Fetal Ultrasound Videos Research Articles

P22.03: Searching for structures of interest in an ultrasound video sequence with an application for detection of breech

The shortage of highly trained sonographers worldwide limits the wider adoption of ultrasound (US) as a diagnostic tool in obstetrics. The aim of this study was to develop a learning software solution to analyse US video-clips acquired along the major uterine longitudinal axis. In particular, the automatic computer analysis was tested to diagnose non-cephalic presentation. Transabdominal 2D fetal US video clips were acquired from 30 subjects participating in the INTERGROWTH-21st project (www.intergrowth21st.org) at the University of Oxford, following a simple and repeatable protocol: a single sweep starting at the symphysis pubis of the mother moving toward the fundus of the uterus over 6-8 seconds was acquired. Data acquisition was carried out using a mid-range US machine (Philips HD9 with a V7-3 transducer). Each video sequence was broken down into smaller sequences, and machine learning techniques were used to look for the fetal skull. When the fetal skull was detected in the last third of the video sequence it was assumed that the fetus is in a non-cephalic presentation. Video clips were acquired and analysed in 100% of the cases. Automatic head detection and thus detection of non-cephalic presentations was possible with an average accuracy of 90% in all the videos obtained through our proposed protocol. We have proposed a general framework for US video analysis. Cost effective tools for automatic US video clips analysis can assist in the diagnosis of fetal presentation; in this case only a longitudinal uterine axis sweep is required by the sonographer. The described tool should be developed to assess other structures of interest and be tested for use in areas where basic US training is lacking.

Hypertext atlas of fetal and neonatal pathology

Hypertext atlas of fetal and neonatal pathology is a free resource for pregraduate students of medicine, pathologists and other health professionals dealing with prenatal medicine. The atlas can be found at . The access is restricted to registered users. Concise texts summarize the gross and microscopic pathology, etiology, and clinical signs of both common and rare fetal and neonatal conditions. The texts are illustrated with over 300 images that are accompanied by short comments. The atlas offers histological pictures of high quality. Virtual microscope interface is used to access the high-resolution histological images. Fetal ultrasound video clips are included. Case studies integrate clinical history, prenatal ultrasonographic examination, gross pathology and histological features. The atlas is available in English (and Czech) and equipped with an active index. The atlas is suitable both for medical students and pathologists as a teaching and reference tool. The atlas is going to be further expanded while keeping the high quality of the images.

Accuracy of assessment of fetal anomalies using compressed fetal ultrasound video clips

Background: Ultrasound is an accepted method of diagnosis and assessment of fetal anomalies, but the performance of ultrasound varies widely between tertiary and general units. Telemedicine involves the delivery of health care across a distance by combining telecommunications and medical expertise, and bridges the health care gap between the city and the regional centers, where access to subspecialist care is usually limited. Our group has shown that real‐time teleultrasound consultation is not only technically feasible, but also offers diagnostic and management modifications.Current real‐time telemedicine relies on digital (ISDN) lines and hence is costly and still limited to larger centres. If the Internet could be used as a method of downloading ultrasound images taken in the field quickly and effectively, this would bring tertiary consultation to even extremely remote centers. Before ultrasound video clips could be transferred across the Internet efficiently, they need to be compressed into smaller files. Our group has previously determined that the maximum compression ratio that was considered clinically acceptable (without compromising image quality) was 1:50 to 1:100.Aim: To assess the ability of experienced ultrasonologists to make accurate diagnoses based on compressed ultrasound clips of fetal anomalies, as well as their confidence in making such diagnoses.Methods: Nine different clips of fetal anomalies and two different normal controls were compressed using commercially available software. The original uncompressed file sizes ranged from 55.6 to 622.7 MB. Final file sizes ranged from 1.6 to 10 MB (i.e. compression ratios of 1:10 to 1:130). Four experienced tertiary ultrasound specialists blinded to the compression factor evaluated the images. The examiners were asked to make a diagnosis from a selection of 48 possible choices. They were also asked to rate their confidence in making the diagnosis (based on the quality of transmission) and image clarity on a scale of 1–7. Data was analyzed using Duncan's multiple range test for variables, with alpha = 0.05.Results: A total of 71 versions of the 11 video clips were played at random to the observers. Intra‐ and interobserver variations were not significant. All of the observers were certain of their diagnoses (mean score above 4), although the certainty scores for cardiac anomalies and for normal fetuses were lower than the scores for other anomalies Of the 71 video clips, 66 were correctly diagnosed (93%). Two were partially correct (2.8%), and three were incorrect (4.2%). All the incorrect diagnoses were cardiac anomalies.Conclusion: Video clips of fetal ultrasound can be digitized, compressed and displayed on the computer without clinically or statistically significant loss of diagnostic certainty or image clarity. For compressions up to 1:100, the diagnostic accuracy was very high, although cardiac anomalies seem to pose the most difficulties. While the potential for consultation via the Internet is enormous, larger clinical trials need to be performed to confirm these results.