Search

The paper includes illustrative and detailed information about implementation of automated diagnostic systems and feature extraction/selection from signals recorded from ophthalmic arteries. The major objective of the paper is to be a guide for the readers, who want to develop an automated diagnostic systems for detection of arterial disorders. Toward achieving this objective, this paper present the techniques which should be considered in developing automated diagnostic systems. The paper will assist to the people in gaining a better understanding of the techniques in the detection of arterial disorders.

The ballistocardiogram (BCG) signal represents the movements of the body in response to cardiac ejection of blood. Recently, many groups have developed low-cost instrumentation for facilitating BCG measurement in the home. The standard method used in the literature for estimating the BCG pulse response has generally been ensemble averaging over several beats. Unfortunately, since the BCG pulse response is likely longer than a typical heartbeat interval, this standard approach does not yield a full-length estimate of the response. This paper describes a simple, novel algorithm for estimating the full-length BCG pulse response using the R-wave timing of a simultaneously acquired electrocardiogram (ECG). With this pulse response, the full signal can be reconstructed, enabling the analysis of slow transient effects in the BCG signal, and of the measurement noise. Additionally, while this paper focuses only on the BCG signal, the same algorithm could be applied to other biomedical signals such as the phonocardiogram or impedance cardiogram, particularly when the heartbeat interval is shorter than the duration of the cpulse response.

Biological ion channels are water filled pores in the cell membrane. They regulate the flow of ions in and out of the cell. Modeling the dynamics of these channels and relating their structure to functionality is crucial in understanding the mechanisms by which they conduct. This paper proposes a novel Finite Element Method (FEM) based simulation framework for modeling of ion channels that does not assume channel symmetry. This is the first framework that allows the use of multiple dielectric constants inside such channels without assuming geometrical symmetry thus providing a more realistic model of the channel. Due to the run-time complexity of the problem, lookup tables must be constructed in memory to store pre-calculated electric potential information. The large number of elements involved in FEM and channel resolution requirements can potentially result in very large lookup tables leading to a performance "bottleneck". This paper answers the following question: Does the accuracy introduced by using an asymmetric model outweigh the inaccuracy caused by having to reduce the size and resolution of electric-field look-up tables? This paper compares the memory footprint of an ion channel simulator that assumes a symmetric channel model versus an asymmetric model. We show that currently available personal computers are sufficient for attaining reasonable levels of accuracy for both. Our results show diminishing returns in accuracy with tables sized greater than 8.5 GB for the asymmetric model.

During an emergency, determining accurate location of victims is crucial for expediting rescue efforts. In this paper, we present a mass-deployable active RFID RSSI-based real time locating system to localize victims at an emergency site. The system architecture, localization algorithms and experimental results are discussed in the paper. The system includes commercial off-the-shelf active RFID readers which are rapidly deployed at known points at the emergency site. A paramedic carrying around a bag of Active RFID tags attaches a tag to each victim as a part of the triage process. Our localization algorithm processes the unreliable RSSI values reported by the above tags using techniques such as tag calibration, tag averaging, time averaging and selective trilateration to estimate the location of the tagged victims. The system was tested in a number of experimental emergency scenarios with satisfactory results. The average error in estimated position of a tagged victim in a 100 x 100 ft emergency site setting was around 12 ft.

This paper presents an active noise control system for infant incubators. Experimental results show that global noise reduction can be achieved for infant incubator ANC systems. An audio-integration algorithm is presented to introduce a healthy audio (intrauterine) sound with the ANC system to mask the residual noise and soothe the infant. Carbon nanotube based transparent thin film speaker is also introduced in this paper as the actuator for the ANC system to generate the destructive secondary sound, which can significantly save the congested incubator space and without blocking the view of doctors and nurses.

This paper presents a soft tissue target stabilization method during needle insertion procedures. The object considered in this study may have fixed boundary sections and limited surface exposed to external manipulation. The target must be stabilized along the needle path during the needle insertion. It is assumed that a paddle with fixed geometry is available for deformable object manipulation. Two approaches were considered for the target stabilization problem. The first approach uses a static paddle placed on the available boundary, at a strategic location, such that the target motion orthogonal to the needle axis is minimized during the needle insertion. The second approach uses a dynamic paddle attached to the available boundary for the active compensation of the target deflection. In this paper we analyze the optimal paddle placement for the two proposed approaches and present initial numerical results for the case of homogeneous and nonhomogeneous deformable objects. The results show that the first approach is sensitive to possible non-homogeneities in the object, therefore it is not robust to modeling errors. The results also show that optimal placement for the second approach is less sensitive to modeling errors, making it more desirable for physical applications.

This paper describes an indentometer device used to identify the linear dynamic and nonlinear properties of skin and underlying tissue using an in vivo test. The device uses a Lorentz force actuator to apply a dynamic force to the skin and measures the resulting displacement. It was found that the skin could be modeled as a Wiener system (i.e. a linear dynamic system followed by a static nonlinearity). Using a stochastic nonlinear system identification technique, the method presented in this paper was able to identify the dynamic linear and static nonlinear mechanical parameters of the indentometer-skin system within 2 to 4 seconds. The shape of the nonlinearity was found to vary depending on the area of the skin that was tested. We show that the device can repeatably distinguish between different areas of human tissue for multiple test subjects.

This paper first describes the AM-FM demodulation of an arterial pressure signal. Although it is known to be efficient on signals modulated by breathing, we demonstrate that in case of lack of respiratory modulation (real or simulated central sleep apnea), the AM-FM algorithm doesn’t perform well in heart rate extraction. We introduce then a new algorithm based on Singular Spectrum Analysis eigen values which performs better cardiac frequency estimation in this context. Respiratory estimation is possible but is beyond the scope of this paper. The error for cardiac frequency estimation is around 0.2 BPM (Beats Per Minute) versus 5.5 BPM for the AM-FM demodulation. Further experimentations will be performed (with this time both cardiac and respiratory assessments) and will deal with real sleep apnea cases.

Electrogastrographic Signal (EGG) is considered to be one of the less interesting from both registration and interpretation point of view. There are several reasons of that two facts. EGG presents gastric myoelectrical activity measured by several electrodes attached on the abdomen. Unfortunately the registration procedure does not deliver a pure signal as EGG is usually associated with some interferences caused by the other organs localized near stomach. On the other hand however there are no databases available, which could allow both comparison and proper interpretation. One of the parameter, among others, which is analyzed owing to proper registration is so called normogastric rhythm, which should cover around 70% of rhythmic behavior of the signal. Proper extraction of the normogastric rhythm is a subject of this paper. Special signal preprocessing steps should be applied before the main tool i.e. Independent Component Analysis (ICA) is applied for normogastric rhythm extraction. Also, to make this analysis possible a special registration procedure has been applied concerning two phases of registration - one with feeding and the other one without with 5 minutes brake between them.

Implantable medical devices will play an important role in modern medicine. To reduce the risk of wire snapping, and replacement and corrosion of embedded batteries, wireless delivery of energy to these devices is desirable. However, current autonomous implants remain large in scale due to the operation at very low frequency and the use of unwieldy size of antennas. This paper will show that the optimal frequency is about 2 orders of magnitude higher than the conventional wisdom; and thereby the power receiving coils can be reduced by more than 100 fold without sacrificing either power efficiency or range. We will show that a mm-sized implant can receive 100's microW of power under safety constraints. This level of power transfer is sufficient to enable many functionalities into the micro-implants for clinical applications.

Respiratory CO2 measurement (capnography) is an important diagnosis tool that lacks inexpensive and wearable sensors. This paper develops techniques to enable use of inexpensive but slow CO2 sensors for breath-by-breath tracking of CO2 concentration. This is achieved by mathematically modeling the dynamic response and using model-inversion techniques to predict input CO2 concentration from the slowly varying output. Experiments are designed to identify model-dynamics and extract relevant model-parameters for a solid-state room monitoring CO2 sensor. A second-order model that accounts for flow through the sensor's filter and casing is found to be accurate in describing the sensor's slow response. The corresponding model-inversion algorithm is however found to be susceptible to noise sources. Techniques to remove spurious noise, while retaining quality of estimate are developed. The resulting estimate is compared with a standard-of-care respiratory CO2 analyzer and shown to effectively track variation in breath-by-breath CO2 concentration. This methodology is potentially useful for measuring fast-varying inputs to any slow sensor.

A patient-specific model-based seizure detection method using statistically optimal null filters (SONF) has been recently proposed to aid the review of long-term EEG [1, 2]. The method relies on the model of a priori known seizure (template pattern) for subsequent detection of similar seizures. Artifacts, non-epileptic EEG rhythms, and at times modeling errors lead to increased false or missed detections. In this paper, we present a new improved model-based seizure detection that introduces a pre-processing block for artifact rejection, an adaptive technique of modeling the template patterns, and a new evolution-based classifier. The proposed classifier tracks the temporal evolution of seizure to improve the classification accuracy. With the help of simulated EEG, we illustrate the significance and need for these modifications. Further, performance of the complete algorithm is tested on single channel depth EEG of seven patients, and compared with the previous approaches. In terms of sensitivity and specificity, the proposed method resulted in 84% and 100%, method of [1] 65% and 84%, and method of [2], 84% and 90% respectively. An overall performance improvement is seen as enhanced detection sensitivity and reduced false positives. This is preliminary result on seven patient data.

Numerous methods have been reported for deriving respiratory information such as respiratory rate from the electrocardiogram (ECG). In this paper the authors present a real-time algorithm for estimation and removal of baseline wander (BW) noise and obtaining the ECG-derived respiration (EDR) signal for estimation of a patient's respiratory rate. This algorithm utilizes a real-time "T-P knot" baseline wander removal technique which is based on the repetitive backward subtraction of the estimated baseline from the ECG signal. The estimated baseline is interpolated from the ECG signal at midpoints between each detected R-wave. As each segment of the estimated baseline signal is subtracted from the ECG, a "flattened" ECG signal is produced for which the amplitude of each R-wave is analyzed. The respiration signal is estimated from the amplitude modulation of R-waves caused by breathing. Testing of the algorithm was conducted in a pseudo real-time environment using MATLAB(TM), and test results are presented for simultaneously recorded ECG and respiration recordings from the PhysioNet/PhysioBank Fantasia database. Test data from patients were chosen with particularly large baseline wander components to ensure the reliability of the algorithm under adverse ECG recording conditions. The algorithm yielded EDR signals with a respiration rate of 4.4 breaths/min. for Fantasia patient record f2y10 and 10.1 breaths/min. for Fantasia patient record f2y06. These were in good agreement with the simultaneously recorded respiration data provided in the Fantasia database thus confirming the efficacy of the algorithm.

In this paper we introduce a noise-resilient edge detection algorithm for brain MRI images. Also, an improved edge detection based on Canny edge detection algorithm is proposed. Computer simulations show that the proposed algorithm is resilient to impulsive noise which makes up for the disadvantages of Canny algorithm, and can detect more edges of MRI brain images effectively. Also, the concept of images fusion is utilized for effective edge detection.

Arterial elasticity has become a topic of importance in the past decades, as it has shown that it can be used to predict cardiovascular diseases and mortality. Several in vivo and ex vivo techniques have been developed to characterize the mechanical properties of vessels. In vivo techniques tend to ignore the anisotropicity of the vessel wall components. While ex vivo techniques tend to be destructive and do not to account for the geometry of the arteries. In this paper we present a technique using sonometry to study the elasticity of soft tubes and excised pig carotids in different directions. The method uses piezoelectric crystals to track the strain in the circumferential and longitudinal directions while the tubes or vessels are being pressurized. We compare the Young's moduli obtained from sonometry experiments performed in two different types of tubes with the mechanical testing done in the material used to make these tubes. We also present data obtained in the excised pig carotids and show the differences in the longitudinal versus the circumferential directions. The technique we propose has a potential for the non destructive study of soft material cylindrical shapes and can be use to study the mechanical properties of vessels.

This paper presents advanced signal processing algorithms to quantify brain connectivity with high temporal resolution using the electroencephalogram (EEG). The experimental paradigm exploits the visual cortex response to flickering images at a given frequency. The envelope of the EEG at the flickering frequency collected by 128 electrodes over the head quantifies the communication amongst the brain areas with high temporal resolution. This work proposes to use the empirical mode decomposition to find the flickering frequency and then use the Hilbert transform to estimate the instantaneous amplitude. A video of the topographical display of the instantaneous power over the array helps us visualize the exquisite communication that occurs during the stimulus presentation.

Pathological myopia, the seventh leading cause of legal blindness in United States, is a condition caused by pathological axial elongation and eyes that deviates from the normal distribution curve of axial length, resulting in impaired vision. Studies have shown that ocular risks associated with myopia should not be underestimated, and there is a public health need to prevent the onset or progression of myopia. Peripapillary atrophy (PPA) is one of the clinical indicators for pathological myopia. In this paper, we introduce a novel method, to detect pathological myopia via peripapaillary atrophy feature by means of variational level set. This method is a core algorithm of our system, PAMELA, an automated system for the detection of pathological myopia. The proposed method has been tested on 40 images from Singapore Cohort study Of the Risk factors for Myopia (SCORM), producing a 95% accuracy of correct assessment, and a sensitivity and specificity of 0.9 and 1 respectively. The results highlight the potential of PAMELA as a possible clinical tool for objective mass screening of pathological myopia.

An original algorithm for measuring diffusion coefficients of optical molecular probes in matrigel from fluorescence data is introduced. The algorithm was developed in Fortran and linked to Graphic User Interface in LabVIEW software that also performs image acquisition and processing. The software models pharmacokinetics of optical molecular probes providing the best fit of experimental data. The paper offers an original way for estimating the diffusion path length through extracellular matrix (ECM) from the rate constants given by the model and from measured diffusion coefficients.

Needles with tip asymmetry deflect upon insertion into soft tissue, an effect that can be used to steer needles within the body. This paper presents a phenomenological characterization of the steering behavior of pre-curved needles, which have tip asymmetry due to curvature of the needle near the tip. We describe needle construction methods and a needle shaft triangulation algorithm to compute the shape of the needle based on images. Experimental results show that pre-curved needles possess greater dexterity than bevel-tipped needles and achieve radii of curvature similar to pre-bent needles. For long pre-curve arc lengths, the radius of curvature of the needle was found to approach the radius of curvature of the pre-curve. Pre-curved needles were found to display behaviors not seen with bevel-tipped needles, such as the insertion velocity influencing the path of the tip within the tissue and the ability to plastically deform the needle during steering.

Image encryption is an effective approach for providing security and privacy protection for medical images. This paper introduces a new lossless approach, called EdgeCrypt, to encrypt medical images using the information contained within an edge map. The algorithm can fully protect the selected objects/regions within medical images or the entire medical images. It can also encrypt other types of images such as grayscale images or color images. The algorithm can be used for privacy protection in the real-time medical applications such as wireless medical networking and mobile medical services.

Motor imagery is a popular paradigm in electrophysiology research and brain computer interface but the evoked EEG signals always contaminated significantly. In this paper we use the Independent Component Analysis to enhance the signal-to-noise ratio of multi trail EEG signals evoked by imaginary hand movement. Infomax algorithm was used to decompose multi channel EEG signals into independent components trail by trail, and then an automatic clustering method was used to group these components into several clusters. For the higher similarity between task relevant components, they can be assembled into one cluster that occupies the highest mean mutual information of pairwise components intra cluster. Furthermore, the reconstructed signals of task relevant cluster showed a high discrepancy features to left versus right hand task, which evaluated by Fisher criterion scores and served as the signal-to-noise ratio measurement.

Ambient Assisted Living (AAL) technology is often proposed as a way to tackle the increasing cost of healthcare caused by population aging. However, the sensing technology for continuous respiratory monitoring at home is lacking. Known approaches of respiratory monitoring are based on measuring either respiratory effect, e.g. tracheal sound recording by a bio-acoustic sensor, or respiratory effort, e.g. abdomen movement measurement by a tri-axial accelerometer. This paper proposes a home respiration monitoring system using a tri-axial accelerometer. Three different methods to extract a single respiratory signal from the tri-axial data are proposed and analyzed. The performance of the methods is evaluated for various possible respiration conditions, defined by the sensor orientation and respiration-induced abdomen movement. The method based on Principal Component Analysis (PCA) performs better than selecting the best axis. The analytical approach called Full Angle shows worse results than the best axis when the gravity vector is close to one of the sensor’s axes (

MR Urography (MRU) plays an important role in pediatric renal diagnoses. While accurate measurement of the kidney is essential for each imaging study, it is challenging to extract the kidney from its background. In this paper, we propose a co-focus elliptical kidney model (CEKM) to integrate the shape prior and build a novel extraction method with both regional and edge information. With a similarity metric defined between a binary mask and CEKM, a kidney template can be obtained using optimization technique. Then, a shape term derived from a distance and orientation description is designed to modify an active contour model. The distance map is used to control the contours in terms of CEKM and orientation map is designed to reduce the artifacts resulting from fake edges. The final kidney was determined with iterative solution. With the priori shape description in a parametric space, the new method can precisely extract the kidney without training, and supply promising results as the experiments demonstrated.

Conventional adaptive active noise control (ANC) methods aim to attenuate the acoustic noise over the frequency band of interest indiscriminately using the sound pressure level (SPL) measurement (or the measurement of the residual error variance). However, SPL does not correctly reflect the human perception of attenuated noise due to the frequency selective sensitivity of human hearing system. A-weighting is a commonly used weighting filter for measuring the noise. This weighting filter quantifies frequency response of the human ears and hearing system. In this paper, we aim to improve the performance of adaptive noise cancellation method from the psychoacoustic point of view by incorporating the A-weighting into the ANC system design. Loudness is used as the psychoacoustic criterion for evaluating the ANC system performance. Simulation results illustrate the effectiveness of the proposed method.

This paper presents an evaluation of the sensitivity of a new thin-film stiffness sensing technology that utilizes commercial electret microphones. The analysis allows comparison of commercial microphones for stiffness sensing applications. A mathematical method to estimate the stiffness sensitivity of a commercial microphone from its acoustic sensitivity is developed. Experimental results are presented on the use of the developed method in a sensor that estimates carbon dioxide concentration by utilizing a carbon nanotube thin film on an electret microphone.

In this paper we reported a novel method to detect and quantify dural ectasia in Marfan syndrome. Firstly, the dural sacs of 8 Marfan patients were segmented by applying an unsupervised Fuzzy C-Means method on T2-weighed magnetic resonance images. Then, for each patient a tubular model of the dural sac was extracted by detecting and removing the existent pathological extrusions. The segmented images together with the resulting tube were then rendered using a marching cubes algorithm. The proposed algorithm represents a first attempt to quantify and to morphologically characterize the pathological ectasia that usually accompanies the Marfan disorder. The generated 3D reconstruction and the opportunity to overlap them with a physiological model provides the clinician with a tool for a panoramic view of the structures and a means for a more accurate inspection of ectasia. In addition the extracted parameters furnish quantitative and reproducible measures that could be useful as discriminative indexes for an automatic and more objective diagnosis.

This paper examines whether an appropriate algorithm, developed for use with neonatal data, could also be used, without alteration, for the detection of seizures in adults with epilepsy. The performance of a feature extraction and SVM classifier system is evaluated on databases of 17 neonatal patients and 15 adult patients. Mean ROC curve areas of 0.96 and 0.94 for neonatal and adult databases respectively show that high accuracy can be achieved independent of age. It is also shown that features contribute differently for neonatal and adult data.

This paper discusses the needs for automated tools to aid in the diagnosis of thyroid nodules based on analysis of fine needle aspiration cytology smears. While conventional practices rely on the analysis of grey scale or RGB color images, we present a multispectral microscopy system that uses thirty-one spectral bands for analysis. Discussed are methods and results for system calibration and cell delineation.

An automatic diagnosis system of nuclear cataract is presented in this paper. Nuclear cataract is graded according to the severity of opacity using slit-lamp lens images. Anatomical structure in the lens image is detected using a modified active shape model (ASM). Based on the anatomical landmark, local features are extracted according to clinical grading protocol. Support vector machine (SVM) regression is employed to train a grading model for grade prediction. The system is tested using clinical images and clinical ground truth. More than five thousands slit-lamp images were tested. The success rate of feature extraction is 95% and the mean grading difference is 0.36. The automatic diagnosis system can help to improve the grading objectivity and save the workload of ophthalmologists.

This paper addresses the issue of heart rate detection from noisy ECG data, and presents a method with low complexity and low memory requirements that can detect QRS complex in the presence of noise and muscle artifacts. On the MIT-BIH arrhythmia database we were able to detect 99.3% of QRS complexes with 0.47% false detection. This method can also be applied to heart rate detection using phonocardio signals.

This paper presents a new method for nonlinear trend estimation of non-stationary signals, by which the trend can be self-adaptively decomposed through calculating the midpoint-based local means. In this method, the so-called midpoints are proposed to construct the local mean of a signal instead of two envelopes in the classical empirical mode decomposition (EMD) algorithm, thus resulting in the midpoint-based empirical decomposition. Furthermore, a negentropy-based statistical method is presented to justify decomposition of the trend. Simulation results indicate that the new algorithm improves the performance of signal decomposition and trend estimation in comparison with the classical EMD algorithm. The proposed method also shows the value in self-adaptively estimating the nonlinear respiratory component from non-invasively measured ventilation signals.

International audience; The morphological studies of neuron structures are of great interests for biologists. However, manually detecting dendrites structures is very labor intensive, therefore unfeasible in studies that involve a large number of images. In this paper, we propose an automated neuron detection and description method. The proposed method uses ratios of probability maps from random walk sessions to detect initial seed-points and minimal cost path integrals with Delaunay triangulations.

Texture analysis of B-mode ultrasound images of carotid atheromatous plaque can be valuable for the accurate diagnosis of atherosclerosis. In this paper, Gabor filters were used to characterize the texture of carotid artery atherosclerotic tissue. B-mode ultrasound images of 10 symptomatic and 9 asymptomatic plaques were interrogated. A total of 40 texture features were estimated for each plaque. The bootstrap method was used to compare the mean values of the texture features extracted from the two groups. After bootstrapping, the mean value and the standard deviation of the energy estimated using the Gabor filters was found to be significantly different between symptomatic and asymptomatic plaques in the first scale of analysis and for all orientations. In addition, a number of texture features that correspond to larger resolution scales were found to be significantly different between the two types of plaques. It is concluded that Gabor-filter-based texture analysis in combination with a powerful statistical technique, such as bootstrapping, may provide valuable information about the plaque tissue type.

Bayesian methods have been widely applied to the ill-posed problem of image reconstruction. Typically the prior information of the objective image is needed to produce reasonable reconstructions. In this paper, we propose a novel generalized Gibbs prior (GG-Prior), which exploits the basic affinity structure information in an image. The motivation for using the GG-Prior is that it has been shown to suppress noise effectively while capturing sharp edges without oscillations. This feature makes it particularly attractive for those applications of Positron Emission Tomographic (PET) where the objective is to identify the shape of objects (e.g.tumors) that are distinguished from the background by sharp edges. We show that the standard paraboloidal surrogate coordinate ascent (PSCA) algorithm can be modified to incorporate the GG-Prior using a local linearized scheme in each iteration process. The proposed GG-Prior MAP reconstruction algorithm based on PSCA algorithm has been tested on simulated, real phantom data. Comparisons the GG-Prior model with other existing prior model clearly demonstrate that the proposed GG-Prior performs better in lowering the noise, and preserving the edge and detail in the image.

Summarization: The metabolic behavior of complex brain tumors, like gliomas and meningiomas, with respect to their type and grade was investigated in this paper. Towards this direction the smallest set of the most representative metabolic markers for each brain tumor type was identified, using ratios of peak areas of well established metabolites, from 1H-MRSI (proton magnetic resonance spectroscopy imaging) data of 24 patients and 4 healthy volunteers. A feature selection method that embeds Fisher's filter criterion into a wrapper selection scheme was applied; support vector machine (SVM) and least squares-SVM (LS-SVM) classifiers were used to evaluate the ratio markers classification significance. The area under the receiver operating characteristic curve (AUROC) was adopted to evaluate the classification significance. It is found that the NAA/CHO, CHO/S, MI/S ratios can be used to discriminate gliomas and meningiomas from Healthy tissue with AUROC greater than 0.98. Ratios CHO/S, CRE/S, MI/S, LAC/CRE, ALA/CRE, ALA/S and LIPS/CRE can identify type and grade differences in gliomas giving AUROC greater than 0.98 apart from the scheme of gliomas grade II vs grade III where 0.84 was recorded due to high heterogeneity. Finally NAA/CRE, NAA/S, CHO/S, MI/S and ALA/S manage to discriminate gliomas from meningiomas providing AUROC exceeding 0.90. Παρουσιάστηκε στο: Annual International Conference of the IEEEE Engineering in Medicine and Biology Society, EMBC

This paper proposed a low-complexity medium access control (MAC) protocol tailored for body sensor networks (BSN) applications. The MAC protocol was designated to handle collision avoidance by reducing the numbers of the overhead packets for handshake control within the BSN. We also suggested a novel message recovery mechanism for getting back the lost physiological information. The adaptive synchronization scheme we have implemented exploited the features of multiple data-rate and adjustable precision design to support differentiated healthcare applications. The MAC protocol was fully implemented using our BSN development platform. The experimental results suggested the improved MAC design was compact and energy-efficient.

In this paper, we propose an automated method for colon registration from supine and prone scans. Four anatomical salient points on the colon are distinguished first. Then correlation optimized warping (COW) method is applied to the segments defined by the anatomical landmarks to find better global registration based on local correlation of segments. To utilize more features along the colon centerline, we extended the COW method by embedding canonical correlation analysis into it for correlation calculation of colon segments. To verify the effectiveness of the proposed method, we tested the algorithm on a CTC dataset of 19 patients with 23 polyps. Experimental results show that by using our method, the estimation error of polyp location could be reduced 68.5% (from 41.6mm to 13.1mm on average) compared to a traditional dynamic warping algorithm.

Limb state feedback is of great importance for achieving stable and adaptive control of FES neuroprostheses. A natural way to determine limb state is to measure and decode the activity of primary afferent neurons in the limb. The feasibility of doing so has been demonstrated by [1] and [2]. Despite positive results, some drawbacks in these works are associated with the application of reverse regression techniques for decoding the afferent neuronal signals. Decoding methods that are based on direct regression are now favored over reverse regression for decoding neural responses in higher regions in the central nervous system [3]. In this paper, we apply a direct regression approach to decode the movement of the hind limb of a cat from a population of primary afferent neurons. We show that this approach is more principled, more efficient, and more generalizable than reverse regression.

Mental health management is fast becoming a major challenge worldwide as the incidence of mental illness has been increasing. It is affecting the quality of life as well as job productivity for a large number of people. Just like physical illnesses, people with mental illnesses can be monitored for a range of conditions and provided medical care as and when necessary. In this paper, we present an IT-enabled framework to support mental health monitoring. This includes comprehensive monitoring of patients for symptoms, behavior, and medication compliance. We utilize context-awareness as a way to develop a system for mental health monitoring. Several examples of future mental health monitoring are also presented.

Conventional endoscope-guided photodynamic therapy (PDT) suffers mostly from motion artifacts, therefore expert hand-eye coordination was always needed during manual operations. In this paper we introduced a visual servo scheme to handle the tracking problem between the focused area and the targeted lesions. The scheme is consisted of real-time feature matching, relative motion cancellation and real-time light dose surveillance. Experiments were carried out both on simulated data and a silicon phantom. It indicates that this scheme outperforms the conventional scheme in terms of reduction in operation time and exposure to healthy tissue.

Estimating patient-specific electrical tissue parameters is of considerable benefit towards personalization of cardiac biophysical models. In this paper, an adaptive inverse parameter estimation algorithm is proposed to estimate the myocardial apparent conductivity from endocardial electrical potential measurements. The forward electrophysiology problem is posed as an Eikonal model and is solved using an anisotropic fast marching method. The conductivity estimation algorithm is validated on patient data obtained using hybrid X-ray/magnetic resonance imaging. Future directions towards improving such estimation procedures are also indicated.

Understanding transport and deposition of inhaled particles in the human airways plays a crucial role in the targeted therapy of pulmonary diseases, and the administration of inhaled medicines. Numerous researchers have studied the inhalation of particles using experiments or computer models. Even though experiments have shown that the airflow in the trachea and the upper branches of the lung is turbulent, the flow is taken to be laminar in most computer models. Only few recently published papers have looked at the turbulent transport of air in the human airways. Even fewer results have been published on the effect of the upper airway structures on the turbulent airflow in the lungs or on the effect of the turbulence on particle deposition. The previously published turbulent models have also mainly used RANS methods to predict the flow. To study the unsteady flow and particle deposition in a human lung, an LES model with a dynamic Smagorinsky subgrid scale model was used. The model equations were solved to study steady inspirational flow at different flow rates for different particle sizes. Results indicate that the upper airway geometry produces turbulence in the flow and the deposition of particles is mainly affected by the particle size and Stokes number. ©2009 IEEE. 2867 2870 2867-2870

This paper presents a novel method that uses eyelid closure and heart rate variability to estimate the driver's drowsiness level. Laboratory experiments were conducted by using a proprietary driving simulator, which induced drowsiness among the test drivers. The purposes of these experiments were to obtain the electrocardiogram (ECG) and the eye-blink video sequences. Also the drivers were monitored through a video camera. The changes in facial expression of the drivers were used as a standard index of drowsiness level. Error-Correcting Output Coding (ECOC) was employed as a multi-class classifier to estimate the drowsiness level. We extended the ordinary ECOC using a loss function for decoding procedure to obtain class tendencies of each drowsiness level. We used the Loss-based Decoding ECOC (LD-ECOC) to classify the drowsiness level. As a result, we obtained an extraordinarily high accuracy for estimation of drowsiness level.

Obstructive sleep apnea (OSA) is a common respiratory disorder during sleep, in which the airways are collapsed and impair the respiration. Apnea is s cessation of airflow to the lungs which lasts at least for 10s. The current gold standard method for OSA assessment is full night polysomnography (PSG); however, its high cost, inconvenience for patients and immobility have persuaded researchers to seek simple and portable devices to detect OSA. In this paper, we report on developing a new system for OSA detection and monitoring, which only requires two data channels: tracheal breathing sounds and the blood oxygen saturation level (S(a)O(2)). A fully automated method was developed that uses the energy of breathing sounds signals to segment the signals into sound and silent segments. Then, the sound segments are classified into breath, snore (if exists) and noise segments. The S(a)O(2) signal is analyzed to find the rises and drops in the S(a)O(2) signal. Finally, a fuzzy algorithm was developed to use this information and detect apnea and hypopnea events. The method was evaluated on the data of 40 patients simultaneously with full night PSG study, and the results were compared with those of the PSG. The results show high correlation (96%) between our system and PSG. Also, the method has been found to have sensitivity and specificity values of more than 90% in differentiating simple snorers from OSA patients.

In this paper we present methods to track and characterize human dynamic skills using motion capture and electromographic sensing. These methods are based on task-space control to obtain the joint kinematics and extract the key physiological parameters and on computed muscle control to solve the muscle force distribution problem. We also present a dynamic control and analysis framework that integrates these metrics for the purpose of reconstructing and analyzing sports motions in real-time.

This paper presents the results of a study into the subjective effect of speckle reduction filtering in echocardiography, as assessed by clinical experts. Echocardiographic videos were filtered by a number of speckle reduction methods to produce a set of test videos with varying levels of speckle content. Six practicing cardiac technicians were asked to rate each video in three scoring categories, with the aim of quantifying their subjective evaluation of the diagnostic usefulness and the level of speckle in each video. The change in expert scores due to filtering is analyzed. Inter-expert difference in the evaluation is investigated, and intra-expert analysis of the association between each score category is also performed. In addition, a number of objective quality metrics are applied to the filtered videos, and the correlation between these metrics and the expert scores is determined. Results indicate that, while there are inter-expert differences, strong intra-expert relationships exist between the score categories. Furthermore each of the three subjective score categories is strongly associated with one of the objective quality metrics.

Neural activity is very important source for data mining and can be used as a control signal for brain-computer interfaces (BCIs). Particularly, Magnetic signals of neurons are enriched with information about the movement of different part of the body such as wrist movement. In this paper, we use MEG (Magneto encephalography) signals of two subjects recorded during wrist movement task in four directions. Data were prepared for BCI competition 2008 for multiclass classification. Our approach for this classification problem consists of PCA as a noise reduction method, ULDA for feature reduction and various linear classifiers such as Bayesian, KNN and SVM. Final results (58%-62% for subject 1 and 36%-40% for subject 2) prove that the suggested method shows better performance compared with other methods.

Nerve cell segmentation is important for better understanding of the connections between broken nerves. In this paper, we propose a registration-based segmentation to improve the segmentation from a series of microscopic image sequence. A global iterative closest point (ICP) registration is first employed to find the corresponding cells between neighboring frames. Then a local ICP registration is adopted to refine shape matching between corresponding cells. When registering the current frame to the previous frame, a missing cell in the current frame can obtain a dummy region with respect to the previous frame. The dummy region is then confirmed as a cell by checking the correspondence of the cell among the previous, current and next frames. Experimental results show that the proposed registration based algorithm recovers the cells that are missed in conventional segmentation.

A wide variety of methods based on fractal, entropic or chaotic approaches have been applied to the analysis of complex physiological time series. In this paper, we show that fractal and entropy measures are poor indicators of nonlinearity for gait data and heart rate variability data. In contrast, the noise titration method based on Volterra autoregressive modeling represents the most reliable currently available method for testing nonlinear determinism and chaotic dynamics in the presence of measurement noise and dynamic noise.

This paper presents an integrated speech enhancement (SE) method for the noisy MRI environment. We show that the performance of SE system improves considerably when the speech signal dominated by MRI acoustic noise at very low SNR is enhanced in two successive stages using two-channel SE methods followed by a single-channel post processing SE algorithm. Actual MRI noisy speech data are used in our experiments showing the improved performance of the proposed SE method.