WU Wenjing, 坪田 悠史, 鈴木 敦郎, 山中 一宏, 川畑 健一, 寺田 崇秀, 山下 啓子, 加藤 扶美, 西田 睦, 佐藤 恵美

Wenjing WU, Yushi TSUBOTA, Asturo SUZUKI, Kazuhiro YAMANAKA, Kenichi KAWABATA, Takahide TERADA, Hiroko YAMASHITA, Fumi KATOU, Mutsumi NISHIDA, Satoh MEGUMI

¹日立製作所研究開発グループ, ²北海道大学病院乳腺外科, ³北海道大学病院放射線診断科, ⁴北海道大学病院超音波センター

¹Research&Development Group, Hitachi,Ltd., ²Breast surgery, Hokkaido University Hospital, ³Diagnostic and Interventional Radiology, Hokkaido University Hospital, ⁴Diagnostic Center for Sonography, Hokkaido University Hospital

キーワード :

［Target］
Ultrasound computed tomography（USCT）is very promising to be applied to early screening for breast cancer, but the incident intensity of transmission signal is constrained by transducer features. The signal emitted with conventional spherical wave could be attenuated seriously especially for high attenuation breasts, which would cause many artifacts. The proposed virtual-source incident pattern using synthetic wave could improve the signal noise ratio and reduce the artifacts effectively.
［Method］
The proposed method could enhance the signal by emitting from multiple piezoelectric elements simultaneously. Each element is synchronized to generate spherical ultrasound wave virtually focused at a predetermined point source. Experiments were conducted using both conventional incident pattern and proposed method with our USCT prototype（200-mm-diameter ring array, 2048-elements transducer）. The images were taken with an elliptical gel phantom with a major and minor diameter of 140 and 120 mm, and surgically removed human breast tissues including tumor regions. In both cases, specimen possessed high attenuation（more than 1 dB/MHz/cm）
［Results］
Comparing the sectional images of elliptical gel phantom, the proposed method showed much less artifacts and the inclusions could be well discerned, while the image obtained by conventional spherical wave pattern had poor discrimination due to serious artifacts. Same effect is validated in images of human breast. The conventional spherical wave pattern failed to show the tumor, while the proposed method detected the tumor correctly.
［Conclusion］
The merits of proposed method are validated by experiments with phantom and ex vivo human breast tissues. It is very effective in artifacts reduction and could expand the applicable ranges of USCT for high attenuation objects.