Real-Time Palpation Imaging for Improved Detection and Discrimination of Breast Abnormalities

摘要

The purpose of this work is to develop and test a new kind of imaging system that the authors call 'palpation imaging.' They have demonstrated that palpation imaging can be a useful tool for improving the discrimination between benign and malignant breast tumors. They implemented their algorithm for imaging tissue elasticity on a commercial ultrasound imaging system and tested the new imaging system on phantoms and in vivo breasts. Palpation images are produced at substantially real-time frame rates with normal ultrasound B-mode and strain images displayed side-by-side. The algorithms are fully integrated into the commercial system and require no system modifications. Breast exams performed on volunteers have shown that palpation imaging techniques are almost identical to the standard clinical breast ultrasound exam. ROC analysis of elasticity image information demonstrates that the diagnostic confidence of breast radiologists is improved when elasticity information is added to standard ultrasound images. Further increases in observer performance will come through improved training, observer tools (automated segmentation), and image quality improvements.