Numerical study of reflectance imaging using a parallel Monte Carlo method

Reflectance imaging of biological tissues with visible and near-infrared light has the significant potential to provide a noninvasive and safe imaging modality for diagnosis of dysplastic and malignant lesions in the superficial tissue layers. The difficulty in the extraction of optical and structural parameters lies in the lack of efficient methods for accurate modeling of light scattering in biological tissues of turbid nature. We present a parallel Monte Carlo method for accurate and efficient modeling of reflectance images from turbid tissue phantoms. A parallel Monte Carlo code has been developed with the message passing interface and evaluated on a computing cluster with 16 processing elements. The code was validated against the solutions of the radiative transfer equation on the bidirectional reflection and transmission functions. With this code we investigated numerically the dependence of reflectance image on the imaging system and phantom parameters. The contrasts of reflectance images were found to be nearly independent of the numerical aperture (NA) of the imaging camera despite the fact that reflectance depends on the NA. This enables efficient simulations of the reflectance images using an NA at 1.00. Using heterogeneous tissue phantoms with an embedded region simulating a lesion, we investigated the correlation between the reflectance image profile or contrast and the phantom parameters. It has been shown that the image contrast approaches 0 when the single-scattering albedos of the two regions in the heterogeneous phantoms become matched. Furthermore, a zone of detection has been demonstrated for determination of the thickness of the embedded region and optical parameters from the reflectance image profile and contrast. Therefore, the utility of the reflectance imaging method with visible and near-infrared light has been firmly established. We conclude from these results that the optical parameters of the embedded region can be determined inversely from reflectance images acquired with full-field illumination at multiple incident angles or multiple wavelengths.     

A decentralized adaptive fuzzy robust strategy for control of upright standing posture in paraplegia using functional electrical stimulation

In this paper, we present a novel decentralized robust methodology for control of quiet upright posture during arm-free paraplegic standing using functional electrical stimulation (FES). Each muscle–joint complex is considered as a subsystem and individual controllers are designed for each one. Each controller operates solely on its associated subsystem, with no exchange of information between them, and the interaction between the subsystems are taken as external disturbances. In order to achieve robustness with respect to external disturbances, unmodeled dynamics, model uncertainty and time-varying properties of muscle–joint dynamics, a robust control framework is proposed. The method is based on the synergistic combination of an adaptive nonlinear compensator with sliding mode control (SMC). Fuzzy logic system is used to represent unknown system dynamics for implementing SMC and an adaptive updating law is designed for online estimating the system parameters such that the global stability and asymptotic convergence to zero of tracking errors is guaranteed. The proposed controller requires no prior knowledge about the dynamics of system to be controlled and no offline learning phase. The results of experiments on three paraplegic subjects show that the proposed control strategy is able to maintain the vertical standing posture using only FES control of ankle dorsiflexion and plantarflexion without using upper limbs for support and to compensate the effect of external disturbances and muscle fatigue.     

Initial in vivo results of a hybrid retinal photocoagulation system

We describe initial in vivo experimental results of a new hybrid digital and analog design for retinal tracking and laser beam control. An overview of the design is given. The results show in vivo tracking rates which exceed the equivalent of 38 degrees/s in the eye. A robotically assisted lesion pattern is created for laser surgery to treat conditions such as diabetic retinopathy and retinal breaks.     

Rhythmogenesis in a hybrid system--interconnecting an olivary neuron to an analog network of coupled oscillators

A hybrid system in which an olivary neuron was interconnected to an analog simulator was used to study possible mechanisms by which the combined behavior of individual olivary neurons evokes synchronized membrane potential oscillations in a large population of neurons. The analog system was composed of four identical, interconnected oscillating units, each of which was capable of producing a damped sinusoidal oscillation in response to a trigger signal. When the units were coupled to each other, a single trigger pulse to one unit could evoke sustained oscillations. The integrity of the connections within the system was required to maintain these oscillations. In the hybrid system the analog system was reciprocally connected to an olivary neuron in a slice preparation. As in the analog system, the hybrid system could generate sustained oscillations following a trigger pulse to one of the units, as well as following a low threshold Ca spike in the neuron. Activation of the low threshold Ca conductance in the olivary neuron was necessary to achieve both gain and in-phase activity within the hybrid system, and thereby sustain the oscillations. The ability of the hybrid system to generate sustained oscillation is frequency dependent. Sustained oscillations were readily obtained at a "preferred frequency" of 5.2 Hz (n = 7) which was independent on the parameters used by the simulator, or on the membrane potential of the neurons. These results, which demonstrate the advantage of a new experimental approach developed to study rhythmogenesis in inferior olivary neurons, support the hypothesis that the inferior olivary nucleus, acting as an interconnected network of oscillating units, can generate an accurate subthreshold oscillation that serves as an internal time reference.     

低等哺乳动物的学习行为训练系统的研制

目的针对缺乏适应于低等哺乳动物条件反射与辨别学习行为的综合训练平台问题,本文设计并构建了一套适用于训练低等哺乳动物建立行为模型的自动训练系统。方法系统包括信号刺激与信号采集分析两大部分：刺激子系统产生条件刺激与非条件刺激信号训练刺激动物;信号采集分析子系统则以专用的生物信号采集装置采集动物的眨眼行为信号与脑电信号,记录并分析动物在条件刺激与非条件刺激下的眨眼行为信号与脑电信号。实验以豚鼠作为研究对象,分组进行条件刺激与非条件刺激配对实验。结果实验结果表明,通过对脑电信号的分析可以准确区分和判定豚鼠对特定声音的反应。豚鼠经过该系统的训练,形成了条件反射与辨别学习能力。结论该系统的成功设计为大脑神经活动和行为之间的相关性研究提供了实验平台。     

Automated prostate recognition: a key process for clinically effective robotic prostatectomy

Clinical trials of PROBOT, a robotic system for prostate surgery, have shown that robotic surgery of soft tissue can be successful. Monitoring of the progress of the resection has shown to be a necessary feature of an effective robotic system for prostate surgery. It should provide the surgeon with a reliable method of assessing the cavity during resection. An automatic system for intraoperative monitoring of the progress of the resection during robotic prostatectomy consists of two subsystems: real-time intraoperative imaging of the prostate and automatic identification of the contour of the gland on each image. The development of a fully automatic scheme for prostate recognition on transurethral ultrasound scans is reported. A genetic algorithm has been developed to automatically adjust a model of the prostate boundary until an optimum fit to the prostate in a given image is obtained. An analysis of its performance on 22 different ultrasound images showed an average error of 6.21 mm. Use of a genetic algorithm and a constrained prostate model have shown to be a robust way to automatically identify the prostate in ultrasound images. The scheme is able to produce approximate prostate boundaries, without any human intervention, on ultrasound scans of varying quality. In addition to soft tissue robotic surgery, the genetic algorithm technique is also applicable to a wide range of computer assisted surgical techniques.     

Automated prostate recognition: a key process for clinically effective robotic prostatectomy

Clinical trials of PROBOT, a robotic system for prostate surgery, have shown that robotic surgery of soft tissue can be successful. Monitoring of the progress of the resection has shown to be a necessary feature of an effective robotic system for prostate surgery. It should provide the surgeon with a reliable method of assessing the cavity during resection. An automatic system for intraoperative monitoring of the progress of the resection during robotic prostatectomy consists of two subsystems: real-time intraoperative imaging of the prostate and automatic identification of the contour of the gland on each image. The development of a fully automatic scheme for prostate recognition on transurethral ultrasound scans is reported. A genetic algorithm has been developed to automatically adjust a model of the prostate boundary until an optimum fit to the prostate in a given image is obtained. An analysis of its performance on 22 different ultrasound images showed an average error of 6.21 mm. Use of a genetic algorithm and a constrained prostate model have shown to be a robust way to automatically identify the prostate in ultrasound images. The scheme is able to produce approximate prostate boundaries, without any human intervention, on ultrasound scans of varying quality. In addition to soft tissue robotic surgery, the genetic algorithm technique is also applicable to a wide range of computer assisted surgical techniques.     

Subsystem throughputs of a clinical picture archiving and communications system

We measured the throughtput rates of individual picture archiving and communications system (PACS) subsystems including the acquisition, archive, display, and communication network as a basis of evaluation the overall throughput of our clinical PACS. The throughput rate of each PACS subsystem was measured in terms of average residence time of individual images in the subsystem. The residence time of an image in a PACS subsystem was determined by the total time the image was required to be processed within the subsystem. The overall throughput of the PACS was measured as the total residence time of an image in the various subsystems. We also measured throughputs of the PACS subsystems using three types of networks (Ethernet; fiber distributed data interface; and UltraNet, UltraNetwork Technologies, San Jose, CA), and the results were compared. Approximately 200 gigabytes of data transactions including magnetic resonance, computed tomography and computed radiography images from our PACS were analyzed. Results showed that PACS throughput was limited by three major factors: (1) low-speed data interface used in the radiologic imaging devices and archive devices; (2) competition for systems processing time among the PACS processes; and (3) network degradation caused by heavy network traffic. We concluded that PACS performance could be improved with a well-designed network architecture, a job prioritizing mechanism, and an image routing strategy. However, device-dependent low-speed data interface has limited PACS performance.     

Differential coding by two olfactory subsystems in the honeybee brain

Sensory systems use parallel processing to extract and process different features of environmental stimuli. Parallel processing has been studied in the auditory, visual, and somatosensory systems, but equivalent research in the olfactory modality is scarce. The honeybee Apis mellifera is an interesting model for such research as its relatively simple brain contains a dual olfactory system, with a clear neural dichotomy from the periphery to higher-order centers, based on two main neuronal tracts [medial (m) and lateral (l) antenno-protocerebral tract (APT)]. The function of this dual system is as yet unknown, and attributes like odor quality and odor quantity might be separately encoded in these subsystems. We have thus studied olfactory coding at the input of both subsystems, using in vivo calcium imaging. As one of the subsystems (m-APT) has never been imaged before, a novel imaging preparation was developed to this end, and responses to a panel of aliphatic odorants at different concentrations were compared in both subsystems. Our data show a global redundancy of olfactory coding at the input of both subsystems but unravel some specificities for encoding chemical group and carbon chain length of odor molecules.     

Inflammatory Response Assessment of a Hybrid Tissue-Engineered Heart Valve Leaflet

Despite substantial research in the past few decades, only slight progress has been made toward developing biocompatible, tissue-engineered scaffolds for heart valve leaflets that can withstand the dynamic pressure inside the heart. Recent progress on the development of hybrid scaffolds, which are composed of a thin metal mesh enclosed by multi-layered tissue, appear to be promising for heart valve engineering. This approach retains all the advantages of biological scaffolds while developing a strong extracellular matrix backbone to withstand dynamic loading. This study aims to test the inflammatory response of hybrid tissue-engineered leaflets based on characterizing the activation of macrophage cells cultured on the surfaces of the tissue construct. The results indicate that integration of biological layers around a metal mesh core—regardless of its type—may reduce the evoked inflammatory responses by THP-1 monocyte-like cells. This observation implies that masking a metal implant within a tissue construct prior to implantation can hide it from the immune system and may improve the implant’s biocompatibility.     

Acute adaptation of the vestibuloocular reflex: signal processing by floccular and ventral parafloccular Purkinje cells

The gain of the vertical vestibuloocular reflex (VVOR), defined as eye velocity/head velocity was adapted in squirrel monkeys by employing visual-vestibular mismatch stimuli. VVOR gain, measured in the dark, could be trained to values between 0.4 and 1.5. Single-unit activity of vertical zone Purkinje cells was recorded from the flocculus and ventral paraflocculus in alert squirrel monkeys before and during the gain change training. Our goal was to evaluate the site(s) of learning of the gain change. To aid in the evaluation, a model of the vertical optokinetic reflex (VOKR) and VVOR was constructed consisting of floccular and nonfloccular systems divided into subsystems based on the known anatomy and input and output parameters. Three kinds of input to floccular Purkinje cells via mossy fibers were explicitly described, namely vestibular, visual (retinal slip), and efference copy of eye movement. The characteristics of each subsystem (gain and phase) were identified at different VOR gains by reconstructing single-unit activity of Purkinje cells during VOKR and VVOR with multiple linear regression models consisting of sensory input and motor output signals. Model adequacy was checked by evaluating the residual following the regressions and by predicting Purkinje cells' activity during visual-vestibular mismatch paradigms. As a result, parallel changes in identified characteristics with VVOR adaptation were found in the prefloccular/floccular subsystem that conveys vestibular signals and in the nonfloccular subsystem that conveys vestibular signals, while no change was found in other subsystems, namely prefloccular/floccular subsystems conveying efference copy or visual signals, nonfloccular subsystem conveying visual signals, and postfloccular subsystem transforming Purkinje cell activity to eye movements. The result suggests multiple sites for VVOR motor learning including both flocculus and nonflocculus pathways. The gain change in the nonfloccular vestibular subsystem was in the correct direction to cause VOR gain adaptation while the change in the prefloccular/floccular vestibular subsystem was incorrect (anti-compensatory). This apparent incorrect directional change might serve to prevent instability of the VOR caused by positive feedback via the efference copy pathway.     

Endoprobe: a system for radionuclide-guided endoscopy

Methods to guide the surgical treatment of cancer utilizing handheld beta-sensitive probes in conjunction with tumor-avid radiopharmaceuticals [such as 18F-fluorodeoxyglucose (FDG)] have previously been developed. These technologies could also potentially be used to assist in minimally invasive techniques for the diagnosis of cancer. The goal of this project is to develop and test a system for performing radionuclide-guided endoscopies. This system (called Endoprobe) has four major subsystems: beta detector, position tracker, endoscope, and user interface. The beta detection unit utilizes two miniaturized solid state detectors to preferentially detect beta particles. The position tracking system allows real-time monitoring of the unit's location. The beta detector and position tracking system's receiver are mounted on the tip of an endoscope. Information from the beta detector and tracking system, in addition to the video signal from the endoscope, are combined and presented to the user via a computer interface. The system was tested in a simulated search for radiotracer-avid areas of esophageal cancer. The search for esophageal cancer was chosen because this type of cancer is often diagnosed with endoscopic procedures and has been reported to have good affinity for FDG. Accumulations of FDG in the normal organs of the abdomen were simulated by an anthropomorphic torso phantom filled with the appropriate amounts of radioactivity. A 1.5- mm-thick gelatin film containing FDG was used to simulate radiotracer uptake in the lining of normal esophagus. Esophageal lesions (both benign and malignant) were simulated by thin disks of gelatin (diameters=3.5-12 mm) containing appropriate concentrations of FDG embedded in the gelatin film simulating normal esophagus. Endoprobe facilitated visual identification and examination of the simulated lesions. The position tracking system permitted the location of the Endoprobe tip to be monitored and plotted in real time on a previously acquired positron emission tomography-computed tomography (PET-CT) image of the phantom. The detection system successfully acquired estimates of the beta flux emitted from areas chosen by the user. Indeed, Endoprobe was able to assist in distinguishing simulated FDG-avid areas as small as 3.5 mm in diameter from normal esophagus (p value <0.025). In addition to FDG, Endoprobe can be used with other positron or electron-emitting radionuclides such as IC or 131I. The next phase of this project will focus on modification of the prototype to make it more suitable for clinical use.     

Development and Characterization of Microfluidic Devices and Systems for Magnetic Bead-Based Biochemical Detection

This paper presents the development and characterization of a generic microfluidic system for magnetic bead-based biochemical detection. Microfluidic and electrochemical detection devices such as microvalves, flow sensors, biofilters, and immunosensors have been successfully developed and individually characterized in this work. Magnetically driven microvalves, pulsed-mode microflow sensors, magnetic particle separators as biofilters, and electrochemical immunosensors have been sep-arately fabricated and tested. The fabricated microfluidic components have been surface-mounted on the microfluidic motherboard for fully integrated microfluidic biochemical detection system. A magnetic bio-bead approach has been adopted for both sampling and manipulating target biological molecules. Magnetic beads were used as both substrate of antibodies and carriers of target antigens for magnetic bead-based immunoassay, which was chosen as a proof-of-concept for the generic microfluidic bio-chemical detection system. The microfluidic and electrochemical immunosensing experiment results obtained from this work have shown that the biochemical sensing capability of the complete microfluidic subsystem is suitable for portable biochemical detection of bio-molecules. The methodology and system, which has been developed in this work, can be extended to generic bio-molecule detection and analysis systems by replacing antibody/antigen with appropriate bio receptors/reagents such as DNA fragments or oligonucleotides for application towards DNA analysis and/or high throughput protein analysis.     

HIFU治疗热场中的有限元分析

本文对生物传热机制及生物热传导方程进行了研究 ,提出了一个混合型的生物热传导模型。针对带有血管的组织建立了高强度聚焦超声治疗下的组织热传导计算模型 ,并利用有限元方法进行了 3 D瞬态温度场分析。同时 ,根据Sapareto Dewey方法给出了组织消融量的仿真结果。     

A serial interface design to integrate bedside devices into patient monitoring systems

While substantial progress has been made toward the development of the Medical Information Bus (MIB), there is a pressing need, today, to integrate bedside devices into patient monitoring systems. To help meet this need, SpaceLabs Inc. has developed serial interfaces that integrate various stand-alone devices into a patient monitoring system. The company makes available a Universal Flexport Protocol to manufacturers who wish to make their devices compatible with the SpaceLabs patient monitoring systems. This paper describes the design of the Flexport serial interfaces and the Universal Flexport Protocol.     

Motion compensation in radiotherapy

Image-guided radiotherapy (IGRT) has helped to dramatically reduce safety margins compensating for positioning uncertainties in radiotherapy. A remaining issue posing problems for photon radiotherapy (RT), but even more so for particle RT, is target motion during treatment delivery. This review outlines the various strategies currently being developed or already in clinical use to compensate for organ motion, predominantly breathing-induced motion of liver and lung targets. Several motion compensation strategies have recently been introduced clinically. Among these are optimized margins encompassing the individual range of target motion, treatment under breath hold, gated treatments, and tumor tracking with a dedicated treatment device. A variety of surveillance strategies for gating and tracking, such as indirect tracking with external fiducial markers and surface scanning devices, direct tracking with implanted electromagnetic markers, fiducial markers, and fluoroscopy, and ultrasound-based tracking are already in clinical use or are under development. Tracked treatment with linear accelerators based on tumor-synchronous MLC- or treatment-table adaptation are moving toward clinical use. A multitude of strategies to reduce the impact of intrafractional target motion in RT have been developed and are increasingly being used clinically. The clinical introduction of advanced strategies currently under development is imminent. After IGRT minimized treatment margins for static tumors, the implementation of motion compensation strategies will achieve the same for targets being subject to intrafractional breathing-induced motion.     

Temporal and instantaneous connectivity of default mode network estimated using Gaussian Bayesian network frameworks

By probing its functional anatomy, the default mode network (DMN) can be considered consisting of two interacting hub and non-hub subsystems. The hub subsystem includes posterior cingulate cortex (PCC), medial prefrontal cortex (MPFC) and bilateral inferior parietal cortex (IPC). The non-hub subsystem contains inferior temporal cortex (ITC) and (para) hippocampus (HC). In this study, Gaussian Bayesian Network (BN) and Gaussian Dynamic Bayesian Network (DBN) were applied separately to detect the instantaneous and temporal connection relationship within each and between the two DMN subsystems. It was found that the directional instantaneous interactions between the two subsystems were primarily “from non-hub to hub”. The temporal interactions between hub and non-hub regions, on the other hand, are less presented between the two subsystems. The hub subsystem demonstrated both strong instantaneous and temporal interactions among the hub regions, while the non-hub regions were only strongly inter-connected instantaneously but temporally isolated with each other. In addition, one of the hub regions, PCC, appears to be a confluent node and important in the functional integration within the network.     

Image navigation as a means to expand the boundaries of fluorescence-guided surgery

Hybrid tracers that are both radioactive and fluorescent help extend the use of fluorescence-guided surgery to deeper structures. Such hybrid tracers facilitate preoperative surgical planning using (3D) scintigraphic images and enable synchronous intraoperative radio- and fluorescence guidance. Nevertheless, we previously found that improved orientation during laparoscopic surgery remains desirable. Here we illustrate how intraoperative navigation based on optical tracking of a fluorescence endoscope may help further improve the accuracy of hybrid surgical guidance. After feeding SPECT/CT images with an optical fiducial as a reference target to the navigation system, optical tracking could be used to position the tip of the fluorescence endoscope relative to the preoperative 3D imaging data. This hybrid navigation approach allowed us to accurately identify marker seeds in a phantom setup. The multispectral nature of the fluorescence endoscope enabled stepwise visualization of the two clinically approved fluorescent dyes, fluorescein and indocyanine green. In addition, the approach was used to navigate toward the prostate in a patient undergoing robot-assisted prostatectomy. Navigation of the tracked fluorescence endoscope toward the target identified on SPECT/CT resulted in real-time gradual visualization of the fluorescent signal in the prostate, thus providing an intraoperative confirmation of the navigation accuracy.     

On the peripheral bladder control system of the dog and urodynamics:In vivo characterisation and hybrid computer simulation

A quantitative model of the peripheral bladder control system of the dog is developed. This model is based on the experimental characterisation of the detrusor muscle and the urethra in 25 female dogs. Intravesical pressure, urinary flow and bladder volume were simultaneously recorded during electrical bladder stimulation, and, from these data, a dynamic model of the bladder control system has been deduced. A general equation for predicting urinary flow patterns is developed, and the hybrid computer EAI-690 is used to simulate the evacuation of the bladder. The effect of different parameters of the system on the micturition pattern is investigated in detail. This work may have interesting clinical applications: for example, in the development of adequate techniques for the evacuation of the paraplegic bladder. Also, the experimental procedures elaborated in this study could be applied in the diagnosis of bladder function and urethral obstruction. Indeed, we have been able to characterise the detrusor muscle and the urethra independently using clinically accessible variables.