Marcello Demi

A System for Real-Time Measurement of the Brachial Artery Diameter in B -Mode Ultrasound Images

The measurement of the brachial artery diameter is frequently used in clinical studies for evaluating the flow-mediated dilation and, in conjunction with the blood pressure value, for assessing arterial stiffness. This paper presents a system for computing the brachial artery diameter in real-time by analyzing B-mode ultrasound images. The method is based on a robust edge detection algorithm which is used to automatically locate the two walls of the vessel. The measure of the diameter is obtained with subpixel precision and with a temporal resolution of 25 samples/s, so that the small dilations induced by the cardiac cycle can also be retrieved. The algorithm is implemented on a standalone video processing board which acquires the analog video signal from the ultrasound equipment. Results are shown in real-time on a graphical user interface. The system was tested both on synthetic ultrasound images and in clinical studies of flow-mediated dilation. Accuracy, robustness, and intra/inter observer variability of the method were evaluated

Real-time Measurement System for Evaluation of the Carotid Intima-Media Thickness With a Robust Edge Operator

Objective. The purpose of this report is to describe an automatic real‐time system for evaluation of the carotid intima‐media thickness (CIMT) characterized by 3 main features: minimal interobserver and intraobserver variability, real‐time capabilities, and great robustness against noise. Methods. One hundred fifty carotid B‐mode ultrasound images were used to validate the system. Two skilled operators were involved in the analysis. Agreement with the gold standard, defined as the mean of 2 manual measurements of a skilled operator, and the interobserver and intraobserver variability were quantitatively evaluated by regression analysis and Bland‐Altman statistics. Results. The automatic measure of the CIMT showed a mean bias ± SD of 0.001 ± 0.035 mm toward the manual measurement. The intraobserver variability, evaluated with Bland‐Altman plots, showed a bias that was not significantly different from 0, whereas the SD of the differences was greater in the manual analysis (0.038 mm) than in the automatic analysis (0.006 mm). For interobserver variability, the automatic measurement had a bias that was not significantly different from 0, with a satisfactory SD of the differences (0.01 mm), whereas in the manual measurement, a little bias was present (0.012 mm), and the SD of the differences was noticeably greater (0.044 mm). Conclusions. The CIMT has been accepted as a noninvasive marker of early vascular alteration. At present, the manual approach is largely used to estimate CIMT values. However, that method is highly operator dependent and time‐consuming. For these reasons, we developed a new system for the CIMT measurement that conjugates precision with real‐time analysis, thus providing considerable advantages in clinical practice.

Assessment of Carotid Stiffness and Intima-Media Thickness From Ultrasound Data Comparison Between Two Methods

Objective. Increased arterial stiffness and carotid intima‐media thickness (IMT) are considered independent predictors of cardiovascular events. The aim of this study was to compare a system recently developed in our laboratory for automatic assessment of these parameters from ultrasound image sequences to a reference radio frequency (RF) echo‐tracking system. Methods. Common carotid artery scans of 21 patients with cardiovascular risk factors and 12 healthy volunteers were analyzed by both devices for the assessment of diameter (D), IMT, and distension (ΔD). In the healthy volunteers, analyses were repeated twice to evaluate intraobserver variability. Agreement was evaluated by Bland‐Altman analysis, whereas reproducibility was expressed as a coefficient of variation (CV). Results. Regarding the agreement between the two systems, bias values ± SD were 0.060 ± 0.110 mm for D, –0.006 ± 0.039 mm for IMT, and –0.016 ± 0.039 mm for ΔD. Intraobserver CVs were 2% ± 2% for D, 5% ± 5% for IMT, and 6% ± 6% for ΔD with the RF echo‐tracking system and 2% ± 1% for D, 6% ± 6% for IMT, and 8% ± 6% for ΔD with our automated system. Conclusions. Although B‐mode‐based devices are less precise than RF‐based ones, our automated system has good agreement with the reference method and comparable reproducibility, at least when high‐quality images are analyzed. Hence, this study suggests that the presented system based on image processing from standard ultrasound scans is a suitable device for measuring IMT and local arterial stiffness parameters in clinical studies.

The First Absolute Central Moment in Low-Level Image Processing

The first absolute central moment is a statistical filter which measures the variability of the gray levels of an image with respect to the local mean. The analysis of the responses of the central and absolute central moments at noiseless isolated step discontinuities shows how the first absolute central moment can be usefull in enhancing these discontinuities. Moreover, experimental results show how a nonstandard form of the absolute central moment should be used to enhance other image key points. At noiseless step discontinuities, the first absolute central moment provides a ridge map similar to the one provided by the GoG magnitude. However, unlike the GoG magnitude, a nonstandard form of the first absolute central moment provides ridges at both edges and lines (pulse functions 1 pixel wide) and gives rise to local extrema of the ridges at line endings, corners, and intersections among different discontinuities. The analysis of the filter output in the presence of additive noise also shows that a generalized form of the first absolute central moment should be used to cope with noise properly. Both theoretical and experimental results show that, if right configurations of the generalized first absolute central moment are used, the filter retains most of its properties when real images are considered. Moreover, since the generalization of the original filter gives rise to a class of nonlinear filters, the recovered edge information can be also usefully combined; two examples are illustrated in this paper. The first one shows how to obtain the zero-crossing map of an equivalent DoG filter, whereas the second one shows how to obtain a local thresholding procedure.

Comparison of two automatic methods for the assessment of brachial artery flow-mediated dilation.

Objectives Brachial artery flow-mediated dilation (FMD) is associated with risk factors providing information on cardiovascular prognosis. Despite the large effort to standardize the methodology, the FMD examination is still characterized by problems of reproducibility and reliability that can be partially overcome with the use of automatic systems. We developed real-time software for the assessment of brachial FMD (FMD Studio, Institute of Clinical Physiology, Pisa, Italy) from ultrasound images. The aim of this study is to compare our system with another automatic method (Brachial Analyzer, MIA LLC, IA, USA) which is currently considered as a reference method in FMD assessment. Methods The agreement between systems was assessed as follows. Protocol 1: Mean baseline (Basal), maximal (Max) brachial artery diameter after forearm ischemia and FMD, calculated as maximal percentage diameter increase, have been evaluated in 60 recorded FMD sequences. Protocol 2: Values of diameter and FMD have been evaluated in 618 frames extracted from 12 sequences. Results All biases are negligible and standard deviations of the differences are satisfactory (protocol 1: −0.27 ± 0.59%; protocol 2: −0.26 ± 0.61%) for FMD measurements. Analysis times were reduced (−33%) when FMD Studio is used. Rejected examinations due to the poor quality were 2% with the FMD Studio and 5% with the Brachial Analyzer. Conclusions In conclusion, the compared systems show a optimal grade of agreement and they can be used interchangeably. Thus, the use of a system characterized by real-time functionalities could represent a referral method for assessing endothelial function in clinical trials.

On the Physical Basis of Pulmonary Sonographic Interstitial Syndrome.

ung sonography is widely accepted and used in emergency medicine and critical care.1–5 Moreover, many pulmonologists are interested in chest sonography for the study of pleural diseases and are increasingly discovering a role for sonography in parenchymal lung diseases.6–9 For those physicians who are devoted to chest sonography, a clear dichotomy between usual sonography and aerated tissue sonography is obvious. Pleural sonography is effective under most circumstances, whereas lung sonography is effective only when certain physical properties of the lung (eg, the bubble system) are lost. In other words, the lung is sonographically explorable only when it is physically comparable with soft tissue. In particular, when using lung sonography, a lung that contains dispersed air and has a density that is not comparable with the density of water does not show anatomic images but rather artifactual images.10 Therefore, lung artifacts are quite consistent with the physical properties of a lung that is not fully consolidated rather than with an anatomic image.11 The physical properties of the subpleural nonconsolidated lung are the hallmarks of many pulmonary diseases, which can be roughly grouped into “interstitial diseases.” If an ultrasound imaging system is used, all of these pulmonary diseases are classified by the generic term “sonographic interstitial syndrome” (B-lines with variable arrangements along the pleural line).5 According to this view, it is not surprising that since 1997,12 vertical lung artifacts, commonly named B-lines, have been associated with pathologic conditions ranging from pulmonary edema to fibrosis, which are characterized by a change in the subpleural physical features in terms of full and empty spaces.11 Gino Soldati, MD Emergency Medicine Unit Valle del Serchio General Hospital Lucca, Italy

A DSP-based real time contour tracking system

In this paper the real-time implementation of an automatic contour tracking procedure is presented. If a rough contour of the desired structure is available on the first frame of a sequence, the contours on the subsequent frames are automatically outlined. The mass center of the first-order absolute moment operator is used to locate the contour of interest on every frame. The contour tracking procedure was implemented on an integrated software/hardware environment based on the interaction of a PC and a DSP processor (TMS320C80). Standard PAL images at 25 frame/s were used as an input signal; the contour was computed in real time and was then drawn on the images and displayed on a VGA monitor. The results are illustrated both on clinical images recorded on a SVHS videocassette and on images of a test object acquired by means of a video camera.

Ultrasonography in lung pathologies: new perspectives.

BackgroundNowadays, ultrasound techniques have not gained importance in the diagnosis and monitoring of lung pathologies yet because of the high mismatch in acoustic impedance between air and intercostal tissues. However, it is evident that B-mode imaging provides important information on pulmonary tissue, although in the form of image artifacts.FindingsNotwithstanding medical evidences, there exists no ultrasound-based method dedicated to the lung, hampering de facto the full exploitation of ultrasound potentials. A chance is given by the experience acquired in other fields, where acoustic attenuation is used to estimate concentrations of suspended particles in liquids and of air-bubbles in aerated foods.ConclusionsCustom hardware must be developed since commercial echographic equipment has been optimized to work with low acoustic impedance mismatches, and, in general, does not provide the primitive radiofrequency (RF) signals nor the possibility to tune key acquisition parameters such as ultrasound carrier frequency and pulse bandwidth, which are surely needed for our application.

The use of lung ultrasound images for the differential diagnosis of pulmonary and cardiac interstitial pathology

AbstractIn recent years, great advances have been made in the use of lung ultrasound to detect pulmonary edema and interstitial changes in the lung. However, it is clear that B-lines oversimplify the description of the physical phenomena associated with their presence. The artifactual images that ultrasounds provide in interstitial pulmonary pathology are merely the ultimate outcome of the complex interaction of a specific acoustic wave with a specific three-dimensional biological structure. This interaction lacks a solid physical interpretation of the acoustic signs to support it. The aim of this paper was to describe the differences between the sonographic interstitial syndrome related to lung diseases and that related to cardiogenic edema in the light of current knowledge regarding the pleural plane’s response to ultrasound waves. SommarioNegli ultimi anni sono stati fatti grandi progressi in ecografia polmonare per rilevare l’edema polmonare e le alterazioni dell’interstizio polmonare. Tuttavia, è chiaro che le linee B semplificano eccessivamente la descrizione di fenomeni fisici associati alla loro presenza. Le immagini di artefatti date dagli ultrasuoni nella patologia polmonare interstiziale sono soltanto il risultato finale di una complessa interazione tra un’onda acustica specifica ed una altrettanto specifica struttura biologica tridimensionale. Questa interazione manca di una solida interpretazione fisica. Lo scopo di questo lavoro è quello di descrivere le differenze ecografiche tra la sindrome interstiziale dovuta a malattia polmonare e quelle relative ad edema cardiogeno, alla luce delle attuali conoscenze in merito alla risposta del piano pleurico agli ultrasuoni.

Detection of artery interfaces: a real-time system and its clinical applications

Analyzing the artery mechanics is a crucial issue because of its close relationship with several cardiovascular risk factors, such as hypertension and diabetes. Moreover, most of the work can be carried out by analyzing image sequences obtained with ultrasounds, that is with a non-invasive technique which allows a real-time visualization of the observed structures. For this reason, therefore, an accurate temporal localization of the main vessel interfaces becomes a central task for which the manual approach should be avoided since such a method is rather unreliable and time consuming. Real-time automatic systems are advantageously used to automatically locate the arterial interfaces. The automatic measurement reduces the inter/intra-observer variability with respect to the manual measurement which unavoidably depends on the experience of the operator. The real-time visual feedback, moreover, guides physicians when looking for the best position of the ultrasound probe, thus increasing the global robustness of the system. The automatic system which we developed is a stand-alone video processing system which acquires the analog video signal from the ultrasound equipment, performs all the measurements and shows the results in real-time. The localization algorithm of the artery tunics is based on a new mathematical operator (the first order absolute moment) and on a pattern recognition approach. Various clinical applications have been developed on board and validated through a comparison with gold-standard techniques: the assessment of intima-media thickness, the arterial distension, the flow-mediated dilation and the pulse wave velocity. With this paper, the results obtained on clinical trials are presented.