this paper the three layered perceptron has been trained by the heart rate data in variety of ANS states yielded by the application of Atropine and Propranolol to 14 healthy male subjects. Six state (control, atropine and propranolol for each of the spine and upright posture) classification based on power spectrum showed average sensitivity of 67.2% and specificity 91.2%. Four state (control, atropine, propranolol and double block for either spine or upright posture) resulted in the average classification sensitivity of 75.7% and specificity 95.5%. The paper revealed that entropy bandwidth and indices originated from characteristic oscillations of blood pressure change improve the classification accuracy.

This paper presents a new application of the Particle Swarm Optimization (PSO) algorithm to optimize Mel Frequency Cepstrum Coefficients (MFCC) parameters, in order to extract an optimal feature set for diagnosis of hypothyroidism in infants using Multi-Layer Perceptrons (MLP) neural network. MFCC features is influenced by the number of filter banks (f(b)) and the number of coefficients (n(c)) used. These parameters are critical in representation of the features as they affect the resolution and dimensionality of the features. In this paper, the PSO algorithm was used to optimize the values of f(b) and n(c). The MFCC features based on the PSO optimization were extracted from healthy and unhealthy infant cry signals and used to train MLP in the classification of hypothyroid infant cries. The results indicate that the PSO algorithm could determine the optimum combination of f(b) and n(c) that produce the best classification accuracy of the MLP.

It has been demonstrated that negative emotions have adverse effects on the immune system of a person. This contributes to increased morbidity and mortality in the elderly population and has a direct impact on quality of life. Positive emotions on the other hand may not only undo the harmful effects of negative emotions but also protect against certain diseases. Hence the use of technology to facilitate emotional regulation that reduces negative emotions may be a good way to promote self-care and support well-being. In this paper we present the early design stages of an emotion detection system that aims to support remote support and self-regulation in situations of intense emotional distress. We provide evidence of the suitability of non-parametric cumulative sum (CUSUM) to indentify emotional changes from neutral to non-neutral and vice versa in real time.

This paper presents a low-complexity video encoding method applicable for wireless image transmission in capsule endoscopes. This encoding method is based on Wyner-Ziv theory, in which side information available at a transmitter is treated as side information at its receiver. Therefore complex processes in video encoding, such as estimation of the motion vector, are moved to the receiver side, which has a larger-capacity battery. As a result, the encoding process is only to decimate coded original data through channel coding. We provide a performance evaluation for a low-density parity check (LDPC) coding method in the AWGN channel.

In diagnosing lung diseases, the structure and shape of airways in lungs are of great importance. In this paper we propose a novel method for segmenting low-contrast 3-D CTA images of airways in lungs. Our approach is an edge-detecting slice-by-slice segmentation method, capable of segmenting low contrasted airway regions. Our segmentation using projections method shows robustness in images with high presence of noise.

The responsible pathological mechanisms of essential tremor are not yet clear. In order to understand the mechanisms of the central network its sources need to be found. The cortical sources of both the basic and first "harmonic" frequency of essential tremor are addressed in this paper. The power and coherence were estimated using the multitaper method for EEG and EMG data from 6 essential tremor patients. The Dynamic Imaging of Coherent Sources (DICS) was used to find the coherent sources in the brain. Before hand this method was validated for the application of finding multiple sources for the same oscillation in the brain by using two model simulations which indicated the accuracy of the method. In all the essential tremor patients the corticomuscular coherence was also present in the basic and the first harmonic frequency of the tremor. The source for the basic frequency and the first harmonic frequency was in the region of primary sensory motor cortex, prefrontal and in the diencephalon on the contralateral side for all the patients. Thus the generation of these two oscillations involves the same cortical areas and indicates the oscillation at double the tremor frequency is a harmonic of the basic tremor frequency.

Autonomous decision making modules in computer vision application allow recognition and classification of different objects, persons, and events in images and video sequences and also make it possible to classify different sensor readings (e.g. images) according to their scientific saliencies. In this paper, we propose a new approach to create the training set for these algorithms by retrieving salient images using electroencephalogram (EEG) and brain computer interface (BCI) and rapid image presentation. To this end, two groups of subjects, namely, expert and novice subjects were asked to participate in our experiments. We show that a relatively high retrieval accuracy can be achieved for most of the subjects. Furthermore, to assess the impact of expertise on the retrieval process, we study their EEG signals separately and show that there is a clear difference in their brainwaves while observing salient images.

In ultrasound elastography, tissue axial strains are obtained through the differentiation of axial displacements. However, the application of the gradient operator amplifies the noise present in the displacement rendering unreadable axial strains. In this paper a novel denoising scheme based on repeated filtering in consecutive fractional Fourier transform domains is proposed for the accurate estimation of axial strains. The presented method generates a time-varying cutoff threshold that can accommodate the discrete non-stationarities present in the displacement signal. This is achieved by means of a filter circuit which is composed of a small number of ordinary linear low-pass filters and appropriate fractional Fourier transforms. We show that the proposed method can improve the contrast-to-noise ratio (CNR(e)) of the elastogram outperforming conventional low-pass filters.

In this paper, a fully automated method for segmenting Left Ventricle (LV) in echocardiography images is proposed. A new method named active ellipse model is developed to automatically find the best ellipse inside the LV chamber without intervention of any specialist. A modified B-Spline Snake algorithm is used to segment the LV chamber in which the initial contour is formed by the predefined ellipse. As a result of using active ellipse model, the segmentation is extricated from dealing with gaps within myocardium boundary which are highly problematic in echocardiography image segmentation. Based on the results obtained from different studies, the proposed method is faster and more accurate than previous approaches. Our method is evaluated on 20 sets of echocardiography images of patients; and acquired results (92.30 ± 4.45% dice's coefficient) indicate the proposed method has remarkable performance.

We present a light-weight, cheap and low-power, wearable system for assisting the visually impaired in performing routine mobility tasks. Our system extends the range of the white cane by providing the user with vibro-tactile cues corresponding to the location of obstacles and a safe path for traversal through a cluttered environment. The presented approach keeps cognitive load to a minimum, and while being autonomous, adapts to the changing mobility requirements of a navigating user. In this paper, we provide an overview of the hardware and algorithmic components of our system, and show results of pilot studies with human test subjects. Our system operates at 20Hz, and significantly improves mobility performance compared to using only the white cane.

Accurate identification of lesion borders is an important task in the analysis of dermoscopy images since the extraction of skin lesion borders provides important cues for accurate diagnosis. Snakes have been used for segmenting a variety of medical imagery including dermoscopy, however, due to the compromise of internal and external energy forces they can lead to under- or over-segmentation problems. In this paper, we introduce a mean shift based gradient vector flow (GVF) snake algorithm that drives the internal/external energies towards the correct direction. The proposed segmentation method incorporates a mean shift operation within the standard GVF cost function. Experimental results on a large set of diverse dermoscopy images demonstrate that the presented method accurately determines skin lesion borders in dermoscopy images.

The initial step of vessel segmentation in 3D is the detection of vessel centerlines. The proposed methods in literature are either dependent on vessel radius and/or have low response at vessel bifurcations. In this paper we propose a 3D tubular structure detection method that removes these two drawbacks. The proposed method exploits the observations on the eigenvalues of the Hessian matrix as is done in literature, yet it employs a direct 3D vector field singularity characterization. The Gradient Vector Flow vector field is used and the eigenvalues of its Jacobian are exploited in computing a parameter free vesselness map. Results on phantom and real patient data exhibit robustness to scale, high response at vessel bifurcations, and good noise/non-vessel structure suppression.

A disposable endoscopic platform with actuation motors inside the body of the endoscope is presented. This platform can enable new medical devices for diagnosis and for minimally invasive surgeries. This paper addresses mechanical and safety issues with existing endoscope technologies by incorporating disposability, safety modules, and lower cable forces. In order to produce path-independent cable forces, motors are incorporated in the body of the endoscope near the bending tip. Results for tip forces are shown accompanied by an analytical model describing the scaling laws for this type of robotic architecture. The system under development will provide a platform for research into haptic control and perceptual feedback.

Cochlear implant patients often complain about their difficulty in understanding speech in noisy environments. Currently a fixed noise suppression algorithm is used in cochlear implants regardless of the characteristics of the speech or noise environment. Access to an intelligent mechanism to determine the noise environment on-the-fly in order to automatically switch between different noise suppression algorithms in real-time can enhance patients experience with cochlear implants. In this paper, we report the first prototype system implementing such a real-time switching mechanism for automatic selection between two noise suppression algorithms designed for two commonly encountered noisy environments. The results obtained indicate the feasibility of this on-the-fly switching for actual deployment in cochlear implants.

In this paper a new non-invasive method for screening patients with obstructive sleep apnea (OSA) during wakefulness is proposed. Eight people with OSA and eight non-apneic individuals participated in this study. The tracheal breath sound was recorded in supine and upright positions during both nose and mouth breathing maneuvers. Spectral analysis of the respiratory sound signals showed the variation in the average power of the sounds at different positions to be a characteristic feature discriminating the two groups. Using this feature, the OSA and non-apneic participants were classified by quadratic discriminant analysis (QDA). The specificity, sensitivity, and classification accuracy of the classifier were found to be 100%, 87.5%, and 95.75%, respectively. These results are encouraging for the use of the proposed method as a fast, simple and screening tool for diagnosis of OSA during wakefulness.

This paper presents real-time signal processing algorithm for detection of onsets and peaks in Photoplethysmogram (PPG) waveform. Algorithm relies on the analysis of amplitude, slope and inter-beat intervals. The presented algorithm consists of four stages for characterizing PPG waveform. Preprocessing stage involves transformation of PPG since the original waveform is less impulsive and robust. In second stage, algorithm seeks for valid pulse detection in transformed signal complying with the amplitude threshold and inter-beat interval. On detection of valid pulses, algorithm then searches backward and forward in transformed signal for the detection of peaks and onsets. Further the detection parameters are made adaptive to comply with varying beat morphologies and fluctuations in baseline. All signal processing steps and decision logics are implemented with low computational complexity to make it applicable for compact ubiquitous health monitoring devices. On evaluation with our database, the algorithm achieved sensitivity of 96.89% and positive predictivity of 94.55% within an acceptance level of 12 ms.

Many recent research works on gait pattern classification indicates that static features are used. This paper describes of extracting novel dynamic features as complimentary features for the gait pattern classification. The dynamic features are obtained by using regression on the delta zero crossing counts (ΔZCC) of the acceleration signal. The classification results using the filterbank features with the novel dynamic features showed an overall accuracy of 97% was achieved. This is an improvement of 3% from using the filterbank features alone.

The optic disc provides important cues for accurate diagnosis of various retinopathic diseases. Accurate segmentation of the optic disc is therefore an important step in the analysis of retinal images. Gradient vector flow (GVF) based segmentation algorithms have been used successfully on a variety of medical imagery, however, due to the compromise of internal and external energy forces, it can lead to less accurate segmentation in certain cases. In this paper, we show, that through incorporation of a mean shift term into the GVF framework, improved segmentation accuracy can be achieved. Experimental results on a large dataset of retinal images demonstrate that the presented method reliably detects the border of the optic disc.

Soft tissue needle guidance and steering for clinical applications has been an active topic of research in the past decade. Although dynamic feedback control of needle insertion systems is expected to provide more accurate target tracking, it has received little attention due to the fact that most available models for needle-tissue interaction do not incorporate the dynamics of motions. In this paper, we study the controllability of rigid or flexible needles inside soft tissues using mechanical-based dynamic models. The results have significant implications on the design of suitable feedback controllers for different types of needle insertion systems.

This paper presents the development of new prismatic actuation approach and its application in human-safe humanoid head design. To reduce actuator output impedance and mitigate unexpected external shock, the prismatic actuation method uses cables to drive a piston with preloaded spring. By leveraging the advantages of parallel manipulator and cable-driven mechanism, the developed neck has a parallel manipulator embodiment with two cable-driven limbs embedded with preloaded springs and one passive limb. The eye mechanism is adapted for low-cost webcam with succinct "ball-in-socket" structure. Based on human head anatomy and biomimetics, the neck has 3 degree of freedom (DOF) motion: pan, tilt and one decoupled roll while each eye has independent pan and synchronous tilt motion (3 DOF eyes). A Kalman filter based face tracking algorithm is implemented to interact with the human. This neck and eye structure is translatable to other human-safe humanoid robots. The robot's appearance reflects a non-threatening image of a penguin, which can be translated into a possible therapeutic intervention for children with Autism Spectrum Disorders.

Compressed Sensing (CS) has recently been applied to dynamic MRI to improve the acquisition speed. Existing methods exploit the information that the dynamic images are sparse in the spatial and temporal-frequency (y-f) domain. In this paper, we propose to use the additional prior information in CS reconstruction that the support of y-f space is partially known from the motion pattern of dynamic MR images. The reconstruction is then formulated as a truncated ℓ(1) minimization problem. Experimental results show that the dynamic image reconstruction quality of the proposed method is superior to that of existing methods when the same number of measurements is used.

International audience; This work proposes a novel foetal electrocardiogram (FECG) extraction approach based on the cyclostationary properties of the signal of interest. The problem of FECG extraction can easily fit in a blind source separation (BSS) framework; taking into account specific statistical nature of the signal, that one wants to extract, leads to an algorithm able to estimate the FECG contribution to ECG recordings where the maternal ECG is predominant. We show that the proposed procedure provides estimates of the FECGs PQRST complexes without incorporating any prior knowledge concerning PQRST features. Discussions about foetal heart rate variability (HRV) estimation and future works conclude this paper.

This paper presents a shared-control approach for Assistive Mobile Robots (AMR), which depends on the user's ability to navigate a semi-autonomous powered wheelchair, using a sparse and discrete human-machine interface (HMI). This system is primarily intended to help users with severe motor disabilities that prevent them to use standard human-machine interfaces. Scanning interfaces and Brain Computer Interfaces (BCI), characterized to provide a small set of commands issued sparsely, are possible HMIs. This shared-control approach is intended to be applied in an Assisted Navigation Training Framework (ANTF) that is used to train users' ability in steering a powered wheelchair in an appropriate manner, given the restrictions imposed by their limited motor capabilities. A shared-controller based on user characterization, is proposed. This controller is able to share the information provided by the local motion planning level with the commands issued sparsely by the user. Simulation results of the proposed shared-control method, are presented.

Capillary - tissue fluid exchange is controlled by the blood pressure in the capillary and the osmotic pressure of blood (pressure of the tissue fluid outside the capillaries). In this paper, we develop a mathematical model to simulate the movement of bacteria into and within a capillary segment. The model is based on Fokker-Planck equation and Navier-Stocks equations that accounts for different boundary conditions. Also, we model the transportation through capillary walls by means of anisotropic diffusivity that depends on the pressure difference across the capillary walls. By solving the model with a numerical method, it was possible to predict the concentration of bacteria at points within the capillary. However, numerical analysis consumes computational time and resources. To efficiently simulate the bacterial clearance, we propose a segmentation model that is based on breaking the capillary network into smaller sections with pre-defined properties in order to reduce the overall computational time. The proposed model shows a great reduction in computational time and provides accurate results when compared to the numerical analysis.

Automatic or semi-automatic segmentation and tracking of artery trees from computed tomography angiography (CTA) is an important step to improve the diagnosis and treatment of artery diseases, but it still remains a significant challenging problem. In this paper, we present an artery extraction method to address the challenge. The proposed method consists of two steps: (1) a geometric moments based tracking to secure a rough centerline, and (2) a fully automatic generalized cylinder structure-based snake method to refine the centerlines and estimate the radii of the arteries. In this method, a new line direction based on first and second order geometric moments is adopted while both gradient and intensity information are used in the snake model to improve the accuracy. The approach has been evaluated on synthetic images as well as 8 clinical coronary CTA images with 32 coronary arteries. Our method achieves 94.7% overlap tracking ability within an average distance inside the vessel of 0.36 mm.

This paper summarizes dynamic measurements of shear modulus constants acquired for spontaneously growing rat mammary tumors. Measurements are compared with histology to determine tumor types. We also report on 3D shear-wave velocity fields acquired from an inhomogeneous hydrogel phantom with known mechanical properties. Phantom measurements enable us to interpret the effects of tissue structure and geometry on viscoelastic parameter estimates, and demonstrate that reproducible measurements are possible in vivo. Viscoelastic properties describe the mechano-environment of cells undergoing malignant transformation and tumor growth.

Several different algorithms have been proposed for automatic detection of epileptic seizure based on both scalp and intracranial electroencephalography (sEEG and iEEG). Which modality that renders the best result is hard to assess though. From 16 patients with focal epilepsy, at least 24 hours of ictal and non-ictal iEEG were obtained. Characteristics of the seizures are represented by use of wavelet transformation (WT) features and classified by a support vector machine. When implementing a method used for sEEG on iEEG data, a great improvement in performance was obtained when the high frequency containing lower levels in the WT were included in the analysis. We were able to obtain a sensitivity of 96.4% and a false detection rate (FDR) of 0.20/h. In general, when implementing an automatic seizure detection algorithm made for sEEG on iEEG, great improvement can be obtained if a frequency band widening of the feature extraction is performed. This means that algorithms for sEEG should not be discarded for use on iEEG - they should be properly adjusted as exemplified in this paper.

As one widely-used parallel-imaging method, Generalized Auto-calibrating Partially Parallel Acquisitions (GRAPPA) technique reconstructs the missing k-space data by a linear combination of the acquired data using a set of weights. These weights are usually derived from auto-calibration signal (ACS) lines that are acquired in parallel to the reduced lines. In this paper, a cross sampling method is proposed to acquire the ACS lines orthogonal to the reduced lines. This cross sampling method increases the amount of calibration data along the direction that the k-space is undersampled and thus improves the calibration accuracy, especially when a small number of ACS lines are acquired. Both phantom and in vivo experiments demonstrate that the proposed method, named cross-sampled GRAPPA (CS-GRAPPA), can effectively reduce the aliasing artifacts of GRAPPA when high acceleration is desired.

International audience; This paper addresses a localization system which is based on a combination of information from two modalities: a Smart Home Person Tracking (SHPT) composed of infrared sensors and an Audio Person Tracking (APT) which uses microphones able to estimate azimuth of acoustic sources. This combination improves precision of localization compared to a standalone or separated module. The localization software facilitates the integration of both SHPT and APT systems, to display the position in real time, to record data and detect some distress situations (some kind of fall). Results on implementation show good adaptation for Smart Home environments and a robust detection

A major concern in human movement research is preventing tripping and falling which is known to cause severe injuries and high fatalities in elderly (>65 years) populations. Current falls prevention technology consists of active interventions e.g., strength and balance exercises, preimpact fall detectors, and passive interventions e.g., shower rails, hip protectors. However it has been found that these interventions with the exception of balance exercises do not effectively reduce falls risk. Recent work has shown that the minimum toe clearance (MTC) can be successfully monitored to detect gait patterns indicative of tripping and falling risk. In this paper, we investigate the feasibility of predicting MTC values of consecutive gait cycles under different walking speeds. The objective is two-fold, first to determine if end point foot trajectories can be accurately predicted and second, if walking speed is a significant parameter which influences the prediction process. The Generalized Regression Neural Networks and the Support Vector Regressor models were trained to predict MTC time series successively over an increasing prediction horizon i.e., 1 to 10 steps. Increased walking speeds resulted in increased MTC variability but no significant increase in mean MTC height. Root mean squared prediction errors ranged between 2.2–2.6mm or 10% of the mean values of the respective test data. The SVM slightly outperformed the GRNN predictions (0.5%–2.1% better accuracy). Best prediction accuracies decreased by 0.5mm for a doubling of walking speed i.e., from 2.5 km/h to 5.5 km/h. The results are encouraging because they demonstrate that the technique could be applied to forecasting low MTC values and provide new approaches to falls prevention technologies.

In this paper we present a novel hybrid approach for multimodal medical image registration based on diffeomorphic demons. Diffeomorphic demons have proven to be a robust and efficient way for intensity-based image registration. A very recent extension even allows to use mutual information (MI) as a similarity measure to registration multimodal images. However, due to the intensity correspondence uncertainty existing in some anatomical parts, it is difficult for a purely intensity-based algorithm to solve the registration problem. Therefore, we propose to combine the resulting transformations from both intensity-based and landmark-based methods for multimodal non-rigid registration based on diffeomorphic demons. Several experiments on different types of MR images were conducted, for which we show that a better anatomical correspondence between the images can be obtained using the hybrid approach than using either intensity information or landmarks alone.

This work shows the theoretical and practical development of a lung simulator for the calibration of Servoventilators of common use in health centers. It shows the development of a prototype device, Shown in the paper the formulation of a model to consider factors that exist in a human respiratory system in order to simulate normal and pathological conditions. Includes the calculation and construction of electronical and fluidic systems that were developed to set up an emulator that allows real lung adequate to connect with any type of servoventilator; as well as the results in terms of graphics of the required functions, highlighting the practical part that behaves like a real lung subsequently introduced into a torso anthropomorphic designed to better emulate real operating conditions of the lung embedded in a actual context closest to where the components behave as does the lung of a patient.

The paper presents an ongoing investigation into the feasibility of distinguishing between healthy young and older adults, but more specifically into the nature of the features that would provide this distinction. The present study compared the performance of forward, backward, and branch and bound feature selection algorithms when applied to electroencephalography and electromyography data. The results showed that the forward selection algorithm outperformed the other techniques for this particular problem. In addition, time domain features were primarily selected over frequency domain features. Validation of the selected subset suggests the approach is appropriate for future investigation.

Modern control theory allows the representation of cardiac dynamics in the state-space, describing the operation of the vascular systems in terms of the cushioning effect of the arterial wall facing compliance changes. In this paper we use state equations to modeling the effect of the compliance variations on the arterial wall. The characteristics of the dynamics and of the calculated parameters of the model allow the distinction of hypertensive and normotensive subjects, in accordance to real clinical data.

Due to the artifacts in electroencephalography (EEG) data, the performance of brain-computer interface (BCI) is degraded. On the other hand, in the motor imagery based BCI system, EEG signals are usually contaminated by the misleading trials caused by improper imagination of a movement. In this paper, we present a novel algorithm to detect the abnormal EEG data using genetic algorithm (GA). After trial pruning, a subset of the EEG data are selected, on which common spatial pattern (CSP) and Gaussian classifier are trained. The performance of the proposed method is tested on Data set IIa of BCI Competition IV, and the simulation result demonstrates a significant improvement for six out of nine subjects.

In this paper, a feature extraction method based on principal component analysis was developed for classification of the vascular access's condition in hemodialysis patients. The assessment of the method was carried out by discriminating between before and after angioplasty sound recordings as well as before angioplasty and reference recordings. The results showed that when before and after angioplasty recordings were compared by patient, the classification agreed with the result of angioplasty procedure. When all the available before and after angioplasty recordings were compared, it was still possible to discriminate them at a good rate. On the other hand, when the reference recordings substituted the after angioplasty recordings, almost a perfect discrimination was achieved.

This paper presents a new approach that improves discriminative training criterion for Hidden Markov Models, and is oriented to pathological voice identification. This technique is aimed at maximizing the Area under the Curve of a receiver operating characteristic curve by adjusting the model parameters using as objective function the Mahalanobis distance and the distance between means of the underlying probability density functions associated with each class. The results show that the proposed technique significantly outperforms the accuracy in a classification system compared with other training criteria. Results are provided using the MEEIVL voice disorders database.

Dendritic spines play an essential role in the central nervous system. Recent experiments have revealed that neuron functional properties are highly correlated with the statistical and morphological changes of the dendritic spines. In this paper, we propose a new multi scale approach for detecting dendritic spines in a 2D Maximum Intensity Projection (MIP) image of the 3D neuron data stacks collected from a 2-photon laser scanning confocal microscope. The proposed method utilizes the curvilinear structure detector in conjunction with the multi scale spine detection algorithm which automatically and accurately extracts and segments the spines with variational sizes along the dendrite. In addition, a slice-based spine detection algorithm is also proposed to detect spines which are hidden from the MIP image within the dendrite area. Experimental results show that our proposed method is effective for automatic spine detection and is able to accurately segment dendrite.

To help people find desired home medical products (HMPs), we developed an intelligent personal health record (iPHR) system that can automatically recommend HMPs based on users' health issues. Using nursing knowledge, we pre-compile a set of "search guide" phrases that provides semantic translation from words describing health issues to their underlying medical meanings. Then iPHR automatically generates queries from those phrases and uses them and a search engine to retrieve HMPs. To avoid missing relevant HMPs during retrieval, the compiled search guide phrases need to be comprehensive. Such compilation is a challenging task because nursing knowledge updates frequently and contains numerous details scattered in many sources. This paper presents a semi-automatic tool facilitating such compilation. Our idea is to formulate the phrase compilation task as a multi-label classification problem. For each newly obtained search guide phrase, we first use nursing knowledge and information retrieval techniques to identify a small set of potentially relevant classes with corresponding hints. Then a nurse makes the final decision on assigning this phrase to proper classes based on those hints. We demonstrate the effectiveness of our techniques by compiling search guide phrases from an occupational therapy textbook.

Laser speckle contrast imaging (LSCI) has classically been used to image regional blood flow changes in animal models. In this paper, we demonstrate the use of LSCI for elucidating blood flow characteristics in individual microvessels with diameters as small as 24µm. We extracted profiles of speckle contrast values within individual vessels, both along their diameters and along their lengths and inferred that they could be attributed to the flow within the vessel. Profiles along the diameter of vessels revealed maxima at the center of vessels, consistent with fluid dynamics. These observed profiles could be fitted with parabolic curves with a mean coefficient of determination of 0.92. Similarly, analysis of speckle contrast values in the axial direction revealed profiles that progressively decreased in discreet quanta at branch points indicating blood flow bifurcations. Flow estimates obtained from speckle contrast values within branches of vessels obeyed the law of mass conservation with a mean error of only 3.5%. This allowed us to elucidate the percentage distribution of blood flow into each of the downstream branches. This ability of LSCI to resolve blood flow distribution in branching microvessel trees in a minimally invasive and dye free environment over a wide field of view promises to find application in both the neuroscience laboratory as well as intraoperative neurosurgery.

Mass spectrometry-based metabolomics is getting mature and playing an ever important role in the systematic understanding of biological process in conjunction with other members of "-omics" family. However, the identification of metabolites in untargeted metabolomics profiling remains a challenge. In this paper, we propose a support vector machine (SVM)-based spectral matching algorithm to combine multiple similarity measures for accurate identification of metabolites. We compared the performance of this approach with several existing spectral matching algorithms on a spectral library we constructed. The results demonstrate that our proposed method is very promising in identifying metabolites in the face of data heterogeneity caused by different experimental parameters and platforms.

The paper offers an enhancement of the classical interrupter technique algorithm dedicated to respiratory mechanics measurements. Idea consists in exploitation of information contained in postocclusional transient states during indirect measurement of parameter characteristics by model identification. It needs the adequacy of an inverse analogue to general behavior of the real system and a reliable algorithm of parameter estimation. The second one was a subject of reported works, which finally showed the potential of the approach to separation of airway and tissue response in a case of short-term excitation by interrupter valve operation. Investigations were conducted in a regime of forward-inverse computer experiment.

This paper addresses error modeling in A-Mode ultrasound- (US-) based registration and integration of model-based weighting into the Random-ICP (R-ICP) algorithm. The R-ICP is a variant of the Iterative Closest Point (ICP) algorithm, and it was suggested for surface-based registration using A-Mode US in the context of skull surgery. In that application area the R-ICP could yield high accuracy even in case of a small number of data points and a very inaccurate user-interactive pre-registration. However, it cannot cope with unequal point uncertainty, which is an important drawback in the context of hip surgery: Uncertainty about the average speed of sound is an error source, whose impact on the registration accuracy increases with the thickness of the scanned soft tissue. It can, therefore, lead to considerable localization errors if a thick soft tissue layer is scanned, and it might vary a lot from data point to data point as the soft tissue thickness is inhomogeneous. The present work investigates how to account for this error source considering also other error sources such as the establishment of point correspondences. Simulation results show that registration accuracy can be substantially improved when model-based weighting is integrated into the R-ICP.