A. Benassi

In vivo quantitative ultrasonic evaluation of myocardial fibrosis in humans.

The aim of this study was to assess in vivo whether the regional ultrasonic reflectivity, evaluated by a real-time integrated backscatter analysis, was related to the local content of connective tissue in human myocardium as estimated by quantitative histology of endomyocardial biopsies. Sixteen patients with presumptive diagnosis of cardiomyopathy were ultrasonically studied by means of an M-mode-based echocardiographic system with quantitative integrated backscatter analysis capabilities. A 2.25-MHz transducer was used. The integrated value of the rectified radiofrequency signal of the interventricular septum was taken as integrated backscatter index and expressed in percent normalized for the pericardial interface (assumed to be 100%). All patients also underwent multiple left ventricular endomyocardial biopsies, which were stained with Massons trichrome and studied with the use of a computer-assisted image analysis system. The percent integrated backscatter index was significantly higher in the presence of connective tissue area greater than 20% (eight patients) versus less than 20% (eight patients): 51 +/- 25% versus 26 +/- 11%, p less than 0.05. A significant correlation (p less than 0.05, R = 0.55) was found between percent integrated backscatter index and percent connective tissue area. In vivo on-line quantitative ultrasound analysis is feasible in man and reliably identifies variations in the regional extent of fibrosis in human myocardium.

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.

The use of frequency histograms of ultrasonic backscatter amplitudes for detection of atherosclerosis in vitro.

This study was designed to determine whether a quantitative analysis of integrated backscatter amplitude distribution is potentially useful in characterizing the atherosclerotic lesion. One hundred measurements (10 X 10 array) were made in fresh aortic regions (2 cm X 2 cm) of nine normal and 19 atherosclerotic arterial walls. A 10 MHz transducer was used. The integrated backscatter distinguished normal from atherosclerotic specimens (-56.7 +/- 4.3 vs -42.5 +/- 8.9 dB, p less than .01). The shape of the integrated backscatter amplitude distribution was analyzed by calculation of skewness and kurtosis of each arterial region. Both skewness values (0.134 +/- 0.325 vs -0.193 +/- 0.491 in normal and atherosclerotic segments, respectively, p = NS) and kurtosis values (0.055 +/- 0.765 vs -0.610 +/- 0.379, p less than .01) discriminated between the two groups. When only the six atherosclerotic specimens with mostly fatty and fibrofatty sites were considered, skewness and kurtosis still distinguished normal from atherosclerotic regions (0.134 +/- 0.325 vs -0.404 +/- 0.232, p less than .05 and 0.055 +/- 0.765 vs -0.558 +/- 0.337, p less than .05, respectively), while integrated backscatter values did not (-56.7 +/- 4.5 vs -52.3 +/- 6.1 dB, p = NS). In conclusion, atherosclerosis may be detected in vitro by the quantitative analysis of integrated backscatter distribution. This variable could also be of help in the identification of less obvious forms of atherosclerotic disease that are not distinguishable on the basis of integrated backscatter amplitude.

Flow quantitation by radio frequency analysis of contrast echocardiography.

Contrast echocardiography has the potential for measuring cardiac output and regional blood flow. However, accurate quantitation is limited both by the use of non-standard contrast agents and by the electronic signal distortion inherent to the echocardiographic instruments. Thus, the aim of this study is to quantify flow by combining a stable contrast agent and a modified echo equipment, able to sample the radio frequency (RF) signal from a region of interest (ROI) in the echo image. The contrast agent SHU-454 (0.8 ml) was bolus injected into anin vitro calf vein, at 23 flow rates (ranging from 376 to 3620 ml/min) but constant volume and pressure. The ROI was placed in the centre of the vein, the RF signal was processed in real time and transferred to a personal computer to generate time-intensity curves. In the absence of recirculation, contrast washout slope and mean transit time (MTT) of curves (1.11–8.52 seconds) yielded excellent correlations with flow: r=0.93 and 0.95, respectively. To compare the accuracy of RF analysis with that of conventional image processing as to flow quantitation, conventional images were collected in the same flow model by two different scanners: a) the mechanical sector scanner used for RF analysis, and b) a conventional electronic sector scanner. These images were digitized off-line, mean videodensity inside an identical ROI was measured and time-intensity curves were built. MTT by RF was shorter than by videodensitometric analysis of the images generated by the same scanner (p<0.001). In contrast, MTT by RF was longer than by the conventional scanner (p<0.001). Significant differences in MTT were also found with changes in the gain setting controls of the conventional scanner. To study the stability of the contrast effect, 6 contrast injections (20 ml) were performed at a constant flow rate during recirculation: the spontaneous decay in RF signal intensity (t1/2=64±8 seconds) was too long to affect MTT significantly.In conclusion, the combination of a stable con trast agent and a modified echocardiographic instrument provides accurate quantitation of flow in anin vitro model; RF analysis is more accurate than conventional processing as to flow quantitation by contrast echocardiography.

Flow quantitation by contrast echocardiography. Effects of intervening tissue and of the angle of incidence between flow and ultrasonic beam.

The combination of a standardized echographic contrast agent with the analysis of the ultrasonic radio frequency (RF) signal allowedin vitro flow quantitation in a circulation model. The purpose of this study was to investigate both the effects of biological tissues, intervening between probe and insonated structure, and the effects of the angle of incidence between flow and ultrasonic beam on RF flow quantitation. Thus, the contrast agent SHU 454 was intravenously injected (0.4 ml) as a bolus into a circulation model, at variable flow rates, while keeping the pressure and volume of the vessel constant. Injections were performed with saline interposed between probe and vessel and after the addition of the subcutaneous tissue of a pig; injections were also performed using the probe normal to the flow and with an angle of incidence of 45°. Echographic data were recorded by a mechanical sector scanner, capable of sampling the RF signal from a region of interest positioned in the center of the vein. Contrast echo time-intensity curves were generated. As expected, both peak intensity and the area under the curves decreased with intervening tissue (− 58 and − 70% of baseline values, respectively, p<0.001). Surprisingly, mean transit time also decreased with intervening tissue (from 1.12±0.25 seconds with saline, to 0.92±0.13 seconds with tissue, p<0.001), thus producing a systematic overestimation of flow (21% on the average). To compensate for signal attenuation, contrast injections were repeated in the presence of tissue after increasing the electronic signal amplification (10 dB), and transit time did not significantly differ from control. Moreover, mean transit time was slightly shorter with an angle of 45° (1.03 ° 0.19 seconds) than with an angle of 90° (1.12±0.25 seconds, p<0.05). However, when the data collected with both angles of incidence were plotted together, the correlation with flow remained very close (r=0.94). In conclusion: intervening tissue influences flow quantitation by contrast echocardiography: these modifications can be compensated for by increasing the electronic signal amplification; the angle of incidence between flow and ultrasonic beam also influences flow quantitation: however, this influence is trivial, and can be neglected for practical purposes.

New standards for cardiology report and data communication: an experience with HL7 CDA release 2 and EbXML

The diffusion of health standards, connected to technological infrastructure, finally opens the road to a new way of interconnecting and integrating different clinical solutions. New standards coming from established Organizations such as HL7 and DICOM, are going to be integrated with Communication Organization to build the Health environment Info-structure of the next decade. The purpose of this work is to show how this is possible, using simple and affordable methods: The recently approved HL7 Clinical Document Architecture Release 2 standard for message packaging and the newly approved ebXML 3.0 specifications. These promising standards can be used in conjunction with Web-Services and Web Service Description Language (WSDL) technology. This type of solution allows to be totally independent from the development platform, from the programming language and the operating system used. EbXML is an XML framework for business-to-business (B2B) environment developed by the ebXML Initiative

HL7 clinical document architecture to share cardiological images and structured data in next generation infrastructure

In medicine and in cardiology different standards are used for treatment of clinical and iconographic information. Among the most relevant there are HL7 for clinical data and DICOM for images and signals. The advent of electronic health record systems (EHR) and the request for data integration coming from different imaging modalities and diagnostic instrumentation, offer us a technological panorama difficult to manage. There is an increasing demand to select the most meaningful information in a simple and effective way, without the duty and the necessity to create from scratch new ways of data communication and sharing. Hence an agreement, in order to be able to feed a new multidisciplinary information databases, is necessary. To reach this goal, we used the emergent development of HL7 in the field of the structure of clinical document architecture (CDA).

An open source based application for integration and sharing of multi-modal cardiac image data in a heterogeneous environment

DICOM format has been recognized as the de facto standard for storage, transferring and sharing of cardiac images along different modalities like magnetic resonance imaging (MRI), nuclear medicine, computer tomography (CT), digital angiography (XA), digital radiology. However, in many medical environments there is a large need to have cardiac images available in formats (i.e. GIF, BMP or JPEG) that are compatible with widely used office automation applications. The here proposed system for multi modality image sharing and conversion is based on both the standard DICOM communication protocol and the common internet file system (CIFS) protocol. DICOM images from any DICOM compliant station can be transferred to a dedicated DICOM server, which has been implemented using the freely available DCMTK DICOM toolkit from Kuratorium OFFIS. An automatic procedure converts DICOM images into the desired image format, creating at the same time a file tree that allows fast and easy retrieval of image data. The entire proposed framework is implemented using open source software with large advantages in terms of cost reduction and software flexibility and robustness.

On-line two-dimensional evaluation of ultrasonic integrated backscatter

We describe an apparatus for the on-line evaluation of integrated backscatter from areas of tissue. The equipment is fully integrated into a B-mode ultrasonic system; there are therefore no new operating procedures to be learned. It provides a simultaneous display of conventional information, together with parameters of tissue characterization. The apparatus is fast and, over a broad diagnostic frequency range, may be used in conjunction with conventional equipment employing transducers.

ON-LINE EVALUATION OF ULTRASONIC INTEGRATED BACKSCATTER

Although it is already known that reflected ultrasonic signals (backscatter) are changed by the structure of the tissue through which they pass, clinicians are still awaiting a practical instrument in which information from backscatter reflections will serve as a diagnostic aid additional to that provided by conventional ultrasonic scans. The equipment described here is both small and fast, and is integrated into a normal ultrasound installation. No new operating procedures have to be learned. The integrated backscatter is calculated on-line and presented on an LED as tissue characterization parameters. In order to minimize noise due to physical movement of the heart during an investigation of the myocardium, the analysis is synchronized with the ECG; and as an aid to the user, the normal system VDU displays both the ECG and the activating trigger pulse derived from the R-wave peak. An A-scan display has been used but this could readily be adapted for B-scan operation and single line analysis. Tests with backscattering models and standard instrumentation have shown no significant difference between results using time domain or frequency domain analysis.