Search

Objectives: To evaluate virtual touch tissue imaging quantification (VTIQ) as a new elastography method concerning its intra- and interexaminer reliability and its ability to differentiate benign from malignant breast lesions in comparison to and in combination with ultrasound (US) B-mode breast imaging reporting and data system (BI-RADS) assessment., Materials and Methods: US and VTIQ were performed by two examiners in 103 women with 104 lesions. Intra- and interexaminer reliability of VTIQ was assessed. The area under the receiver operating curve (AUC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of BIRADS, VTIQ, and combined data were compared., Results: Fifty-four of 104 lesions were malignant. Intraexaminer reliability was consistent, and interexaminer agreement showed a strong positive correlation (r = 0.93). The mean VTIQ values in malignant lesions were significantly higher than those in benign (7.73 m/s ± 1.02 versus 4.46 m/s ± 1.87; P < 0.0001). The combination of US-BIRADS with the optimal cut-off for clinical decision making of 5.18 m/s yielded a sensitivity of 98%, specificity of 82%, PPV of 86%, and NPV of 98%. The combination of BIRADS and VTIQ led to improved test validity., Conclusion: VTIQ is highly reliable and reproducible. There is a significant difference regarding the mean maximum velocity of benign and malignant lesions. Adding VTIQ to BIRADS assessment improves the specificity.

Purpose: To compose diagnostic standard operating procedures for both clinical and imaging assessment for vulvar and vaginal cancer, for vaginal sarcoma, and for ovarian cancer., Methods: The literature was reviewed for diagnosing the above mentioned malignancies in the female pelvis. Special focus herein lies in tumor representation in MRI, followed by the evaluation of CT and PET/CT for this topic., Conclusion: MRI is a useful additional diagnostic complement but by no means replaces established methods of gynecologic diagnostics and ultrasound. In fact, MRI is only implemented in the guidelines for vulvar cancer. According to the current literature, CT is still the cross-sectional imaging modality of choice for evaluating ovarian cancer. PET/CT appears to have advantages for staging and follow-up in sarcomas and cancers of the ovaries.

Aim: The goal of this article is to provide an overview of diagnostic standard operating procedures for both clinical and imaging assessment of cervical and endometrial carcinoma, sarcoma of the uterus, and primary pelvic non-Hodgkin's lymphoma., Methods: The literature was reviewed for methods used to diagnose malignancies in the female pelvis with a special focus on the role of MRI as the imaging method of choice. Furthermore, CT findings and staging criteria for the mentioned malignancies are also provided., Conclusion: Whereas ultrasound still remains the imaging modality of choice in clinical practice for the early diagnosis of female pelvic malignancies, MRI is more frequently recognized as a diagnostic tool for its accuracy in tumor identification. MRI also plays a crucial role in the 3D pretreatment planning for brachytherapy especially in cervical cancer. In the future, PET/CT might achieve an important role for staging lymph nodes or distant metastases as well as tumor recurrence.

Aim: To compare the significance of the two-dimensional and three-dimensional sonography in the diagnosis of meniscal tears under clinical conditions., Methods: Sixty menisci of knees with clinical symptoms (44 medial and 16 lateral menisci) were examined by an identical transducer in a two- and three-dimensional sonography technique. The findings were compared with the diagnosis made with subsequent arthroscopy., Results: In the diagnosis of meniscal lesions the two- and three-dimensional sonography reached a sensitivity of 92% and 100% and a specificity of 83% and 88%, respectively. The positive predictive value of these methods was 58% and 67%. The negative predictive value was 98% and 100%. Because of the good results with the two-dimensional sonography, there were no statistically significant differences between both methods. The three-dimensional sonography, however, proved to be superior, to the two-dimensional sonography in the analysis of subgroups (medial and lateral menisci, menisci with and without clinical symptoms). The negative predictive value of the three-dimensional sonography was 100% for all of these subgroups., Conclusion: Although this study shows no significant difference in the results of two- and three-dimensional sonography, the analysis of subgroups displays a slightly improved significance for diagnosis of meniscal tears by three-dimensional sonography. The high negative predictive value shows that three-dimensional sonography, performed by an experienced examiner, may be useful to exclude meniscal tears. This result may help focus further cost-intensive or invasive examinations.

Background: Two new blood-flow-diagnosis techniques have recently been developed as supplements to the established color techniques: the MEM (maximum entropy method) technique and color flow amplitude (power Doppler). These are capable of representing blood flow in distinctly more slowly flowing areas than is possible with the conventional Doppler technique., Methods: Both methods make use of the Doppler technique in part, yet analyze the reflected signal in a different manner, in so doing optimizing the relation between the noise and the signal. Measurements were obtained on two anatomic flow models to test the sensitivity of both techniques under slow flow conditions., Results: The slowest flow the MEM technique was capable of recording was 0.5 mm/s, albeit utilizing a 5-MHz transducer for the MEM technique and a 10-MHz transducer for the angio technique. One may thus assume that the MEM technique would be still more sensitive when utilizing a 10-MHz transducer. The advantage of the MEM technique is its real-time flow representation: The angio technique requires a few seconds of acquisition time. This could have serious consequences during clinical utilization. Doppler sonography was merely capable of detecting a minimum flow velocity of 15 mm/s. The angio technique is less dependent on the angle during flow representation than the MEM technique., Conclusions: The distinctly higher sensitivity of these two new color techniques offers new possibilities in the clinical sector as far as the perfusion diagnosis of organs and tumors is concerned.

On the assumption that the architecture of blood vessels of malignant tumors, formed by neoangiogenesis, shows characteristics that are different from those of blood vessels of benign tumors or physiological findings, we have tried in the present study to investigate the behavior of these different vessels under increased blood pressure. Using a special stand, the same sonographic section could be stably maintained during an examination time of approx. 4 minutes. Using a new computer program, the color pixels of the employed Angio color technique were quantified and recorded as a function of the measured blood pressure. To increase blood pressure, the patient had to press a hand grip, which practically always caused a systolic blood pressure elevation of more than 15-20 mmHg. Seventy patients with sonographically detected breast tumors were examined; 54 (14 benign and 40 malignant tumors) could be included in the evaluation. We found four typical types of curves: Curve type 1 is associated with an instantaneous increase in blood flow with increased blood pressure, followed by a drop in the blood pressure, with a slow decrease in blood flow as the blood pressure drops (with 29 malignant and 3 benign tumors). Curve type 2 shows a continuous increase in blood flow-though somewhat delayed with respect to the rise in blood pressure-which is also observed when the blood pressure drops (exclusively benign tumors). In curve type 3, maximum blood flow is reached after the blood pressure maximum, and then the blood flow decreases (1 benign and 3 malignant tumors). Curve type 4 features decreased blood flow in spite of increased blood pressure (3 benign and 5 malignant tumors). The described quantification method, in combination with the stand, permits for the first time analysis of a tumor under increased blood pressure as to its blood flow behavior over time in an examination using a challenge test. Here one can find two distinctive curve types (types 1 and 2) that correlate mostly with malignant (type 1) or benign (type 2) breast tumors. Should this tendency be substantiated by additional large-scale studies, it would seem that a new ultrasonic possibility for differential diagnosis has been found.

Aim: The clinical value of two-dimensional ultrasound in examining the menisci of the knee compared to other methods is controversial. We studied the utility diagnosed by three-dimensional sonography in evaluating meniscal tears., Method: 96 menisci with standardised artificial lesions were examined in a bath of Ringer solution. Two-dimensional ultrasound was compared to three-dimensional ultrasound, which creates three-dimensional reconstruction of sequential ultrasonographic images., Results: The three-dimensional ultrasound had a higher sensitivity of 88% vs 69% with a specificity of 83% vs 94% for the diagnosis of meniscal tears compared with the two-dimensional method. The sensitivity in imaging was 54% in each case for longitudinal tears, 63% vs 96% for horizontal tears, 67% vs 37% for oblique tears and 54% vs 0% for radial tears. The difference of these results, however, was not statistically significant. Radial tears were more frequently diagnosed by three-dimensional sonography., Conclusion: In our model, sonographic diagnosis of meniscal tears was improved only partially by using three-dimensional ultrasound. Whether in vivo other factors alter the minimal differences between the two techniques is the subject of ongoing investigation.

Aim: The sensitivity of the two procedures for slow blood flow velocities in two flow phantoms examined and compared to conventional colour Doppler., Method: In two phantoms with vessels between 0.3 and 6 mm diameter the slowest measurable blood flow with the different techniques was determined. The influence of various application angles was also studied., Results: The slowest blood flow velocity detected with the MEM technique was 0.6 mm/s and with the angio-technique 0.4 mm/s. Two different scan-heads were used; a 5 MHz probe for the MEM procedures and a 10 MHz probe for ultrasound angiography. Using a 10 MHz scan-head will most likely lead to increased sensitivity of MEM. Blood flow representation with the MEM technique is real-time, while several seconds of acquisition time are required for the angio-technique, which is a disadvantage during clinical use. Conventional Doppler was merely able to represent 15 mm/s blood flow. The angio-technique was less dependent on the application angle than the MEM procedure., Conclusion: The definitely increased sensitivity of the two colour methods open up new areas in the diagnosis of organ and tumour perfusion.

Sonographic diagnostics has been essentially enlarged by blood flow measurements with the color Doppler technique. Nevertheless, the method has certain limitations, especially when employed to visualize extremely slow blood flow velocities, which occur e.g. in malignant tumors due to neovascularization. Recently, a new technique, the so-called angio-color procedure, has been presented, which is supposed to overcome the above-mentioned limitations. The way of coding the signal differs considerably from that in conventional color Doppler: The amplitude is color coded, not the frequency deviations. Results and first experiences with the new procedure in various domains were discussed at a 'Heidelberg Round Table' and are summarized in the present article. In the obstetrics/gynecology there are clear advantages in the representation of blood flow in the placenta. In eutrophic fetuses a blood flow over the entire placental breadth was detected, in dystrophic fetuses, however, only on the margins of the placenta. Conventional color Doppler could not visualize blood flow in the placenta at all. Blood flow assessment in the fetal organs also provided different results with the two procedures: the angio-color method showed color pixels even in the peripheral areas. In 8 malignant breast tumors the same observation was made, the angio-color procedure seemed to visualize more color pixels. In 4 of 6 benign lesions blood flow could be measured with the traditional color Doppler, but in 5 of 6 patients with the angio-color Doppler. This was also observed in ovarian tumors. Angio-color representation of renal transplants with normal function resulted in blood flow information reaching deep into the renal capsule, whereas in cases of renal transplants with impaired function a clearly reduced perfusion was visualized. These differences were not so evident with the conventional Doppler method. In lymph node diagnosis, too, marked differences between malignant and reactive lymph nodes could be visualized. In the diagnosis of arteriosclerotic plaques, blood flow in the residual lumen could also be detected more exactly with the angio-color technique. Color representation of liver tumors also showed a higher color signal density with ultrasound angiography. In summary, ultrasound angiography can provide additional information in the color representation of vessels with slow flow velocities. However, an obvious problem is the unability to quantify detected low blood flow. The new procedure will not replace but usefully complete the established sonographic techniques.

Objective: The aim of the present study was to find a possible relationship between biological behavior of breast cancer and sonographically detectable blood flow., Methods: 157 patients with ductal invasive breast cancer were examined by means of a new sonographic procedure, the MEM (Maximum Entropy Method), able to detect considerably lower blood flow velocities than doppler sonography. In the absence of objectifyable quantification methods, findings were allotted to 3 classes, depending on the visual color information obtained. Blood flow was correlated to tumor size, lymph node and receptor status, ploidy and S-phase-fraction., Results: Most patients with small tumors, no lymph node metatases, positive receptors, diploid genome and low S-phase found to have low blood flow., Conclusion: This close relation between established prognostic factors and results of sonographic blood flow examination with the MEM might indicate a new preoperative prognostic factor, which, however, will have to be proven by larger studies.

In the published study a totally new method for the diagnosis of malign and benign tumors of the female breast is presented. During sonographic examinations with the maximum entropy method we found out, purely by chance, that artifacts produced by vibrations of the thorax can be used to differentiate malignant from benign lesions. These artifacts in form of small colour pixels can be provoked by the vocal fremitus of the internal medicine or by humming a deep sound. In case of malignancy, the colour artifacts are visible inside of the tumor, in benign lesions, however, only in the surrounding tissue. On the preoperative day, 95 patients were examined with the method described above. In 91% of the cases the diagnosis of the dignity corresponded with the histological results. Seven times a benign lesion was classified as a malignant tumor and twice malignancy was not correctly diagnosed. Our explanation of these phenomena is the following: Benign lesions of the female breast show a displacing growth and a definable boundary. Thus the vibrations are damped and not conducted into the tumor. In contrast, malignant lesions grow invasively into the surrounding tissue. Consequently, the vibrations can be conducted into the center of the tumor by this dissemination.

Following the first three-dimensional (3D) ultrasound imaging of organs using 3D ringstructures, which we developed in 1986, the method had to be improved and refined because of sources of error resulting from the ring structure. The newly developed 3D method enades its application in clinical routine. By rotating the sectional plane around a horizontal axis using an especially constructed 3D probe, in 5 s 60 ultrasound sections of the examined tissue are registered. These 60 sections are calculated transparently and than compiled into a transparent spatially reconstructed image. In spite of the enormous information, the 3D picture is projected on the screen after a very short time. There is no loss in information. The spatially reconstructed body can be cut along any imaginable plane so that planes can be demonstrated which normally are inaccessible to conventional sonography. This new method presents possibilities which were formerly confined to CT and MRI. We tested this method mainly in prenatal and tumor diagnoses. It has been shown that the dignity of tumors can be differentiated in many cases already by means of the 3D-image. An indistinct image is characteristic for a malign tumor, and a sharply defined image represents a benign lesion. The diagnosis of fetal malformations can profit from this new method. Exact localizations and volume determinations are now sonographically possible. New diagnostic possibilities by means of ultrasound are forthcoming. But this needs further technical investigations. Both the presentation with a closed surface and the transparent presentation will have indications in the future. Our goal is a free applicability of the probe.

Three-dimensional sonography is a logical development of conventional sonography, and is a consequence of the fact that every examiner tries to imagine the three-dimensional appearance of an organ on the basis of the tomographic images of it. In a long development process starting with the first 3D images in 1986/87 and lasting until today, we succeeded in developing a 3D system suitable for routine examinations, which the manufacturer is now marketing commercially. With a 3D transducer a coordinated sequence of tomograms is obtained; all the images are then computed transparently and subsequently imaged three-dimensionally as a crystalline volume. Movement of the 3D image on the computer screen is important for three-dimensional identification. The volume thus computed can then be bisected in longitudinal, transverse and horizontal planes and thus examined in real-time without any interference such as fetal movements. The horizontal sections cannot be obtained by conventional sonography. On the basis of more than 600 patients examined by this new method it was established that 3D imaging of malformations improves diagnosis and assessment, that examinations deliver accurate and reproducible results, and that section analysis can substantially facilitate diagnosis of fetal malformations.

Previous experimental and clinical studies showed, that three-dimensional (3-D), i.e. dynamic spatial reconstruction of 2-D ultrasound images is feasible. A rotating acoustic plane helps to visualise the cross-sections serially and in a coordinated manner. When reconstructing the 2-D ultrasound images, surface reconstruction from planar contours of each cross-section is performed (so that the organ limits are defined) and the cross-sectional pictures are reconstructed into 3-D organ images. This approach had been described by the authors on a previous occasion. Another step forward is "transparent" spatial imaging eliminates the source of error involved in contouring. It is important to emphasise, that the 3-D system now described by the authors, yields a spatial image of the organ in question, and does not just present 3 planes of the organ (some authors have been claiming, that such an imaging method is also three-dimensional). Our system enables the imaging of several cross-sectional planes cut through the spatially calculated organ. This results in cross-sections, which cannot be performed by conventional sonography, thus disclosing new perspectives and possibilities. In the present article, the authors describe a clinical pilot study in the diagnosis of carcinomas of the breast. In our opinion, 3-D imaging of the tumours enables the physician to diagnose the status of the tumours; this is demonstrated by the examples of 35 malignant and benign tumours of the breast. In 95% of the cases of malignant tumours and 80% of the benign tumours, the diagnosis proved correct. This paper describes the fundamental possibilities and limitations of the new method.

Malignant tumours can be differentiated from the benign ones by their vascular blood supply. We employed three methods to analyse the blood flow: the angiodynograph, the duplex system, and the continuous-wave Doppler method, and measured the blood flow in 151 patients on the preoperative day. Histology revealed a malignoma in 92 cases and a benign tumour in 59 cases. It was found, that in more than 90% of the malignomas, a high blood flow was identified in or around the tumour by means of the colour method (angiodynography), which could be quantified by the pulsed-wave Doppler. A significantly lower blood flow was evident in the benign tumours with a markedly increased resistance index (by 80%) established by means of the pulsed-wave Doppler. Continuous-wave (CW) Doppler showed a significantly higher blood flow with almost all malignomas in the entire breast, than was the case with benign changes. Our studies showed, that the enhanced blood flow in and around malignant tumours can be visualised by means of update technology, angiodynography being particularly suitable for demonstrating the flow by the B-mode. Quantification, however, is, at present, only possible by employing the duplex method with pulsed-wave Doppler. CW Doppler is suitable only for blood flow diagnosis of the entire organ, but it requires great precision of working method and is time consuming; tumour blood flow cannot be visualised on-target by this method.

Three-dimensional reconstruction of an organ requires a coordinated sequence of cross-sectional scans. In ultrasound this can be solved by rotating the scan plane horizontally or vertically. Two scan heads have been built, one to rotate the plane around a horizontal, the second around a vertical axis. There are two ways to reconstruct the scans taken by ultrasound into a three-dimensional image: --to contour the surface of the organ in each scan and to reconstruct these contours to a ring-shaped structure. This procedure needs a lot of time; moreover, contouring must be performed via cursor on the screen. --calculate a transparent image of the organ. To obtain the best spatial image by the "transparent method", the image must be moved on the screen. Both methods of reconstruction are shown and compared in this paper. First experiences show that they can be applied in tumour diagnostic and in the diagnosis of malformations in early pregnancy. Further clinical studies will have to prove this.

The problem of gaining a coordinated sequence of section cuts by ultrasound is solved by rotating the plane section around a vertical and a horizontal axle. There are two possibilities in presenting the 3D image of the examined organ: ring structure image or transparent image. Precondition in the ring structure image is time consuming and defective since it has to be done by hand using a cursor. These pitfalls can be avoided when using a method without contouring by showing the 3D image transparently. Each section cut has to be calculated transparently. Both methods are presented in this paper and compared as to their applicability in medicine. The problem of the transparent 3D presentation lies in the fact that it cannot be well presented on printed paper, as is done her. The moving picture on the computer screen gives an optimal 3D image. This new method seem to be useful in tumor diagnostic and diagnostic of malformations in early pregnancies.

This is the first report in literature on three-dimensional imaging of organs via sonography. The relevant experiments were performed in vitro. Whereas MRI and CT can produce three-dimensional images of bodies by means of appropriate computer programmes, we had to search for special techniques in sonography that would communicate to the computer the exact positioning and arrangement of the individual segments to be reconstructed to supply a three-dimensional image. In MRI and CT the individual segments are arranged parallel to one another; the distance between the individual segments is known; all the computer has to do is to add up these segments to produce a spatial image. In contrast to this, conventional sonography cannot supply parallel segments or sections due to the unevenness of the human skin. Hence, it was not possible to use the computer programmes compiled for the three-dimensional reconstruction of MRI and CT images, in sonography; special transducer guides had to be constructed before this could be realised. One of our special constructions enabled parallel shifting of the transducer to obtain parallel segments or sections, and another construction enabled rotation of the transducer to obtain segments or sections differing from one another by a known angle of displacement. In this manner the computer was able to reconstruct an organ to supply a three-dimensional image--a first-time achievement. Using these devices, we examined a kidney in a water-bath. By means of outlining of the individual sonographic segments, only their surfaces are depicted, and these are reconstructed to produce a three-dimensional body by means of newly developed computer programmes.(ABSTRACT TRUNCATED AT 250 WORDS)

Clinical and experimental studies proved the usefulness of three-dimensional sonographic diagnosis by means of a newly constructed transducer providing for a coordinated sequence of cuts, and hence initial trials investigated its application in sonographic diagnosis of the infant hip. Differences were found in the three-dimensional anatomic image between normal and dysplastic hips. It is no longer necessary to draw any lines or to perform any calculations of angles as required in "conventional" hip sonography. In contrast to conventional hip sonography, the use of a sector transducer is meaningful with this method, since the drawbacks of the sector transducer stated by Graf, do not apply. Clinical studies will have to confirm the possible use of this new method in diagnosis disturbances in infant hip maturation.

After satisfactory experimental results with three-dimensional data acquisition and presentation of the kidney in vitro, we wanted to show how our technique could be used for medical application in vivo, such as ultrasonic 3D reconstruction of organs. The sector scans were taken by rotating the scan head 10 degrees in any direction around the length of the axis. The organ was scanned with 18 images and reconstructed. The digital images, using the organ contours, allowed 3D reconstruction of the original organ, relying on the computer memory. The first results in reconstructing uteri show how this scan head can be used in combination with the computer programs for medical application. For the first time, it has become possible to present computer-generated views of an organ cross-section that has been impossible to obtain by means of traditional ultrasound techniques. Recent experiments using this method show new ways of diagnosing tumors.

This is a report on the clinical use of sonographic imaging. Experimental studies have shown that parallel sonographic sections are available only in vitro, which means no in-vivo three-dimensional reconstruction is possible as it is with magnetic resonance and computed tomography. The idea that a coordinated sequence of sections can be obtained by spatial rotation of the plane of sound so that the individual sections differ from one another by defined angular distances, enables three-dimensional reconstruction of sonographic in-vivo sections. A transducer was constructed that enables the production of sonographic sections rotating around a fixed center of rotation. Its clinical usefulness was tried and confirmed in the imaging of early pregnancies, benign and malignant carcinomas of the breast, and in imaging a gallbladder with a solitary gallstone. Three-dimensional imaging can be achieved either as a ring-shaped structure or with a continuous surface. This points to the possibility of diagnosing malformations early in pregnancy if the number of sections is sufficiently high. In tumour imaging the malignant tumour seems to be clearly distinguishable from the benign one; three-dimensional diagnosis is likely to furnish important additional criteria in diagnosis. Further clinical studies will have to verify this.

In the study reported here three-dimensional sonographic imaging of organs was achieved for the first time. To make this possible it was first necessary to ensure, by appropriate guidance of the probe, that the sequence of sonographic sections was coordinated in their spatial arrangement. This was accomplished by constructing a probe guide with which parallel sonographic sections could be demonstrated. The distance between these sections was known, so that with the aid of suitable computer programs three-dimensional reconstruction of a kidney examined in a water bath was possible. Since, however, due to the uneven surface of the body, it will hardly be possible to obtain parallel sonographic sections of an organ, a new solution had to be found to ensure the necessary coordinated sequence of sections. The solution lay in rotating the probe. A further device was constructed in which the probe could be rotated farther, by known angles, from section to section. The pivotal point was at the center of the probe tip. The computer knew the angular distance between these sections and reconstruction to a three-dimensional image was therefore possible. Prior the three-dimensional reconstruction the ultrasonographic sections had to be contoured, since only the surface of the organ was available for three-dimensional image construction. Three-dimensional imaging of an organ can be achieved on the one hand by binary representation and on the other with a continuous organ surface. The advantage of binary representation is that the original sonographic data are incorporated in the image-producing process without any computer manipulation; with a continuous surface the distance between the individual sections has to be interpolated.(ABSTRACT TRUNCATED AT 250 WORDS)

This is the first report on the 3-dimensional representation of organs with US. The prerequisite for this was a coordinated transducer movement, in such a manner that the organ under examination was represented by US sections differing in only one of the space coordinates. Such transducer movement was made possible by 2 devices. In the first instance, longitudinal movement of the transducer resulted in the production of parallel sections of the organ while, in the other instance, rotation of the transducer head permitted sections arranged around a fixed center-point. Using a special computer program, the sections were contoured in such a manner that only the surface of the organ was represented. These sections were then arranged in space. The 3-dimensional representation can be effected both by binary image representation and by representation with closed (intact) body surface. The advantage of the binary image representation is the fact that the organ surfaces "extracted" from the original US sections are directly incorporated within the 3-dimensional image build-up, with no further computer manipulations. It can be seen that the rotation of the transducer head represents the practicable possibility for the use in the clinical setting.

A new method of three-dimensional (3-D) reconstruction of 2-D ultrasound images of the kidney is described. It is based on a coordinated spatial reconstruction of sequential cross-sectional images. The ultrasound head is moved longitudinally between two rails (parallel sections) and rotated. With a suitable computer program and contouring of each cross-section (so that the organ limits are defined for the computer) these cross-sectional pictures can be reconstructed into 3-D organ images. The kidney can then be presented spatially either as a binary picture or with closed surface. Ultrasound investigators are still unaccustomed to colour reproduction of 3-D reconstructed organs. It remains to be seen whether the method is valuable in routine clinical use.