Correlation between signal-to-noise ratios and region of interest sensitivity ratios of bipolar EEG measurements

We have developed a parameter, which describes how well the measurement is concentrated on the region of interest source area compared to other source areas in the volume conductor. The parameter concept is called the region of interest sensitivity ratio (ROISR). We assume that ROISR is also connected to the SNR of the measurement. The objective of the present study was to investigate the assumed correlation between the ROISR and SNR of the measurement with three-layer spherical head model. We studied how the source distribution and orientation affect the correlation and thus how applicable the ROISR is in analysing the sensitivity distributions of measurements. We simulated bipolar EEG-evoked potential measurements with 16 combinations of four-source distribution and four-source orientation models. The results indicate that the ROISR correlates with the SNR of the measurement with all tested source distributions and orientations. Thus the ROISRs concept can be applied to analyse measurement setups by modelling and analysing the sensitivity distributions.     

Region of Interest Sensitivity Ratio in Analyzing Sensitivity Distributions of Electrocardiographic Measurements

Modern devices such as wearable and implantable systems provide new methods for bioelectric measurement while creating new needs to assess the efficacy of the measurements. A modeling related analysis method, called the region of interest sensitivity ratio (ROISR), has been developed and applied hitherto in analyzing EEG measurements. ROISR describes how well the sensitivity of a measurement is concentrated within the region of interest (ROI). The objectives of the present study were to demonstrate and evaluate the applicability of this method in analyzing the specificity of ECG leads and to compare the results with those previously published. Here the specificities of the 117 leads of a Dalhousie body surface ECG mapping system to different segments of the left ventricular myocardium and the whole right ventricular free wall were analyzed. The ROISR method was also validated with epicardial dipole simulations. The results are consistent with those from previous clinical studies, and the ROISR method is thus applicable in future studies where, e.g., sensitivity distributions of implantable ECG measurements are analyzed.     

Spatial sensitivity of acousto-optic and optical near-infrared spectroscopy sensing measurements

Near-infrared spectroscopy (NIRS) is a popular sensing technique to measure tissue oxygenation noninvasively. However, the region of interest (ROI) is often beneath a superficial layer, which affects its accuracy. By applying focused ultrasound in the ROI, acousto-optic (AO) techniques can potentially minimize the effect of physiological changes in the superficial layer. Using absorption perturbation experiments in both transmission and reflection modes, we investigated the spatial sensitivity distributions and mean penetration depths of an AO system based on a digital correlator and two popular NIRS systems based on i. intensity measurements using a single source and detector configuration, and ii. spatially resolved spectroscopy. Our results show that for both transmission and reflection modes, the peak relative sensitivities of the two NIRS systems are near to the superficial regions, whereas those of the AO technique are near to the ROIs. In the reflection mode, when the ROI is deeper than 14 mm, the AO technique has a higher absolute mean sensitivity than the two NIRS techniques. As the focused ultrasound is moved deeper into the turbid medium, the mean penetration depth increases accordingly. The focused ultrasound can shift the peak relative sensitivity of the AO measurement toward its focused region.     

Influence of Image Metrics When Assessing Image Quality from a Test Object in Cardiac X-ray Systems

Modern fluoroscopic systems used for invasive cardiology typically acquire digital images in a 1,024 × 1,024 × 12 bits. These images are maintained in the original format while they remain on the imaging system itself. However, images are usually stored using a reduced 512 × 512 × 8-bits format. This paper presents a method for digital analysis of test objects images. The results obtained using image-intensifier and flat-detector systems are given for the original and reduced matrices. Images were acquired using a test object (TO) and a range of polymethyl methacrylate (PMMA) thicknesses from 4 to 28 cm. Adult patient protocols were evaluated for 16–28 cm of PMMA using the image-intensifier system. Pediatric protocols were evaluated for 4–16 cm of PMMA using the flat-detector system. The TO contains disks of various thicknesses to evaluate low contrast sensitivity and a bar pattern to evaluate high-contrast spatial resolution (HCSR). All available fluoroscopic and cine modes were evaluated. Entrance surface air kerma was also measured. Signal-to-noise ratio (SNR) was evaluated using regions of interest (ROI). HCSR was evaluated by comparing the statistical analysis of a ROI placed over the image of the bar pattern against a reference ROI. For both systems, an improvement of approximately 20% was observed for the SNR on the reduced matrices. However, the HCSR parameter was substantially lower in the reduced metrics. Cardiologists should consider the clinical influence of reduced spatial resolution when using the archived images.     

Treatment planning for prostate brachytherapy using region of interest adjoint functions and a greedy heuristic

We have developed an efficient treatment-planning algorithm for prostate implants that is based on region of interest (ROI) adjoint functions and a greedy heuristic. For this work, we define the adjoint function for an ROI as the sensitivity of the average dose in the ROI to a unit-strength brachytherapy source at any seed position. The greedy heuristic uses a ratio of target and critical structure adjoint functions to rank seed positions according to their ability to irradiate the target ROI while sparing critical structure ROIs. This ratio is computed once for each seed position prior to the optimization process. Optimization is performed by a greedy heuristic that selects seed positions according to their ratio values. With this method, clinically acceptable treatment plans are obtained in less than 2 s. For comparison, a branch-and-bound method to solve a mixed integer-programming model took more than 50 min to arrive at a feasible solution. Both methods achieved good treatment plans, but the speedup provided by the greedy heuristic was a factor of approximately 1500. This attribute makes this algorithm suitable for intra-operative real-time treatment planning.     

脊柱结核和恶性肿瘤弥散加权成像中感兴趣区设置方法初探

为了探讨脊柱结核和脊柱恶性肿瘤MR弥散加权成像(DWI)中感兴趣区(ROI)设置对表观弥散系数(ADC)值测量的影响,为ADC值测量选择较为合理、有效的测量方法。我们收集2009年8月-2010年6月间32例经病理和临床随访证实的脊柱结核和恶性肿瘤患者的常规MRI和DWI资料。以常规MRI为指导,在DWI图像中病变最大层面,分别以整个病变区域取单个ROI和在整个病变区域内取面积相同的多个ROI,测量其信号强度(SI)和ADC值并进行统计学分析。由受试者工作特性(ROC)曲线确定临界点后,比较两种方法所测得的ADC值诊断脊柱结核的灵敏度、特异度和准确率。结果发现单个ROI测得的ADC均值较多个ROI低,但差异没有统计学意义(P>0.05)。单个ROI或多个ROI时脊柱结核与肿瘤的ADC均值差别都有统计学意义(P<0.05),多个ROI测得的ADC值诊断脊柱结核的灵敏度、特异度和准确率高于前者。因此,对于脊柱结核和肿瘤,多个ROI测得的ADC值较单个ROI测得的ADC值能更好地反映水分子的弥散状态,在脊柱结核与恶性肿瘤的鉴别诊断中更有价值。     

Numerical evaluation of shaped surface coil sensitivity at 63 MHz

Surface coils are widely used in magnetic resonance studies due to their superior signal to noise (SNR) properties. When shaping planar coils to cylindrical surfaces, the region with maximum sensitivity migrates from the coil plane towards the centre of the shaping radius. The influence of the coil current, the probe and tissue dimensions, the electrical tissue properties and the operating frequency on the B1 field strength of a coil has been studied using statistical methods. This analysis allows the dependence of the axial SNR distribution of circular and square surface coils on shaping radius and coil dimensions to be evaluated quantitatively using 3D finite element methods. An empirically derived equation describing the dependence of the SNR distributions on coil geometry and depth allows the optimum coil dimensions to be predicted for a given shaping radius and desired optimized depth of sensitivity. Simulations are validated experimentally using both B1 and SNR mapping techniques. A comparison between the axial SNR of circular and square coils demonstrated equal SNR distributions of coils with equivalent area at depth.     

An optimization framework to maximize signal‐to‐noise ratio in simultaneous multi‐slice body imaging

Parallel imaging is essential for the acceleration of abdominal and pelvic 2D multi‐slice imaging, in order to reduce scan time and mitigate motion artifacts. Controlled Aliasing In Parallel Imaging Results IN Higher Acceleration (CAIPIRINHA) accelerated imaging has been shown to increase the signal‐to‐noise ratio (SNR) significantly compared with in‐plane parallel imaging with similar acceleration. We hypothesize that for CAIPIRINHA‐accelerated abdominal imaging the consistency of image quality and SNR is more difficult to achieve due to the subject‐specific coil sensitivity profiles, caused by (1) flexible coil placement; (2) variations in anatomy; and (3) variations in scan coverage along the superior–inferior direction. To test this, a mathematical framework is introduced that calculates the (retained) SNR for in‐plane and simultaneous multi‐slice (SMS)‐accelerated acquisitions. Moreover, this framework was used to optimize the sampling pattern by maximizing the local SNR within a region of interest (ROI) through non‐linear, RF‐induced CAIPIRINHA slice shifts. The framework was evaluated on 14 healthy subjects and the optimized sampling pattern was compared with in‐plane acceleration and CAIPIRINHA acceleration with linear slice shifts, which are primarily used in brain imaging. We demonstrate that the field of view (FOV) in the superior–inferior direction, the coil positioning and the individual anatomy have a large impact on the image SNR (changes up to 50% for varying coil positions and 40% differences between subjects) and image artifacts for simultaneous multi‐slice acceleration. Consequently, sampling patterns have to be optimized for acquisitions employing different FOVs and ideally on an individual basis. Optimization of the sampling pattern, which exploits non‐linear shifts between slices, showed a considerable SNR increase (10–30%) for higher acceleration factors. The framework outlined in this article can be used to optimize sampling patterns for a broad range of accelerated body acquisitions on an individual basis. Copyright © 2015 John Wiley & Sons, Ltd.     

多b值弥散加权成像的综合应用研究

该研究介绍了一种分析人大脑水分子弥散信息的新方法。该方法通过综合利用多b值的弥散加权成像信息,使用新的算法,计算出一个新的弥散参数(Rd),并重建出该参数的图像。在对25名正常成人脑组织的定量研究结果表明,相比于表观弥散系数(ADC),该方法能明显提高弥散图像的信噪比,有效区分大脑不同组织结构,提高了对水分子弥散运动的敏感性。     

On the analysis of single versus multiple channels of electromagnetic brain signals

OBJECTIVE: When extracting information from electromagnetic (EM) brain function through recordings such as the electroencephalogram (EEG) it is often assumed that signal processing techniques must be applied to multiple simultaneous recordings in order to obtain useful results. However, sometimes only a single channel of EEG recording is available or desirable. In this paper we objectively assess a novel methodology which exploits only a single measurement channel to extract information of interest relatively independent of channel location (relative to the source of interest). METHODS: The method relies on a combination of a matrix of delay vectors constructed from the single channel measurement, along with constrained independent component analysis, which incorporates prior information into the process. MATERIALS: Here, we use synthetically generated seizure EEG, composed of real, normal multi-channel EEG onto which is superimposed synthetic epileptic "seizure-like" activity, at different signal-to-noise (SNR) levels, through an equivalent current dipole model. RESULTS: We show that the method can extract desired information from single channels with a reasonable accuracy even at very small SNR and from channels distant from the focus of the activity. This provides a powerful technique capable of extracting multiple sources underlying single channel recordings and will be useful in situations where only single channel EM recordings of brain function are desirable, such as would be the case in wearable or implantable recording devices.     

基于最佳截断嵌入码块编码和离散小波变换的医学图像复合压缩算法

根据医学图像信息相对集中的特点,提出了一种基于最佳截断嵌入码块编码和离散小波变换的医学图像任意形状感兴趣区域复合压缩方法,通过对图像感兴趣区域和背景区采用不同的编码方式,提高了医学图像压缩比,并确保了医学图像感兴趣区域的高质量重建。实验表明:该方法在重建图像质量和压缩比方面均达到了较好的性能。     

EEG assessment of brain activity: Spatial aspects,segmentation and imaging

High temporal resolution and sensitivity to index different functional brain states makes the EEG a powerful tool in psychophysiology. Its full potential can now be utilized since recording technology and computational power for the large data masses has become affordable. However, basic traditional strategies in EEG need reviewing.Conventional, spontaneous or evoked EEG traces which are used for various complex analyses give ambiguous information on EEG power (amplitude) and phase for a given point on the scalp. Principally, analysis should first be done over space, then over time, to avoid ambiguities or pre-selections. First or second spatial derivative computations can provide “reference-free” data for analyses over time. We propose to use direct, spatial approaches for the analysis of the scalp EEG field distributions when simultaneous recording in several EEG channels can be examined.The ambiguity of the conventional EEG waveshapes results in different, equally “correct” scalp maps of EEG power of the same multichannel data for different reference electrodes. An exeption are scalp maps of EEG power computed against the common, average reference, as they are related to the reference-free spatial distribution (maps) of the maximal and minimal (extreme) field values over time, and thus are directly interpretable in terms of net orientation of the generator process.A proposed, reference-free EEG segmentation into epochs of periodically stationary spatial distributions of the mapped scalp EEG fields uses the locations of maximal and minimal (extreme) field values at each moment in time as classifiers, and thus avoids the priviledging of two arbitrarily chosen recording points in the field.     

How should low-contrast detail detectability be measured in fluoroscopy?

The relationship and precision of four methods for measuring the low-contrast detail detectability in fluoroscopic imaging were studied. These included the physical measurement of the accumulation rate of the square of the signal-to-noise ratio (SNR(rate)2), two-alternative forced-choice (2-AFC) experiments, sixteen-alternative forced-choice (16-AFC) experiments and subjective determination of the threshold contrast. The precision and sensitivity of the threshold contrast measurement were seen to be modest in the constancy testing of fluoroscopic equipment: only large changes in system performance could be reliably detected by that method. The measurement of the SNR(rate)2 is suggested instead. The relationship between the results of the various methods were studied, and it was found that human performance can be related to SNR(rate)2 by introducing the concept of the effective image information integration time (t(eff)). When measured for an unlimited observation time, it depicts the saturation of human performance in detecting a static low-contrast detail in dynamic image noise. Here, t(eff) was found to be about 0.6 s in 2-AFC tests and 0.3 s in 16-AFC tests.     

A 16-channel SQUID-device for biomagnetic investigations of small objects

Biomagnetic investigations in basic physiological research using animals require measurement devices different from commercial biomagnetometers used in human investigations. Two major problems have to be tackled in the design of such biomagnetometers. First, the spatial sampling needs to be much higher. Second, the distance between pick-up coils and the sources needs to be much shorter in order to compensate the worse signal-to-noise ratio (SNR) due to the smaller pick-up coils. We designed and built a 16-channel biomagnetic measurement system meeting these design criteria. The pick-up coil diameter of this new biomagnetometer is 6.7 mm, thus allowing 16 channels on an area of 3.2x3.2 cm2. The pick-up coils are located 3 mm above the dewar outer bottom, hence the closest distance to the cortical surface can be a few millimetres. We provide as an example of first measurements performed with the new biomagnetometer investigations of epileptic spikes in adult rabbits by simultaneous magnetoencephalogram (MEG) and electrocorticogram (ECoG) recordings. The high SNR of the recorded MEG and the simultaneously detected electric potentials allow investigations of the spatio-temporal pattern of neuronal processes of epileptiform spikes with signal strengths of about 3.5 pT.     

Reinvestigating hyperpolarized (129)Xe longitudinal relaxation time in the rat brain with noise considerations

The longitudinal relaxation time of hyperpolarized (HP) (129)Xe in the brain is a critical parameter for developing HP (129)Xe brain imaging and spectroscopy and optimizing the pulse sequences, especially in the case of cerebral blood flow measurements. Various studies have produced widely varying estimates of HP (129)Xe T(1) in the rat brain. To make improved measurements of HP (129)Xe T(1) in the rat brain and investigate how low signal-to-noise ratio (SNR) contributes to these discrepancies, we developed a multi-pulse protocol during the washout of (129)Xe from the brain. Afterwards, we applied an SNR threshold theory to both the multi-pulse protocol and an existing two-pulse protocol. The two protocols yielded mean +/- SD HP (129)Xe T(1) values in the rat brain of 15.3 +/- 1.2 and 16.2 +/- 0.9 s, suggesting that the low SNR might be a key reason for the wide range of T(1) values published in the literature, a problem that might be easily alleviated by taking SNR levels into account.     

Estimation of habituation and signal-to-noise ratio of cortical evoked potentials to oesophageal electrical and mechanical stimulation

Electrical and mechanical stimulation of the oesophagus has been recently proposed to examine the physiological effects of autonomic stimulation in humans. Cortical evoked potentials (EPs) to oesophageal stimulation provide an assessment of afferent fibres and central processing. However, habituation takes place during averaging of cortical EPs and reduces the signal-to-noise ratio (SNR) as the number of stimuli increases. The SNR of cortical EPs to oesophageal stimulation is computed for 15 normal subjects. Habituation is characterised by the Euclidean distance between the EEG response to single stimuli and the averaged EP, to serve as an objective measure of similarity between the averaged EP and the single-stimulus EEG. With electrical stimulation, the SNR is highest (0.41±0.21) for 1–12 stimuli and then significantly decreases to 0.2±0.08 for 13–24 stimuli (p<0.001). With balloon distension (BD), the SNR is highest (0.22±0.16) for 1–12 stimuli and lowest (0.12±0.14) for 13–24 stimuli, but these SNRs are not significantly different from each other. Both electrical and mechanical stimulation of the oesophagus produce rapidly adapting EPs. The SNR of the EPs is higher with electrical stimulation than with BD. The EPs response to BD has a higher variability and is more noisy. Consequently, these results suggest that the overall cortical EP response to electrical stimulation of the oesophagus is more reproducible than that due to balloon distension.     

基于S-G滤波的超声造影分析参数灌注成像

为弥补以往超声造影定量分析软件将感兴趣区(ROI)作为均质的整体区域进行分析的不足,提出一种基于S-G滤波器的在时域内直接处理数据平滑问题的定量分析方法,针对ROI区域内部像素点进行灌注成像参数分析,快速并可最大限度地保留原始有用信息。首先获得超声造影的原始DICOM动态医学数据,获取ROI内每个像素点的时间强度曲线,用S-G滤波器对时间强度曲线进行分析,得到反映灌注时相和灌注强度的多项定量分析参数,如到达时间(AT)、达峰时间(TTP)、达峰强度(PI)、平均通过时间(MTT)、曲线下面积(AUC)和上升斜率(RS)等;然后再对像素点参数分别进行彩色编码并以不同颜色显示,从而准确量化和直观显示超声造影的各项灌注参数。为评估该分析方法的稳定性,比较3位研究者分别对15个肝细胞癌病灶的分析结果。经统计,所有参数的组间相关系数(ICC)>0.80,验证了该方法的可行性和稳定性。     

Boosting the SNR by adding a receive‐only endorectal monopole to an external antenna array for high‐resolution,T2‐weighted imaging of early‐stage cervical cancer with 7‐T MRI

The aim of this study was to investigate the signal‐to‐noise ratio (SNR) gain in early‐stage cervical cancer at ultrahigh‐field MRI (e.g. 7 T) using a combination of multiple external antennas and a single endorectal antenna. In particular, we used an endorectal monopole antenna to increase the SNR in cervical magnetic resonance imaging (MRI). This should allow high‐resolution, T₂‐weighted imaging and magnetic resonance spectroscopy (MRS) for metabolic staging, which could facilitate the local tumor status assessment. In a prospective feasibility study, five healthy female volunteers and six patients with histologically proven stage IB1–IIB cervical cancer were scanned at 7 T. We used seven external fractionated dipole antennas for transmit–receive (transceive) and an endorectally placed monopole antenna for reception only. A region of interest, containing both normal cervix and tumor tissue, was selected for the SNR measurement. Separated signal and noise measurements were obtained in the region of the cervix for each element and in the near field of the monopole antenna (radius < 30 mm) to calculate the SNR gain of the endorectal antenna in each patient. We obtained high‐resolution, T₂‐weighted images with a voxel size of 0.7 × 0.8 × 3.0 mm³. In four cases with optimal placement of the endorectal antenna (verified on the T₂‐weighted images), a mean gain of 2.2 in SNR was obtained at the overall cervix and tumor tissue area. Within a radius of 30 mm from the monopole antenna, a mean SNR gain of 3.7 was achieved in the four optimal cases. Overlap between the two different regions of the SNR calculations was around 24%. We have demonstrated that the use of an endorectal monopole antenna substantially increases the SNR of 7‐T MRI at the cervical anatomy. Combined with the intrinsically high SNR of ultrahigh‐field MRI, this gain may be employed to obtain metabolic information using MRS and to enhance spatial resolutions to assess tumor invasion.     

Spatial variation of acoustic properties is related with mechanical properties of trabecular bone

In clinical applications, ultrasound parameters are measured as an average value over a region of interest (ROI) or as a value at a single measurement point. Due to natural adaptation to loading conditions, trabecular bone is structurally, compositionally and mechanically heterogeneous and anisotropic. Thus, spatial variation of ultrasound parameters within ROI may contain valuable information on the mechanical integrity of trabecular bone. However, this issue has not been thoroughly investigated. In the present study, we aimed at investigating the significance of the spatial variation of ultrasound parameters for the prediction of mechanical properties of human trabecular bone. For this aim, parametric maps of apparent integrated backscattering (AIB), integrated reflection coefficient (IRC), speed of sound (SOS), average attenuation (AA) and normalized broadband ultrasound attenuation (nBUA) were calculated for femoral and tibial bone cylinders (n = 19-20). Further, the effect of time window length on the AIB, variation of AIB within ROI and association between AIB and bone mechanical properties were characterized. Based on linear correlation analysis, spatial variation of AIB, assessed as standard deviation of measurements within ROI, was a strong predictor of bone ultimate strength (r = -0.82, n = 19, p < 0.01). Further, the time window length affected absolute values of AIB and strength of correlation between AIB and bone ultimate strength. Interestingly, linear combination of mean IRC and spatial variation of AIB within ROI was the strongest predictor of bone ultimate strength (r = 0.92, n = 19, p < 0.01). In conclusion, our findings suggest that the measurement of two-dimensional parametric maps of ultrasound parameters could yield information on bone status not extractable from single point measurements. This highlights the potential of parametric imaging in osteoporosis diagnostics.     

基于深度学习的颈动脉超声图像内中膜厚度测量

目的 颈动脉血管内中膜厚度(IMT)是衡量动脉粥样硬化程度的重要标准.一般采用人工标定进行测量,该过程耗时且繁琐,由此提出一种总体性能较好的全自动分割(AS)算法.方法 该算法首先利用卷积神经网络(CNN)识别出颈动脉血管远端,进而提取包含颈动脉内膜、中膜部分的感兴趣区域(ROI).采用基于堆栈式自编码器(SAE)构造的模式分类器将ROI中的像素进行分类.最后利用分类区域的面积信息和位置信息对分类结果进行甄别,运用曲线拟合提取边界完成测量任务.结果 针对本研究所用图像库中的84幅颈动脉超声图像进行实验,金标准(GT)由两名专家4次测量的平均值产生,其与AS之间的绝对误差和标准差为(13.3±20.5) μm,协方差系数为0.990 7.结论 实验结果表明,此算法总体性能较好,能够实现超声颈动脉血管内中膜全自动、快速、准确分割,从而满足临床需要.