Search

In this paper, we explore the application of total variation regularization method for bioluminescence tomography (BLT) with an adaptive regularization parameter choice approach. Since BLT is a seriously ill-posed problem, therefore, l(2) regularized methods are frequently adopted to recover the bi-oluminescent sources. However, l(2) regularized methods typically lead to smooth reconstructions. In this paper, we investigated the use of total variation (TV) regularization to improve the quality of BLT reconstruction. Furthermore, the regularization parameter in TV method was chosen adaptively to make the proposed algorithm more stable. Results on simulation data provide evidence that the reconstructed source can be localized accurately compared with l(2) method. Meanwhile, the effectiveness of utility of the parameter choice were illustrated. Finally, different levels of noisy data were added to validate the performance of the proposed algorithm.

This paper proposes a novel approach for segmenting fetal ultrasound images. This problem presents a variety of challenges including high noise, low contrast, and other US imaging properties such as similarity between texture and gray levels of two organs/ tissues. In this paper, we have proposed a Conditional Random Field (CRF) based framework to handle challenges in segmenting fetal ultrasound images. Clinically, it is known that fetus is surrounded by specific maternal tissues, amniotic fluid and placenta. We exploit this context information using CRFs for segmenting the fetal images accurately. The proposed CRF framework uses wavelet based texture features for representing the ultrasound image and Support Vector Machines (SVM) for initial label prediction. Initial results on a limited dataset of real world ultrasound images of fetus are promising. Results show that proposed method could handle the noise and similarity between fetus and its surroundings in ultrasound images.

In this paper, particle swarm optimization (PSO) algorithm is used to estimate the permittivities of the tissue layers at microwave frequency band. According to the literature, microwave radiometry (MWR) is potentially a promising cancer detection technique. In addition, breast cancer is an appropriate candidate of MWR due to the breast's exclusive physiology. Several algorithms have been evaluated for analyzing the measurement data and solving the inverse scattering problem in MWR, and different levels of accuracy have been reported. In this paper, the potential of PSO in solving this problem is demonstrated at 1-2.25 GHz. Two distinct algorithms are developed for the two considered scenarios. In the first scenario, we assume no a priori knowledge of the tissue under the test, whereas, in the second scenario, a priori knowledge is assumed. It is noteworthy that, there are only a few research articles studying PSO for permittivity estimation. However, since these studies underestimate the loss encountered by the test samples, the methods are not valid for body tissue case. Here, measurement-based loss coefficients, reported in the existing literature, are included in the calculations. It is shown that the algorithm converges relatively fast, and, distinguishes between different tissues with an acceptable accuracy.

A signal processor/compressor dedicated to implantable neural recording microsystems is presented. Signal compression is performed based on Haar wavelet. It is shown in this paper that, compared to other mathematical transforms already used for this purpose, compression of neural signals using this type of wavelet transform can be of almost the same quality, while demanding less circuit complexity and smaller silicon area. Designed in a 0.13-μm standard CMOS process, the 64-channel 8-bit signal processor reported in this paper occupies 113 μm x 110 μm of silicon area. It operates under a 1.8-V supply voltage at a master clock frequency of 3.2 MHz.

In contrast to the muscle recruitment during voluntary walking, only a limited number of muscles are activated during functional electrical stimulation (FES)-based walking. This implies that a trajectory designed or recorded from the normal human walking data may not be the best choice for tracking control. Another major challenge during FES-based walking is the rapid onset of muscle fatigue. Two methods to reduce fatigue during FES-based walking are employing an orthosis and minimizing muscle activations. To deal with these aforementioned challenges, this paper presents firstly a dynamic model representing FES-elicited walking constrained by an orthosis and a walker. Secondly, this paper deals with the design of optimal stimulation and force profiles (instead of gait-trajectories from able-bodied humans) that minimize muscle activations via FES and arm reaction forces from the walker. Ten walking steps are simulated to show the feasibility of the walking model and optimization algorithm.

This paper presents a time-varying cortical connectivity analysis for to a source localization approach within the inner layers of the newborn brain. (Additional details can be found in the comprehensive book on Time-Frequency Signal Analysis and Processing (see http://www.elsevier.com/locate/isbn/0080443354). In addition, the most recent upgrade of the original software package that calculates Time-Frequency Distributions and Instantaneous Frequency estimators can be downloaded from the web site: www.time-frequency.net. This was the first software developed in the field, and it was first released publicly in 1987 at the 1st ISSPA conference held in Brisbane, Australia, and then continuously updated). Multivariate Granger causality in the timefrequency domain as a representation of time-varying cortical connectivity in the brain has been investigated for the adult case. This is, however, not the case in newborns as the nature of the transient changes in the newborn EEG is different from that of adults. This paper aims to evaluate the performance of the time-varying versions of the two popular Granger causality measures, namely Partial Directed Coherence (PDC) and direct Directed Transfer Function (dDTF). The parameters of the time-varying AR, that models the inter-channel interactions, are estimated using Dual Extended Kalman Filter (DEKF) as it accounts for both non-stationarity and non-linearity behaviors of the EEG. Using simulated data, we show that fast changing cortical connectivity between channels can be measured more accurately using the time-varying PDC. The performance of the time-varying PDC is also tested on a neonatal EEG exhibiting seizure.

This paper introduces a kernel adaptive filter implemented with stochastic gradient on temporal differences, kernel Temporal Difference (TD)(λ), to estimate the state-action value function in reinforcement learning. The case λ=0 will be studied in this paper. Experimental results show the method's applicability for learning motor state decoding during a center-out reaching task performed by a monkey. The results are compared to the implementation of a time delay neural network (TDNN) trained with backpropagation of the temporal difference error. From the experiments, it is observed that kernel TD(0) allows faster convergence and a better solution than the neural network.

Large variations in Surface Electromyogram (SEMG) signal across different subjects make the process of automated signal classification as a generalized tool, challenging. In this paper, we propose a domain adaptation methodology that addresses this challenge. In particular we propose a hierarchical sample selection methodology, that selects samples from multiple training subjects, based on their similarity with the target subject at different levels of granularity. We have validated our framework on SEMG data collected from 8 people during a fatiguing exercise. Comprehensive experiments conducted in the paper demonstrate that the proposed method improves the subject independent classification accuracy by 21% to 23% over the cases without domain adaptation methods and by 14% to 20% over the existing state-of-the-art domain adaptation methods.

This paper describes the initial groundwork car- ried out as part of the European Commission funded Transatlantic Tumor Model Repositories project, to develop a new markup language for computational cancer modelling, TumorML. In this paper we describe the motivations for such a language, arguing that current state-of-the-art biomodelling languages are not suited to the cancer modelling domain. We go on to describe the work that needs to be done to develop TumorML, the conceptual design, and a description of what existing markup languages will be used to compose the language specification.

This paper presents the design of a wireless event detection and in building location awareness system. The systems architecture is based on using a body worn sensor to detect events such as falls where they occur in an assisted living environment. This process involves developing event detection algorithms and transmitting such events wirelessly to an in house network based on the 802.15.4 protocol. The network would then generate alerts both in the assisted living facility and remotely to an offsite monitoring facility. The focus of this paper is on the design of the system architecture and the compliance challenges in applying this technology.

Three-dimensional models of the spine are commonly used to diagnose, to treat, and to study spinal deformities. Creating these models is however time-consuming and, therefore, expensive. We propose in this paper a reconstruction method that finds the most likely 3D reconstruction given a maximal error bound on a limited set of landmark locations supplied by the user. This problem can be solved using second-order cone programming, leading to a globally convergent method that is considerably faster than currently available methods. A user can, with our current implementation, interactively modify the landmark locations and receive instantaneous feedback on the effect of those changes on the 3D reconstruction instead of blindly selecting landmarks. The proposed method was validated on a set of 53 patients who had adolescent idiopathic scoliosis using real and synthetic tests. Test results showed that the proposed method is considerably faster than currents methods (about forty times faster), is extremely flexible, and offers comparable accuracy., IEEE EMBC (Engineering in Medicine and Biology Conference), August 30 - September 3, 2011, Boston, MA, USA

In order to improve the quality of life of persons who lost their limb due to an accident or a sickness, many myoelectric artificial arms have been proposed. To control the motion of the elbow joint of the myoelectric arm, the EMG signals of the biceps brachii and the triceps brachii muscles are frequently used as input signals. However, since both muscles are the biarticular muscles between the shoulder joint and elbow joint, the amount of the EMG signals of these muscles are affected by the motion of the shoulder joint. In this paper, a control method for a myoelectric arm is proposed in order to cancel the influence of the shoulder joint motion on EMG signals is removed by using the neuro-fuzzy modifier in order to realize proper elbow motion. The effectiveness of the proposed method has been evaluated by performing experiments.

Algebraic Reconstruction Technique (ART) is a widely employed method in computed tomography since it has certain advantages, such as allowing reconstruction of data with missing projections in some angle ranges, over other techniques such as Filtered Back Projection (FBP). Recently, a regularisation technique for ART, RegART, was introduced which provides greatly reduced noise levels. However, a serious drawback of both ART and RegART is the computational complexity of the methods. In this paper, we present a fast version of RegART, which makes use of nVidia's CUDA technology, and show that this approach performs favourably compared to FBP.

The complexity associated with musculoskeletal modeling, simulation, and neural control of the human spine is a challenging problem in the field of biomechanics. This paper presents a novel method for simulation of a 3D trunk model under control of 48 muscle actuators. Central pattern generators (CPG) and artificial neural network (ANN) are used simultaneously to generate muscles activation patterns. The parameters of the ANN are updated based on a novel learning method used to address the kinetic redundancy due to presence of 48 muscles driving the trunk. We demonstrated the feasibility of the proposed method with numerical simulation of experiments involving rhythmic motion between upright standing and 55 degrees of flexion. The tracking performance of the model is accurate to within 2° while reciprocal muscle activation patterns were similar to the observed experimental coordination patterns in normal subjects. The suggested method can be used to map high-level control strategies to low-level control signals in complex biomechanical and biorobotic systems. This will also provide insight about underlying neural control mechanisms.

Epilepsy is one of the most common brain disorders in the world. The spontaneous seizure onset influences the daily life of epilepsy patients. The studies on feature extraction and feature classification from Electroencephalography(EEG) signal in seizure prediction methods have shown great improvement these years. However, the variation issue of EEG signal (being awake, being asleep, severity of epilepsy, etc.) poses a fundamental difficulty in seizure prediction problem. The traditional off-line training method trains the model using a fixed training set, and expects the performance of the model to remain stable even after a long period of time, and thus suffers from variation issue. In this paper, we propose an on-line retraining method to leverage the recent input data by gradually enlarging the training set and retraining the model. Also, a simple post-processing scheme is incorporated to reduce false alarms. We develop our method based on the state of the art machine learning based classification of bivariate patterns method. The performance of the method is evaluated on Electrocorticogram(ECoG) recording from Freiburg database as well as long-term scalp EEG recording from CHB-MIT EEG Database and National Taiwan University Hospital. The proposed method achieves 74.2% sensitivity on ECoG database and 52.2% sensitivity on scalp EEG database, while improving the sensitivity of off-line training method by 29.0% and 17.4% in ECoG database and EEG database respectively. The experimental result suggests that on-line retraining can greatly improve the reliability and is promising for future seizure prediction method development.

This paper presents the evaluation of the accuracy of an elastic registration algorithm, based on the particle filter and an optical flow process. The algorithm is applied in brain CT and MRI simulated image datasets, and MRI images from a real clinical radiotherapy case. To validate registration accuracy, standard indices for registration accuracy assessment were calculated: the dice similarity coefficient (DICE), the average symmetric distance (ASD) and the maximal distance between pixels (Dmax). The results showed that this registration process has good accuracy, both qualitatively and quantitatively, suggesting that this method may be considered as a good new option for radiotherapy applications like patient's follow up treatment.

The Cognitive State Assessment Competition 2011 was organized by the U.S. Air Force Research Laboratory (AFRL) to compare the performance of real-time cognitive state classification software. This paper presents results for QUASAR's data classification module, QStates, which is a software package for real-time (and off-line) analysis of physiologic data collected during cognitive-specific tasks. The classifier's methodology can be generalized to any particular cognitive state; QStates identifies the most salient features extracted from EEG signals recorded during different cognitive states or loads.

This paper presents an approach for measuring and monitoring human body joint angles using inertial measurement unit (IMU) sensors. This type of monitoring is beneficial for therapists and physicians because it facilitates remote assessment of patient activities. In our approach, two IMUs are mounted on the upper leg and the lower leg to measure the Euler angles of each segment. The Euler angles are sent via Bluetooth protocols to a pc for calculating the knee joint angle. In our experiments, we utilized a motion capture system to accurately measure the knee joint angle and used this as the ground truth to assess the accuracy of the IMU system. The range of average error of the system across a variety of motion trials was 0.08 to 3.06 degrees. In summary, the accuracy of the IMU measurement system currently outperforms existing wearable systems such as conductive fiber optic sensors and flex-sensors.

Many computer users suffer varying degrees of visual impairment, which hinder their interaction with computers. In contrast with available methods of vision correction (spectacles, contact lenses, LASIK, etc.), this paper proposes a vision correction method for computer users based on image pre-compensation. The blurring caused by visual aberration is counteracted through the pre-compensation performed on images displayed on the computer screen. The pre-compensation model used is based on the visual aberration of the user's eye, which can be measured by a wavefront analyzer. However, the aberration measured is associated with one specific pupil size. If the pupil has a different size during viewing of the pre-compensated images, the pre-compensation model should also be modified to sustain appropriate performance. In order to solve this problem, an adjustment of the wavefront function used for pre-compensation is implemented to match the viewing pupil size. The efficiency of these adjustments is evaluated with an "artificial eye" (high resolution camera). Results indicate that the adjustment used is successful and significantly improves the images perceived and recorded by the artificial eye.

Sparse approximation is a novel technique in applications of event detection problems to long-term complex biomedical signals. It involves simplifying the extent of resources required to describe a large set of data sufficiently for classification. In this paper, we propose a multivariate statistical approach using dynamic principal component analysis along with the non-overlapping moving window technique to extract feature information from univariate long-term observational signals. Within the dynamic PCA framework, a few principal components plus the energy measure of signals in principal component subspace are highly promising for applying event detection problems to both stationary and non-stationary signals. The proposed method has been first tested using synthetic databases which contain various representative signals. The effectiveness of the method is then verified with real EEG signals for the purpose of epilepsy diagnosis and epileptic seizure detection. This sparse method produces a 100% classification accuracy for both synthetic data and real single channel EEG data.

Infering causal relationships from observed time series has attracted much recent attention. In cases of nonlinear coupling, adequate inference is often hindered by the need to specify coupling details that call for many parameters and global minimization of nonconvex functions. In this paper we use an example to investigate a new concept, termed here running entropy mapping, whereby time series are mapped onto other entropy related time sequences whose analysis via a linear parametric time series methods, such as partial directed coherence, is able to expose the presence of formerly linearly undetectable causal relationships.

Microarray analysis can contribute considerably to the understanding of biologically significant cellular mechanisms that yield novel information regarding co-regulated sets of gene patterns. Clustering is one of the most popular tools for analyzing DNA microarray data. In this paper, we present an unsupervised clustering algorithm based on the K-local hyperplane distance nearest-neighbor classifier (HKNN). We adapted the well-known nearest neighbor clustering algorithm for use with hyperplane distance. The result is a simple and computationally inexpensive unsupervised clustering algorithm that can be applied to high-dimensional data. It has been reported that the NFkB1 gene is progressively over-expressed in moderate-to-severe Alzheimer's disease (AD) cases, and that the NF-kB complex plays a key role in neuroinflammatory responses in AD pathogenesis. In this study, we apply the proposed clustering algorithm to identify co-expression patterns with the NFkB1 in gene expression data from hippocampal tissue samples. Finally, we validate our experiments with biomedical literature search.

Optical imaging in vivo is an important tool for allowing researchers to understand neural ensemble interactions during awake behavior, sleep, anesthesia and during seizure activity. A major bottleneck in the overall efficiency of neural imaging experiments is the need for post-hoc analysis of imaging data. Computational capabilities are now at the point where real- or near-real-time multivariate analysis of imaging data is possible as data is acquired. In this paper we address the feasibility of performing real-time data analysis with a desktop computer, MATLAB, and a graphics processing unit (GPU). Important components of any real-time functional imaging analysis system are 1) dimensional reduction of the data, 2) visualization of the reduced vector space and 3) rapid calculation of functional connectivities. The ability to assess sources of variability in the data, and connectivity estimates on the fly, are potentially transformative for the way imaging laboratories perform their work. Here, we present benchmarks for analysis of functional imaging data using dimensional reduction methods and estimation of functional connectivities using least-squares and ridge regression methods.

fNIRS recordings are increasingly utilized to monitor brain activity in both clinical and connected health settings. These optical recordings provide a convenient measurement of cerebral hemodynamic changes which can be linked to motor and cognitive performance. Such measurements are of clinical utility in a broad range of conditions ranging from dementia to movement rehabilitation therapy. For such applications fNIRS is increasingly deployed outside the clinic for patient monitoring in the home. However, such a measurement environment is poorly controlled and motion, in particular, is a major source of artifacts in the signal, leading to poor signal quality for subsequent clinical interpretation. Artifact removal techniques are increasingly being employed with an aim of reducing the effect of the noise in the desired signal. Currently no methodology is available to accurately determine the efficacy of a given artifact removal technique due to the lack of a true reference for the uncontaminated signal. In this paper we propose a novel methodology for fNIRS data collection allowing for effective validation of artifact removal techniques. This methodology describes the use of two fNIRS channels in close proximity allowing them to sample the same measurement location; allowing for the introducing of motion artifact to only one channel while having the other free of contamination. Through use of this methodology, for each motion artifact epoch, a true reference for the uncontaminated signal becomes available for use in the development and performance evaluation of signal processing strategies. The advantage of the described methodology is demonstrated using a simple artifact removal technique with an accelerometer based reference.

Optic disc segmentation from retinal fundus image is a fundamental but important step for automatic glaucoma diagnosis. In this paper, an optic disc segmentation method is proposed based on peripapillary atrophy elimination. The elimination is done through edge filtering, constraint elliptical Hough transform and peripapillary atrophy detection. With the elimination, edges that are likely from non-disc structures especially peripapillary atrophy are excluded to make the segmentation more accurate. The proposed method has been tested in a database of 650 images with disc boundaries marked by trained professionals manually. The experimental results by the proposed method show average m(1), m(2) and m(VD) of 10.0%, 7.4% and 4.9% respectively. It can be used to compute cup to disc ratio as well as other features for application in automatic glaucoma diagnosis systems.

Most pulse oximeters determine blood oxygen saturation (SpO(2)) after calculating a coefficient, R, that represents the normalized ratiometric contributions of the pulsatile red and near-infrared photoplethysmograms (PPGs) acquired by the sensor. This paper presents a new approach that uses principle component analysis (PCA) to separate the signal and noise components of unfiltered PPGs and provide the determination of R. Also, rather than use peak-to-valley time intervals to obtain R, this technique uses eigenvalue and eigenvector data obtained during PCA to optimize these time intervals and improve the R calculation. Early analyses on unfiltered PPGs from 16 subjects indicate that these R values compare to those obtained from FFT-based methods and yield SpO(2) values consistent with those reported by a commercial unit. All signal data are considered during the PCA process, so this technique shows promise to precisely segment clean versus noise-corrupted PPGs.

Head motion during brain CT studies can degrade the reconstructed image by introducing distortion and loss of resolution, thereby contributing to misdiagnosis of diseases. In this paper, we have proposed a correlation coefficient and Least Squares Support Vector Machines (LS-SVM) based approach to detect and mitigate motion artifacts in FDK based three-dimensional cone-beam tomography. Motion is detected using correlation between adjacent x-ray projections. Artifacts, caused by motion, are mitigated either by replacing motion corrupted projections with their counterpart 180° apart projections under certain conditions, or by estimating motion corrupted projections using LS-SVM based time series prediction. The method has been evaluated on 3D Shepp-Logan phantom. Simulation results validate our claims.

Glioma histologies are the primary factor in prognostic estimates and are used in determining the proper course of treatment. Furthermore, due to the sensitivity of cranial environments, real-time tumor-cell classification and boundary detection can aid in the precision and completeness of tumor resection. A recent improvement to mass spectrometry known as desorption electrospray ionization operates in an ambient environment without the application of a preparation compound. This allows for a real-time acquisition of mass spectra during surgeries and other live operations. In this paper, we present a framework using sparse kernel machines to determine a glioma sample’s histopathological subtype by analyzing its chemical composition acquired by desorption electrospray ionization mass spectrometry.

Research indicates that long-term monitoring of vital signs and activity in elderly adults may provide opportunities for maintaining quality-of-life and extending independence into later years. Such a strategy requires development of a system to collect this data while imposing minimal intrusion into the lives of those being monitored. To further this goal, we have developed a hydraulic bed sensor to non-invasively monitor heartbeat and respiration during sleep. This paper describes the refinement of our developed prototype and signal processing methods, along with an evaluation of the robustness of our algorithms and results from testing. An evaluation of our sensor on a group of five diverse subjects (ranging in age from 24 to 67, two with cardiac history), in three different positions, demonstrates accuracy within 8 beats per minute up to 97.5% of the time.

In this paper we implement the 7-point checklist, a set of dermoscopic criteria widely used by clinicians for melanoma detection, on smart handheld devices, such as the Apple iPhone and iPad. The application developed is using sophisticated image processing and pattern recognition algorithms, yet it is light enough to run on a handheld device with limited memory and computational speed. When combined with a commercially available handheld dermoscope that provides proper lesion illumination, this application provides a truly self-contained handheld system for melanoma detection. Such a device can be used in a clinical setting for routine skin screening, or as an assistive diagnostic device in underserved areas and in developing countries with limited healthcare infrastructure.

The estimation of current distributions from electroencephalographic recordings poses an inverse problem, which can approximately be solved by including dynamical models as spatio-temporal constraints onto the solution. In this paper, we consider the electrocardiography source localization task, where a specific structure for the dynamical model of current distribution is directly obtained from the data by fitting multivariate autoregressive models to electroencephalographic time series. Whereas previous approaches consider an approximation of the internal connectivity of the sources, the proposed methodology takes into account a realistic structure of the model estimated from the data, such that it becomes possible to obtain improved inverse solutions. The performance of the new method is demonstrated by application to simulated electroencephalographic data over several signal to noise ratios, where the source localization task is evaluated by using the localization error and the data fit error. Finally, it is shown that estimating MVAR models makes possible to obtain inverse solutions of considerably improved quality, as compared to the usual instantaneous inverse solutions, even if the regularized inverse of Tikhonov is used.

In this paper the statistical properties of the swallowing sound is discussed. This knowledge is required for the acoustical modeling of the swallowing mechanism as it is important to select an appropriate type of the system (i.e. linear vs. nonlinear) for modeling. The tests of linearity and gaussianity were performed. The results of the statistical test of gaussianity showed a nonGaussian distribution of the swallowing sound signals. Also, the test of linearity exhibited the nonlinear characteristics of the model that represents the swallowing sound generation.

In this paper we introduce a robust classification framework for tongue-movement ear pressure signals based around an ensemble voting methodology. The ensemble members are comprised of different combinations of sensor inputs i.e. two in-ear microphones and an acoustic gel sensor positioned under the chin of the individual and classification using three different base models. It is shown that by using all nine ensemble members when compared to the individual (base) models, the average misclassification rate can be reduced from 23% to 2.8% when using the majority voting strategy. The correct classification rate is improved from 76% to 92.4% when utilizing either the borda count or condorcet methods. This is achieved through a combination of rejection based on ambiguity in the ensemble and diversity in the misclassified instances across the ensemble members.

In this paper, a novel technique based on signal processing of breath sounds during wakefulness for prediction of obstructive sleep apnea (OSA) is proposed. We recorded tracheal breath sounds of 35 people with various severity of OSA and 17 non-apneic individuals; the breath sounds were recorded in supine and upright positions during both nose and mouth breathing at medium flow rate. Power spectrum, Kurtosis and Katz fractal dimension of the recorded signals in every posture and breathing maneuver were calculated. We used one-way ANOVA to select the features with most significant differences between the groups followed by the Maximum Relevancy Minimum Redundancy (mRMR) method to reduce the number of characteristic features to three, and investigated the separability of the groups based on the three selected features. The results are encouraging for classification of patients using the selected features. Once being verified on a larger population, the proposed method offers a fast, simple and non-invasive screening tool for prediction of OSA during wakefulness.

Many methods for inferring genetic regulatory networks have been proposed. However inferred networks can hardly be used to analyze the dynamics of genetic regulatory networks. Recently nonlinear differential equations are proposed to model genetic regulatory networks. Based on this kind of model, the stability of genetic regulatory networks has been intensively investigated. Because of difficulty in estimating parameters in nonlinear model, inference of genetic regulatory networks with nonlinear model has been paid little attention. In this paper, we present a method for estimating parameters in genetic regulatory networks with SUM regulatory logic. In this kind of genetic regulatory networks, a regulatory function for each gene is a linear combination of Hill form functions, which are nonlinear in parameters and in system states. To investigate the proposed method, the gene toggle switch network is used as an illustrative example. The simulation results show that the proposed method can accurately estimates parameters in genetic regulatory networks with SUM logic.

The detection of swallowing events by acoustic means represents an important tool to assess and diagnose swallowing disorders as well as to objectively monitor ingestive behavior of individuals. Acoustic sensors used to register swallowing sounds may also capture sound artifacts arising from intrinsic speech and external noise affecting the detection. In this paper we tested if subsonic frequencies are less prone to artifacts from speech, chewing and other intrinsic sounds than sonic frequencies. A simple method using a throat and an ambient microphone was employed to compare the swallowing detection accuracy by acoustic signals acquired in the sonic (20-2500 Hz) and subsonic (≤ 5 Hz) ranges. Averaged recall values were higher than 85% for both ranges. However, averaged precision values of 50% for subsonic frequencies and of 42% for sonic frequencies were caused by a high number of false positives. These results indicated no significant difference between averaged precision values which may suggest that subsonic frequencies were not less prone to intrinsic sound artifacts than frequencies in the sonic range. Further examination with the addition of a signal classification layer is proposed as a future step to confirm this statement.

Arrhythmia (i.e., irregular cardiac beat) classification in electrocardiogram (ECG) signals is an important issue for heart disease diagnosis due to the non-invasive nature of the ECG exam. In this paper, we analyze and criticize the results of some arrhythmia classification methods presented in the literature in terms of how the samples are chosen for training/testing the classifier and the impact this choice has on their performance (i.e., accuracy/sensitivity/specificity). From our implementation, we also report new accuracies for these methods, establishing a new state-of-the-art method, in terms of results.

This paper presents a method to automatically recognize events and driving activities during the use of a powered wheelchair (PW). The method uses a support vector machine classifier, trained from sensor-based data from a datalogging platform installed on the PW. Data from a 3D accelerometer positioned on the back of the PW were collected in a laboratory space during PW driving tasks. 16-segmented events and driving activities (i.e. impacts from different side on different objects, rolling down or up on incline surface, going across threshold of different height) were performed repeatedly (n=25 trials) by one operator at three different speeds (slow, normal, high). We present results from an experiment aiming to classify five different events and driving activities from the sensor data acquired using the datalogging platform. Classification results show the ability of the proposed method to reliably segment 100% of events, and to identify the correct event type in 80% of events.

This paper introduces SAGE — an algorithm that uses the spatial clustering of objects to enhance their classification. It assumes that discrete objects can be identified and classified based on their individual appearance, and further that they tend to appear in spatial clusters (for example, circinate exudates). The algorithm builds spatial distribution maps for objects and confounds for a given image, and adjusts individual object confidence levels to reflect their spatial clustering. SAGE may be combined with a wide range of object identification and classification methods; we demonstrate it using a Multi-Layered Perceptron (MLP) Neural Network and a Support Vector Machine (SVM) classifier types for both dark and bright retinal lesions. Using ROC analysis SAGE improves classifier performance as much as 83%.

Thermal infrared imaging (TIRI) employs a focal plane array (FPA) of infrared detectors, with associated optics and optoelectronics to remotely detect and topographically map thermal emittance. Thermal and optical properties of human physioanatomy are not fully understood yet confounding diagnostic interpretation of human TIRI's. Elucidation of the specific physical mechanism via which thermal emission arises from human anatomy in-vivo requires empirical investigation under objective clinical protocols. This paper characterizes the fundamental architecture of the clinical TIRI system with a view to facilitation of objective protocol development, elucidation of the mechanism/s of human thermal infrared emittance, and eventual validation of TIRI as a diagnostic medical tool. Relevant recent and ongoing empirical studies by the authors are also summarized.

Hemorrhage is the main cause of deaths that occurs within first 24 hours after a traumatic pelvic injury. Therefore, it is very important to determine hemorrhage quickly. Hemorrhages are detected using a CT scan. However, it is very time consuming for physicians to look for hemorrhage in all CT slices. Therefore, an automated system is needed. This paper proposes an automated hemorrhage detection technique by incorporating anatomical information of pelvic region. The results showed method performs comparably to manual methods. A statistical test is conducted to see if the volume of hemorrhage detected using this technique is significantly different from the volume assessed manually.

International audience; In this paper, we propose a new joint watermarking/encryption algorithm for the purpose of verifying the reliability of medical images in both encrypted and spatial domains. It combines a substitutive watermarking algorithm, the quantization index modulation (QIM), with a block cipher algorithm, the Advanced Encryption Standard (AES), in CBC mode of operation. The proposed solution gives access to the outcomes of the image integrity and of its origins even though the image is stored encrypted. Experimental results achieved on 8 bits encoded Ultrasound images illustrate the overall performances of the proposed scheme. By making use of the AES block cipher in CBC mode, the proposed solution is compliant with or transparent to the DICOM standard.

In this paper we describe a class of analog algorithms based on the concept of cellular neural networks (CNNs) for detecting osteoarthritis (OA) from x-ray images. The indicator of OA that we examine is the presence of bony spurs or osteophytes in the vicinity of the weight bearing joints of the fingers. Results on a series of hand x-ray images are shown to be promising.

Due to the rapidly aging population around the world, senile dementia is growing into a prominent problem in many societies. To monitor the elderly dementia patients so as to assist them in carrying out their basic Activities of Daily Living (ADLs) independently, sensors are deployed in their homes. The sensors generate a stream of context information, i.e., snippets of the patient's current happenings, and pattern mining techniques can be applied to recognize the patient's activities based on these micro contexts. Most mining techniques aim to discover frequent patterns that correspond to certain activities. However, frequent patterns can be poor representations of activities. In this paper, instead of using frequent patterns, we propose using correlated patterns to represent activities. Using simulation data collected in a smart home testbed, our experimental results show that using correlated patterns rather than frequent ones improves the recognition performance by 35.5% on average.

Accuracy and robustness are fundamental requirements of any automated method used for segmentation of medical images. Model-based segmentation (MBS) is a well established technique, where uncertainties in image content can be to a certain extent compensated by the use of prior shape information. This approach is, however, often problematic in cases where image information does not allow for generating a strong feature response, one example being soft tissue organs in CT data, which typically appear in low contrast. In this paper, we enhance our recently proposed framework for voxel classification-based refinement of MBS using a level-set segmentation technique with shape priors. We also introduce a novel feature weighting methodology that improves the performance of the classifier, demonstrating results superior to the previous feature selection method. Results of fully automated segmentation of low contrast organs in head and neck CT are presented. Compared to our previous approach, we have achieved an increase of up to 22% in segmentation accuracy.

This paper describes a novel driving control scheme of electric power-assisted wheelchairs for assistive driving on various large disturbance roads. The "electric power-assisted wheelchair" which assists the driving force by electric motors is expected to be widely used as a mobility support system for elderly people and disabled people; however, there are lots of large disturbance roads such as uphill roads and rough roads and operators need to row the hand-rims with the larger power load on such roads in order to obtain the enough driving velocity. For example the wheelchair might move backward on uphill roads due to the driving torque shortage. Therefore this study proposes a fuzzy algorithm based adaptive control scheme in order to realize the assistive driving without the operator's power load on large disturbance roads. The proposed fuzzy rules are designed from the driving distance information and the control parameters are inferred by the fuzzy algorithm. The assisted torque can be adjusted so that the enough distance and velocity are kept even on large disturbance roads. Driving experimental results are provided to verify the effectiveness of the proposed control system.

Deep brain stimulation (DBS) effectively alleviates the pathological neural activity associated with Parkinson's disease. Its exact mode of action is not entirely understood. This paper explores theoretically the optimum stimulation parameters necessary to quench oscillations in a neural-mass type model with second order dynamics. This model applies well established nonlinear control system theory to DBS. The analysis here determines the minimum criteria in terms of amplitude and pulse duration of stimulation, necessary to quench the unwanted oscillations in a closed loop system, and outlines the relationship between this model and the actual physiological system.

This paper presents a novel, unsupervised spike classification algorithm for intracranial EEG. The method combines template matching and principal component analysis (PCA) for building a dynamic patient-specific codebook without a priori knowledge of the spike waveforms. The problem of misclassification due to overlapping classes is resolved by identifying similar classes in the codebook using hierarchical clustering. Cluster quality is visually assessed by projecting inter- and intra-clusters onto a 3D plot. Intracranial EEG from 5 patients was utilized to optimize the algorithm. The resulting codebook retains 82.1% of the detected spikes in non-overlapping and disjoint clusters. Initial results suggest a definite role of this method for both rapid review and quantitation of interictal spikes that could enhance both clinical treatment and research studies on epileptic patients.

This paper investigated the performance of affine optical flow (AFOF) in motion tracking of the arterial wall from B-mode ultrasound images and the effect of its combination with multiscale image analysis on the accuracy of the process. Multiscale AFOF (MAFOF) exploits the information obtained with AFOF from the approximation sub-images at different spatial resolution levels of the images, obtained using a 2D discrete wavelet transform. Both AFOF and MAFOF were evaluated through their application to synthetic image sequences of the common carotid artery. Multiscale image analysis increased the accuracy in motion tracking, with MAFOF yielding average displacement error reductions of 9% with respect to AFOF. The methods were also effectively applied to real ultrasound image sequences of the carotid artery. The results showed that MAFOF could be considered as a reliable estimator for arterial wall motion from B-mode ultrasound images.

Current standard quantitative 3D spectral-domain optical coherence tomography (SD-OCT) analyses of various ocular diseases is limited in detecting structural damage at early pathologic stages. This is mostly because only a small fraction of the 3D data is used in the current method of quantifying the structure of interest. This paper presents a novel SD-OCT data analysis technique, taking full advantage of the 3D dataset. The proposed algorithm uses machine classifier to analyze SD-OCT images after grouping adjacent pixels into super pixel in order to detect glaucomatous damage. A 3D SD-OCT image is first converted into a 2D feature map and partitioned into over a hundred super pixels. Machine classifier analysis using boosting algorithm is performed on super pixel features. One hundred and ninety-two 3D OCT images of the optic nerve head region were tested. Area under the receiver operating characteristic (AUC) was computed to evaluate the glaucoma discrimination performance of the algorithm and compare it to the commercial software output. The AUC of normal vs glaucoma suspect eyes using the proposed method was statistically significantly higher than the current method (0.855 and 0.707, respectively, p=0.031). This new method has the potential to improve early detection of glaucomatous structural damages.