Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference最新文献

A truly wearable non-invasive blood pressure (NIBP) sensor--light-weight, compact, unobstrusive, and essentially unnoticeable to the patient--could revolutionize healthcare delivered beyond the traditional walls of medical facilities, offering new ways to care for patients in their everyday surroundings. This paper presents results from our work towards the development of a self-contained, wearable blood pressure sensor. A PPG-based approach to blood pressure monitoring is presented. The design enables significant miniaturization of traditional oscillometric devices without the need for occlusive circumferential pressures. It will be shown how natural raising and lowering of the arm replaces the need for bulky actuators. Additionally, a dual-accelerometer height sensor that is tetherless is proposed and supported by experimental results.

We reported previously that the skills transferred from practicing using a bimanual grasp to skills in right hand are small but significant. In this study on healthy right-handed people we compared how well skills learned while training using a bimanual grasp transferred to the left and right hands performance individually. As before, the task was to make target-directed reaching movements while grasping a planar robotic device that systematically disturbed movements at the handle. Results showed that skills learned while practicing with a bimanual grasp generalized (transferred) to both the dominant and the non-dominant arm equally well, with the right limb benefiting the most because it began with less error.

Based on several years of experiments, we propose a model of information systems for residential healthcare, and technical guide to select available hard and software technologies. An implementation is described, based on Emails. The system is under experimentation within the framework of the French national project AILISA.

In this paper, we present several fuzzy inference systems for monitoring patient status in an operating room. The algorithms used include recursive fuzzy inference (RFIS), and non-recursive with sequential patterns as inputs. The RFIS algorithm combines current patient status data with previous output of the inference system, therefore is able to reinforce the current finding based on previous sequential system output. The results show that the RFIS system can be tuned towards higher sensitivity for more critical status, while generating smoother inference output.

In this paper, we discuss a design for a wearable electrocardiograph device constructed with small, low-powered "mote" sensors for use in wireless sensor networks. A wearable, wireless three-lead electrocardiograph sensor module is utilized and the initial tests are presented that illustrate the viability of this design. This device can be integrated into a suite of wearable wireless sensors used for patient monitoring and other applications.

Maintaining intracranial pressure (ICP) below 20-25 mmHg is an important clinical goal in the treatment of patients with traumatic brain injury (TBI). It is well known that the partial pressure of arterial CO2 (PaCO2) can affect cerebral blood flow, cerebral blood volume, and therefore ICP. The end-tidal CO2 (ETCO2) is usually monitored by clinicians as a proxy for PaCO2. We show examples where subclinical fluctuations in ETCO2 are associated with clinically significant fluctuations in ICR. We estimated ICP from past and present values of ETCO2 with a linear estimator. The variance of the ICP residuals was 37 percent of the variance of the ICP signal at frequencies above 0.33 mHz. We suggest that a large proportion of clinically significant ICP fluctuations could be eliminated or reduced if the patients ventilation and CO2 levels were more tightly regulated.

Prolonged electrical stimulation can lead to temporary or permanent changes in neural response. Stimulation of neurons at levels sufficient to cause overlapping zones of excitation can induce multiple effects, leading to permanent damage to neurons or temporary depression not detectable through histopathological analysis. The present study focuses on determining the effects of prolonged, continuous electrical stimulation in the retina. One hour stimulation was performed in the rat retina and electrically evoked responses in the superior colliculus were recorded before and after the continuous stimulation. Comparison of the pre and post stimulation responses indicates a depression in the excitability of the neurons.

Mutual adaptation between man and machine is necessary for the development of more efficient devices that allows easy adaptation. The interaction with intelligent machines involves adaptation processes from both the user and the machine. The human body has the ability to change its body schema to include external tools in it, using this fact we proposed the design of intelligent machines with biofeedback to the user, permitting in this way, development of subconscious control of external devices. We propose the case of an EMG prosthetic hand with biofeedback to study the adaptation process between man-machine. Our system includes an EMG classification system to acquire the intended movement from the user. We use electrical stimulation as a provider of tactile feedback, to interact with the human body. We use a functional Magnetic Resonance Image study while using the prosthetic device while receiving biofeedback to measure the activation levels in the amputee's brain.

Despite steady progress in the miniaturization of pulse oximeters over the years, significant challenges remain since advanced signal processing must be implemented efficiently in real-time by a relatively small size wearable device. The goal of this study was to investigate several potential digital signal processing algorithms for computing arterial oxygen saturation (SpO(2)) and heart rate (HR) in a battery-operated wearable reflectance pulse oximeter that is being developed in our laboratory for use by medics and first responders in the field. We found that a differential measurement approach, combined with a low-pass filter (LPF), yielded the most suitable signal processing technique for estimating SpO(2), while a signal derivative approach produced the most accurate HR measurements.

The linear regression model has been widely used in the analysis of gene expression and microarray data to identify a subset of genes that are important to a given metabolic function. One of the key challenges in applying the linear regression model to gene expression data analysis arises from the sparse data problem, in which the number of genes is significantly larger than the number of conditions. To resolve this problem, we present a knowledge driven regression model that incorporates the knowledge of genes from the Gene Ontology (GO) database into the linear regression model. It is based on the assumption that two genes are likely to be assigned similar weights when they share similar sets of GO codes. Empirical studies show that the proposed knowledge driven regression model is effective in reducing the regression errors, and furthermore effective in identifying genes that are relevant to a given metabolite.