Why Does a 3D Ultrasound Sometimes Look Strange?

3D ultrasound can look distorted because it is a surface rendering built from sound-wave data. Learn about shadows, missing surfaces, placenta, movement and other artifacts.

A strange-looking 3D ultrasound usually reflects the way the image was acquired or rendered. The familiar surface image is generated from ultrasound echoes. If the face is partly hidden, moving, touching another structure or poorly separated from surrounding tissue, the rendering can create holes, flattened areas, extra-looking tissue or asymmetric contours. [8, 11, 20]

Educational visual for why a 3D ultrasound can look strange. Educational illustration; not a diagnostic ultrasound image.
Visual overview of why a 3D ultrasound can look strange. Educational illustration; not a diagnostic ultrasound image.

3D ultrasound can look distorted because it is a surface rendering built from sound-wave data. Learn about shadows, missing surfaces, placenta, movement and other artifacts.

Educational schematic of factors that make 3D facial images look unusual. Educational illustration; not a diagnostic ultrasound image.
Strange-looking 3D faces usually reflect acquisition and rendering, not a diagnosis from one still. Educational illustration; not a diagnostic ultrasound image.

Why the eyes can look hollow

Surface-rendered 3D ultrasound primarily depicts external surfaces. The eyelids and surrounding tissues may be visible, but the rendering does not behave like a photograph of open eyes. Shadowing and the chosen rendering threshold can make the orbital region appear deep or hollow.

Educational pipeline diagram titled why a rendered 3D ultrasound can look different from a photograph, from ultrasound echoes through 2D frames, volume acquisition and surface rendering to a displayed 3D face. Educational illustration; not a diagnostic ultrasound image.
Why a rendered 3D ultrasound can look different from a photograph: echoes become 2D frames, then a volume, then a displayed surface. Educational illustration; not a diagnostic ultrasound image.

Why the nose can look flat or wide

The nose projects from the face and is sensitive to viewing angle. If the facial surface is oblique, partially shadowed or touching nearby tissue, the rendered nose may look flattened, broadened or incomplete. A single rendered image is therefore unsuitable for judging subtle facial morphology.

Why part of the face can disappear

Surface rendering requires usable data at the boundary between the fetal face and its surroundings. Where the boundary is not captured clearly, the algorithm may fail to display a surface. This can create a missing cheek, lip edge or forehead region.

Why extra 'tissue' can appear

Placenta, uterine wall, umbilical cord or a limb can lie close to the face. Rendering may display these structures together, creating apparent lumps or bridges that are not part of the facial surface.

Motion artifacts

During volume acquisition, fetal or maternal motion can cause spatial inconsistency. The result may be a stepped, duplicated or warped surface. This is one reason the best still image is selected from multiple attempts rather than treating every acquisition as equally representative. [8, 15]

Educational diagram of motion artifacts on a 3D rendering. Educational illustration; not a diagnostic ultrasound image.
Motion during volume capture can warp, step, or duplicate the rendered surface. Educational illustration; not a diagnostic ultrasound image.

Do strange images mean something is wrong?

Not by themselves. Diagnostic conclusions should come from the complete ultrasound examination interpreted by qualified professionals, not from the cosmetic appearance of one 3D rendering. Standard 2D planes remain central to prenatal diagnosis, with 3D used as an adjunct in selected circumstances. [1, 2, 11]

Implication for 3D modeling

Artifact recognition is essential when converting ultrasound data into a digital or physical surface. A responsible workflow should separate observed anatomy from noise and should avoid confidently inventing structures that the acquisition did not show.

Educational infographic comparing 2D, 3D and 4D ultrasound side by side. Educational illustration; not a diagnostic ultrasound image.
2D, 3D and 4D ultrasound: a side-by-side comparison. Educational illustration; not a diagnostic ultrasound image.

Medical disclaimer

This article is intended for educational purposes only and does not provide medical advice, diagnosis, or treatment. Always consult a qualified healthcare professional regarding your individual medical care. This content is based on peer-reviewed scientific literature and current clinical guidelines available at the time of publication. Every article requires physician review before publication.

Key Takeaways

  • Surface-rendered 3D ultrasound primarily depicts external surfaces.
  • The nose projects from the face and is sensitive to viewing angle.
  • Surface rendering requires usable data at the boundary between the fetal face and its surroundings.
  • Placenta, uterine wall, umbilical cord or a limb can lie close to the face.
  • During volume acquisition, fetal or maternal motion can cause spatial inconsistency.

Frequently Asked Questions

Why are the eyes dark or hollow?

Surface rendering and shadowing can make the orbital region look deep. It is not a photograph of the eyes.

Why is the nose flattened?

Viewing angle, contact with nearby tissue and incomplete surface data can alter the rendered contour.

Should I diagnose a problem from a strange 3D image?

No. Diagnostic interpretation requires the complete medical examination.

References

  1. ACOG. Ultrasound Exams. American College of Obstetricians and Gynecologists. Standard obstetric ultrasound is commonly performed at 18-22 weeks; specialized techniques may include 3D ultrasound. https://www.acog.org/womens-health/faqs/ultrasound-exams
  2. ISUOG Practice Guidelines (updated): performance of the routine mid-trimester fetal ultrasound scan. Ultrasound Obstet Gynecol. 2022;59:840-856. https://www.isuog.org/resource/isuog-practice-guidelines-updated-performance-of-the-routine-mid-trimester-fetal-ultrasound-scan.html
  3. FDA. Ultrasound Imaging. U.S. Food and Drug Administration. Ultrasound does not use ionizing radiation; keepsake images are reasonable when produced during a medically indicated examination without additional exposure. https://www.fda.gov/radiation-emitting-products/medical-imaging/ultrasound-imaging
  4. AIUM. Prudent Clinical Use and Safety of Diagnostic Ultrasound. 2019. Supports prudent, medically appropriate use and attention to acoustic output and exposure time. https://www.aium.org/resources/official-statements/view/prudent-clinical-use-and-safety-of-diagnostic-ultrasound
  5. AIUM. Prudent Use and Safety of Diagnostic Ultrasound in Pregnancy. Official statement. https://www.aium.org/resources/official-statements/obstetric-ultrasound
  6. NHS. 20-week screening scan. The anomaly scan is usually performed between 18 and 21 weeks and assesses fetal anatomy and growth. https://www.nhs.uk/pregnancy/your-pregnancy-care/20-week-scan/
  7. MedlinePlus. Fetal development. U.S. National Library of Medicine. https://medlineplus.gov/ency/article/002398.htm
  8. Connor A, Matthew J, Chung E, Wright R, Rutherford MA. Factors influencing the successful acquisition and quality of three-dimensional souvenir obstetric ultrasound facial images: a narrative review and quantitative evaluation. Ultrasound. 2026. PMID: 42294041. DOI: 10.1177/1742271X261436619. DOI: 10.1177/1742271X261436619 PMID: 42294041 Source
  9. Accuracy and repeatability of fetal facial measurements in 3D ultrasound: A longitudinal study. Early Hum Dev. 2024. PMID: 38701668. DOI: 10.1016/j.earlhumdev.2024.106021. DOI: 10.1016/j.earlhumdev.2024.106021 PMID: 38701668 Source
  10. Three-dimensional sonographic visualization of the fetal face. AJR Am J Roentgenol. PMID: 9456969. DOI: 10.2214/ajr.170.2.9456969. DOI: 10.2214/ajr.170.2.9456969 PMID: 9456969 Source
  11. Two- and three-dimensional sonographic assessment of the fetal face. 1. A systematic analysis of the normal face. Ultrasound Obstet Gynecol. PMID: 15027008. DOI: 10.1002/uog.984. DOI: 10.1002/uog.984 PMID: 15027008 Source
  12. Diagnostic evaluation of the fetal face using 3-dimensional ultrasound. Ultrasound Q. 2008;24:215-223. PMID: 19060688. DOI: 10.1097/RUQ.0b013e31819073c2. DOI: 10.1097/RUQ.0b013e31819073c2 PMID: 19060688 Source
  13. Three-dimensional and four-dimensional ultrasound applications in fetal medicine. Curr Opin Obstet Gynecol. 2009;21:167-174. PMID: 19996869. DOI: 10.1097/GCO.0b013e328329243c. DOI: 10.1097/GCO.0b013e328329243c PMID: 19996869 Source
  14. Role of 3-D ultrasound in clinical obstetric practice: evolution over 20 years. Ultrasound Med Biol. 2015. PMID: 25748522. DOI: 10.1016/j.ultrasmedbio.2014.12.009. DOI: 10.1016/j.ultrasmedbio.2014.12.009 PMID: 25748522 Source
  15. 3D/4D sonography - any safety problem. J Perinat Med. 2016;44:125-129. PMID: 26376219. DOI: 10.1515/jpm-2015-0225. DOI: 10.1515/jpm-2015-0225 PMID: 26376219 Source
  16. 4D ultrasound study of fetal facial expressions at 20-24 weeks of gestation. Int J Gynaecol Obstet. 2014;126:275-279. PMID: 24996686. DOI: 10.1016/j.ijgo.2014.03.036. DOI: 10.1016/j.ijgo.2014.03.036 PMID: 24996686 Source
  17. Fetal face shape analysis from prenatal 3D ultrasound images. Sci Rep. 2024. PMID: 38388522. PMCID: PMC10884000. DOI: 10.1038/s41598-023-50386-9. DOI: 10.1038/s41598-023-50386-9 PMID: 38388522 Source
  18. Three-dimensional ultrasonic images of normal fetus. Study of pregnancies 20-36 weeks; satisfactory recognizable images were reported particularly from 24-34 weeks. PMID page. PMID: 11601189 Source
  19. Three-dimensional ultrasonographic assessments of fetal development. PMID: 9469279. PMID: 9469279 Source
  20. Three-dimensional reconstruction of fetal abnormalities using ultrasonography and magnetic resonance imaging. PMID: 29656679. PMID: 29656679 Source

Glossary · Frequently asked questions · All Learning Center articles