Development of a hybrid tomosynthesis and acoustic imaging modality for early detection of breast cancer

2017 Fiscal Year Annual Research Report

• Project/Area Number: 17H05075
• Research Institution: Hokkaido University
• Principal Investigator: 彭 浩 北海道大学, 国際連携研究教育局, 招へい教員 (70790195)
• Project Period (FY): 2017-04-01 – 2020-03-31
• Keywords: Breast cancer / Tomosynthesis / Acoustic imaging / X-ray

Outline of Annual Research Achievements

System design from signal induction to readout electronics was completed and appropriate hardware and software was procured, including a 5Mhz ultrasound transducer with 80% bandwidth and multiplexing signal acquisition and processing system. Development of pathway to simulate X-ray- induced acoustic (XA) imaging is complete. Monte Carlo simulation of X-ray transport and interaction, as well as that for acoustic wave propagation are being performed to model sensitivity and spatial resolution needs on both the X-ray and acoustic sides for breast imaging purposes using GATE, k-Wave, MATLAB, and virtual breast phantoms. A noise-equivalent pressure (NEP) / signal-to-noise ratio (SNR) framework has been developed, showing potential for NEP to be <1 Pa dependent on area of transducer surface. The design of the transducer is proprietary however, so this has yet to be tested under realistic conditions. The optimum configurations of the system, in terms of image SNR and contrast- to-noise ratio (CNR) are being investigated, with the goal being to differentiate between breast tissue subgroups of different molecular densities, with various micro-calcification sites having or not having fiducial markers. Radiosensitizers are being investigated quantitatively for their dose/contrast enhancement for this imaging modality being developed. Minimum required dose under different conditions are being investigated.

Current Status of Research Progress

2: Research has progressed on the whole more than it was originally planned.

Reason

System design and equipment purchases were completed without issue and we conduct experiments at scheduled pace.

Strategy for Future Research Activity

Continue investigating optimum configurations of the system for breast imaging, in terms of image SNR and CNR. Confirm theoretical estimations, including NEP, in realistic conditions, especially given additional sources of noise in signal processing chain and imaging environment. Further investigate potential of radiosensitizers in XA mammographic imaging.
Move on to further development of XA image reconstruction. Use procured hardware to develop an accurate model that accounts for non-uniform X-ray attenuation. The signal-limited nature of the measurements will be modeled using a statistical reconstruction approach. One possibility to be evaluated in our study is whether the angular information from multiple projections in the tomosynthesis can help us improve the reconstruction/location of targets, when compared with a standard mammography. In other words, among multiple x-ray tube positions, a strategy for their use in image reconstruction might be able to provide superior image quality than others due to several physical factors, such as tissue overlapping or scattering/attenuation along a projection line.
Investigate optimal geometric efficiency of transducer elements in terms of image quality. Investigate alternative reconstruction methods/models, effects of X-ray attenuation, and further potential applications. Present findings as milestones are met in conferences and journal papers.