Could volume sweep imaging assist in breast cancer detection?

Thursday, December 1 | 9:30 a.m.-10:30 a.m. | R3-SSIN07-3 | S404
A new system that automatically detects palpable breast lumps without the need of a sonographer or radiologist will be explained in this RSNA session.

Dr. Thomas Marini from the University of Rochester Medical Center in New York will talk about the system, which uses ultrasonic volume sweep imaging and S-Detect software (Samsung). Images from volume sweep imaging are fed directly into S-Detect for interpretation. Images of palpable breast lumps in this study were acquired with a Butterfly iQ probe operated by a medical student without prior ultrasound training.

S-Detect analyzed 115 palpable masses. Researchers found "substantial" agreement between the system and standard-of-care interpretations (Cohen's k = 0.79), expert volume sweep imaging ultrasound reports (k = 0.73), expert standard-of-care ultrasound reports (k = 0.73), and pathological diagnosis (k = 0.8). All comparable data were found to be statistically significant.

The researchers also noted greater than 80% agreement between S-Detect and the other standard references when it came to interpretations of volume sweep imaging for mass features, such as margins and shape. Also, S-Detect showed a sensitivity of 100% and specificity of 86% for all pathologically proven cancers designated as "possibly malignant."

Find out more about the system and how it may be able to help in areas where supplemental breast ultrasound isn't widely available in this Thursday session.

Page 1 of 507
Next Page