While three-dimensional ultrasound is not yet as fast as two-dimensionalimaging, reconstructed 3-D images may provide diagnostic informationthat standard scanning techniques cannot provide, according toRalph Ranalli, product manager for Philips
While three-dimensional ultrasound is not yet as fast as two-dimensionalimaging, reconstructed 3-D images may provide diagnostic informationthat standard scanning techniques cannot provide, according toRalph Ranalli, product manager for Philips Ultrasound.
Philips' strategy in its 3-D development program is to applythe technology to problems posed by 2-D techniques, Ranalli said.
"Because it (3-D ultrasound) is still a research project,we are looking at everything ultrasound does at this point,"he said.
Initial work by the vendor showed how 3-D volumetric reconstructioncan create image planes not possible in standard scanning dueto transducer access problems (SCAN 4/10/91).
Philips has also developed an image projection technique using3-D volumetric reconstruction that produces high-resolution 2-Dimages of anatomical structures not easily scanned with standardtransducers. Three-D image projection can portray objects, suchas a curved fetal hand, with a clarity not seen before in 2-Dimages, he said.
The technique involves searching through a 3-D volume and isolatingthe bright spots that represent bone or other objects. This datais then processed and presented in an image showing the full bonecurvature, Ranalli said.
Three-D image projection allows fetal sonographers to betterfocus on specific areas, such as the face and hands. Reconstructedimages may aid sonographers in obtaining fetal bone measurements,which are critical to high-risk ob/gyn assessment.
Philips showed projection images at its booth at the AmericanInstitute of Ultrasound in Medicine meeting in San Diego lastmonth. The technique was also discussed by Dr. Dolores H. Pretoriusduring an AIUM categorical course on ob/gyn ultrasound. Pretoriusis chief of ultrasound at the University of California, San Diego,and a leading researcher in fetal assessment.
What New Research Reveals About ChatGPT and Ultrasound Detection of Thyroid Nodules
March 13th 2024In a comparison of image-to-text large language models (LLMs), ChatGPT 4.0 offered a 95 percent sensitivity rate and an 83 percent AUC that were comparable to that of two senior radiologists and one junior radiologist interacting with LLM to differentiate between malignant and benign thyroid nodules on ultrasound.
ECR Study Finds Mixed Results with AI on Breast Ultrasound
March 6th 2024While adjunctive use of AI led to significantly higher specificity and accuracy rates in detecting cancer on breast ultrasound exams in comparison to unassisted reading by breast radiologists, researchers noted that 12 of 13 BI-RADS 3 lesions upgraded by AI were ultimately benign, according to research presented at the European Congress of Radiology.