TY - GEN
T1 - Real-time coherence imaging of suspicious breast masses recommended for aspiration or biopsy
AU - Sharma, Arunima
AU - Gonzalez, Eduardo A.
AU - Ambinder, Emily
AU - Myers, Kelly
AU - Oluyemi, Eniola
AU - Bell, Muyinatu A.Lediju
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - Our group previously demonstrated that coherence-based imaging improves the certainty of fluid mass content in breast ultrasound images, and therefore has the potential to decrease unnecessary biopsies. However, this potential was demonstrated with post-processed raw channel data after patient examination. This work is the first to implement real-time short-lag spatial coherence (SLSC) imaging on an FDA-approved clinical ultrasound scanner to investigate a combined B-mode and coherence-based imaging approach to improve the diagnostic capabilities of breast ultrasound imaging. Real-time SLSC imaging was provided through GPU-based SLSC scripts that we developed and installed on our FDA-approved Alpinion ECUBE12R ultrasound scanner. A total of 47 hypoechoic breast masses from 32 patients were examined. First, the board-certified breast radiologists performing the patient exam predicted the mass contents and the BI-RADS category of the mass using real-time clinical B-mode images provided by the scanner. Then, the scanner was switched to present real-time SLSC images of the masses alongside real-time B-mode images, and the radiologist performing the scan repeated the predictions. The inclusion of SLSC images resulted in correct predictions of 100% of complicated cysts as BI-RADS 2 lesions (compared to 50% of complicated cysts being predicted as BI-RADS 2 lesions with only B-mode images). In addition, SLSC correctly identified presence of non-solid content in clusters of cysts, and no change in the BI-RADS category of malignant masses was observed after the inclusion of SLSC imaging. These results demonstrate the real-time potential of SLSC to decrease unnecessary biopsies by correctly identifying complicated cysts as benign fluid lesions.
AB - Our group previously demonstrated that coherence-based imaging improves the certainty of fluid mass content in breast ultrasound images, and therefore has the potential to decrease unnecessary biopsies. However, this potential was demonstrated with post-processed raw channel data after patient examination. This work is the first to implement real-time short-lag spatial coherence (SLSC) imaging on an FDA-approved clinical ultrasound scanner to investigate a combined B-mode and coherence-based imaging approach to improve the diagnostic capabilities of breast ultrasound imaging. Real-time SLSC imaging was provided through GPU-based SLSC scripts that we developed and installed on our FDA-approved Alpinion ECUBE12R ultrasound scanner. A total of 47 hypoechoic breast masses from 32 patients were examined. First, the board-certified breast radiologists performing the patient exam predicted the mass contents and the BI-RADS category of the mass using real-time clinical B-mode images provided by the scanner. Then, the scanner was switched to present real-time SLSC images of the masses alongside real-time B-mode images, and the radiologist performing the scan repeated the predictions. The inclusion of SLSC images resulted in correct predictions of 100% of complicated cysts as BI-RADS 2 lesions (compared to 50% of complicated cysts being predicted as BI-RADS 2 lesions with only B-mode images). In addition, SLSC correctly identified presence of non-solid content in clusters of cysts, and no change in the BI-RADS category of malignant masses was observed after the inclusion of SLSC imaging. These results demonstrate the real-time potential of SLSC to decrease unnecessary biopsies by correctly identifying complicated cysts as benign fluid lesions.
KW - breast imaging
KW - coherence beamforming
KW - real-time
KW - ultrasound
UR - http://www.scopus.com/inward/record.url?scp=85178629936&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85178629936&partnerID=8YFLogxK
U2 - 10.1109/IUS51837.2023.10307040
DO - 10.1109/IUS51837.2023.10307040
M3 - Conference contribution
AN - SCOPUS:85178629936
T3 - IEEE International Ultrasonics Symposium, IUS
BT - IUS 2023 - IEEE International Ultrasonics Symposium, Proceedings
PB - IEEE Computer Society
T2 - 2023 IEEE International Ultrasonics Symposium, IUS 2023
Y2 - 3 September 2023 through 8 September 2023
ER -